UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Sustainable vegetable greenhouse production through bio-conversion of greenhouse solid wastes and re-utilization Cheuk, William Wai Lun 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2004-901610.pdf [ 41.24MB ]
Metadata
JSON: 831-1.0058673.json
JSON-LD: 831-1.0058673-ld.json
RDF/XML (Pretty): 831-1.0058673-rdf.xml
RDF/JSON: 831-1.0058673-rdf.json
Turtle: 831-1.0058673-turtle.txt
N-Triples: 831-1.0058673-rdf-ntriples.txt
Original Record: 831-1.0058673-source.json
Full Text
831-1.0058673-fulltext.txt
Citation
831-1.0058673.ris

Full Text

SUSTAINABLE VEGETABLE GREENHOUSE PRODUCTION THROUGH BIO-CONVERSION OF GREENHOUSE SOLID WASTES ANDRE-UTILIZATION by WILLIAM WAI LUN CHE UK B.A.Sc, The University of British Columbia, 1996 B.B.A., Simon Fraser University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Chemical and Biological Engineering Faculty of Applied Science We accept this thesis as conforming to the reauired standQd THE UNIVERSITY OF BRITISH COLUMBIA November, 2003 © William Wai Lun Cheuk, 2003 11 ABSTRACT Current practices of handling greenhouse wastes are not the sustainable ways to conserve agricultural lands and ground waters. This study developed a sustainable growing practice in the vegetable greenhouse industry. Waste handling (shredding) and the biodegradable plastics were investigated first. Then, different composting control algorithms and substrate recipes were tested in both lab scale and pilot scale composting. With a good control algorithm, composting of greenhouse wastes could reach the requirement for Process to Reduce Further Pathogens, PFRP (55 OC for 3 days). Ammonia emission might be a problem but it could be reduced by using air-recirculation or removed by a biofilter with compost as medium. Recirculation cooling control was found to be a more effective method, to maintain the process temperature below the set point, than any kind of temperature feedback control. Less leachate and condensate were found from the reactors with air recirculation control. Systems with air recirculation for cooling and aeration showed higher degradation rates, and also more consistent moisture content of the final compost. Alder bark was found out to be a better choice of bulking agent than hemlock bark in terms of better substrate structure, more carbon loss, less nitrogen loss, and higher process temperature. Shredding was proven to be not necessary before composting of prunings and it also helped minimizing the amount of leachate. Bulking agents (alder bark) of about 20-30% (in weight) were found necessary for composting prunings. For Ul year-end wastes, a ratio of 62% vines, 13% used sawdust and 25% alder bark was recommended for in-vessel composting. Using conventional management techniques in greenhouse tomatoes, a similar yield using a 2:1 sawdust to amendment mix by could be achieved compared to conventional sawdust medium. Significant reduction of crown and root rot disease caused by Fusarium oxysporum f. sp. radicis- lycopersici in susceptible tomatoes was achieved by addition of the greenhouse compost amendment to seedling plugs or blocks, and by mixing with the sawdust medium. A mixture of 2:1 sawdust to amendment by volume was shown to be effective. The reduction in disease resulted in 74% improved yield over a full growing season under high disease pressure. IV TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES xi LIST OF FIGURES ...xv LIST OF EQUATIONS ...xix LIST OF ABBREVIATIONS xx ACKNOWLEDGEMENTS xxii CHAPTER I OVERVIEW OF THE THESIS PROJECT 1 1.1 INTRODUCTION 1 1.1.1 Vegetable Greenhouse Industry 1 1.1.2 Solid Wastes Problem 2 1.2 OBJECTIVES ' 4 1.2.1 Sustainable Development 4 1.2.2 Hypothesis 7 1.2.3 Specific Objectives 7 1.2.3.1 Bio-Conversion of Vegetable Greenhouse Wastes 7 1.2.3.2 Re-Utilization of the Bio-converted Products 8 1.3 REFERENCES 10 CHAPTERII WASTES TRACKING, WASTES CHARACTERISTICS AND GREENHOUSE 12 11.1 BACKGROUND 12 11.2 SPECIFIC OBJECTIVES 1 13 11.3 METHODOLOGY 13 11.4 RESULTS 14 V 11.4.1 Plant Wastes 14 11.4.2 Non-plant Wastes .....16 11.5 DISCUSSIONS 19 II. 5.1 Conventional Disposal 19 11.6 CONCLUSIONS ...21 11.7 REFERENCES .'..22 CHAPTER III MATERIALS HANDLING 23 111.1 BACKGROUND AND LITERATURE RESEARCH 23 111.2 SPECIFIC OBJECTIVES 28 111.3 METHODOLOGY 28 111.4 RESULTS AND DISCUSSIONS 30 III, 4.1 Shredder Test 1 30 III. 4.2 Shredder Test 2 31 111.5 CONCLUSIONS 38 111.6 REFERENCES 40 CHAPTER IV UTILIZATION OF BIO-DEGRADABLE TWINES 42 IV. 1 BACKGROUND AND LITERATURE RESEARCH.... 42 IV. 1.1 Trends in Bio-degradable Plastics 42 IV. 1.2 Use of Bio-degradable Plastics in Horticulture 45 IV.2 SPECIFIC OBJECTIVES 48 IV.3 METHODOLOGY .......48 IV.3.1 Selection of Twine Candidates 49 IV.3.2 Utilization of Selected Twines 54 TV.3.3 Measurement of Changes in Physical Properties of Twines 56 vi IV.3.4 Compostability 59 IV.3.5 Statistical Analysis 61 IV.4 RESULTS AND DISCUSSIONS 62 rV.4.1 On-site Observations and Surveys 612 IV.4.2 Linear Density 65 rv.4.3 Tensile Strength Test 67 IV. 4.4 Compostability study 69 IV.5 CONCLUSIONS 73 IV. 6 REFERENCES 76 CHAPTER V BIO-CONVERSION PROCESS < ...79 V. 1 BACKGROUND AND LITERATURE RESEARCH 79 V. 1.1 Composting Methods 79 V . l .2 Criteria and Specification of Composting : 85 V.2 SPECIFIC OBJECTIVES 98 V.3 METHODOLOGY 99 V.3.1 Lab Scale Experiment System Design 99 V.3.2 Pilot Scale Experiment System Design 108 V.4 RESULTS AND DISCUSSIONS 114 V.4.1 Lab Scale Composting Data 114 V.4.1.1 Labi Test 114 V.4.1.2 Lab 2 Test 119 V.4.1.3 Lab 3 Test 123 V.4.1.4 Lab 4 Test 128 V.4.1.5 Lab 5 Test 136 V.4.1.6 Lab 6 Test 145 V.4.1.7 Lab 7 Test 150 V.4.1.8 Lab 8 Test 154 V.4.1.9 Lab 9 Test 159 V.4.2 Pilot Scale Composting Data 167 V. 4.3 Cost Analysis of Greenhouse in-situ Composting Systems 182 V.5 CONCLUSIONS 190 V. 6 REFERENCES . 193 CHAPTER VI COMPOST QUALITY 203 VI. 1 BACKGROUND AND LITERATURE RESEARCH 203 VI.2 SPECIFIC OBJECTIVES 208 VI.3 METHODOLOGY 209 VI.4 RESULTS AND DISCUSSIONS 212 VI. 4.1 Compost Analysis 212 VI.4.2 Bioassays on Compost Extract 218 VI. 4.3 Bioassays on Wastes, Leachate and Condensate 228 VI.5 CONCLUSIONS 237 VI. 6 REFERENCES 238 CHAPTER VII UTILIZATION OF COMPOST .....242 VII. 1 BACKGROUND AND LITERATURE RESEARCH .....242 VII.2 SPECIFIC OBJECTIVES 243 VII.3 METHODOLOGY 243 VII. 3.1 Growing Media Analysis 243 VII.3.2 Growing Trial 245 VII.3.3 Fruit Quality . 249 VII.4 RESULTS AND DISCUSSIONS , 250 VII.4.1 Growing Media Tests 250 VII.4.1.1 Pre-Season Media Analysis 250 VII.4.1.2 Post-Season Media Analysis 254 VII. 4.2 Growing Trial , 256 VII.4.2.1 Fruit Yield..... 256 VII.4.2.2 Fruit Quality.... 257 VII.4.2.3 Plant Growth 258 VII.4.2.4 Disease 258 VII. 4.2.5 Medium Acidity 258 VII.5 CONCLUSIONS ...260 VII. 6 REFERENCES 262 CHAPTER VIII DISEASE SUPPRESSION TRIALS 264 VIII. l BACKGROUND AND LITERATURE RESEARCH 264 VIII.2 SPECIFIC OBJECTIVES 265 VIII.3 METHODOLOGY 266 VIII. 3.1 Seedling Tests 267 VIII.3.2 Preparation of FORL Spores ....268 VIII.3.3 Seedling Scoring 268 VIII.3.4 Yield Test 273 VIII.4 RESULTS AND DISCUSSIONS ...277 VIII.4.1 Tomato Seedling Tests 277 VIII. 4.1.1Test3 277 ix VIII.4.1.2Test4 : 279 VIII.4.1.3Test6 279 VIII.4.2 Yield Test 282 VIII.4.3 Microbial Analysis 285 VIII.5 CONCLUSIONS.... 286 VIII. 6 REFERENCES 287 CHAPTER IX MARKETING STUDY OF COMPOST AND COMPOST EXTRACT 289 IX. 1 BACKGROUND 289 IX.2 METHODOLOGY 290 LX.3 RESULTS AND DISCUSSIONS .....293 LX.3.1 Market for the Organic Amended Products 293 LX.3.1.1 Retail Market 293 LX.3.1.2 Floriculture Market 299 LX.3.1.3 Nursery Market 303 IX.4 CONCLUSIONS ...309 IX.5. REFERENCES 313 OVERALL CONCLUSIONS & RECOMMENDATIONS FOR FUTURE WORK 314 APPENDIX A Waste Tracking Data Sheet 319 APPENDIX B Greenhouse Solid Waste Management Survey 320 APPENDLX C Sample Aeration Requirement Calculations 324 APPENDLX D Rugters Temperature Feedback: C-Program Code of control Algorithm and Flowchart 327 APPENDLX E Linear Temperature Feedback: C-Program Code of control Alorithm and Flowchart 331 APPENDIX F Sample Fertilizer Recipe of Growth Trial (Chapter VII) 336 c . ' • X APPENDIX G ANOVA - Plant Growth (for Chapter VII) .. .337 APPENDIX H ANOVA - Test 3 (for Chapter VIII) 343 APPENDIX I ANOVA - Test 4 (focr Chapter VIII) 347 APPENDIX J ANOVA - Test 6 (for Chapter VIII) 354 APPENDIX K ANOVA - Disease suppression yield test (for Chapter VIII) 365 xi LIST OF TABLES Table II. 1 Organic wastes generation 16 Table II.2 Greenhouse organic waste characteristics 16 Table III.l Comparison of shredding pepper and tomato year-end wastes 36 Table IV. 1 Criteria for twines 51 Table IV.2 Selected twine candidates and their compositions 54 Table IV.3 Typical greenhouse conditions 55 Table IV.4 Settings of tensile tester 59 Table IV.5 Time in service of each type of twine 63 Table IV.6 Evaluation matrix 74 Table V . l Comparison of the three biological stabilization systems 85 Table V.2 C/N ratio of various wastes 87 Table V.3 List of lab tests 105 Table V.4 Lab 1 Compost feedstock recipes 116 Table V.5 Equations for the time in hours (t) required to attain no viable organisms (equal to 121og io )* of different pathogens at different temperatures (T) above 45 °C. 117 Table V.6 Loss of carbon over composting period as calculated via mass balance (with manure vs without manure) 119 Table V.7 Lab 2 compost feedstock recipes 121 Table V.8 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) 122 Table V.9 Lab 3 Compost feedstock recipes 124 Table V. 10 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) - Lab 3 126 Table V . l 1 Comparison of ammonia emissions 126 Table V. l2 Different aeration methods for Lab 4 128 xu Table V. 13 Leachate and condensate from reactors 131 Table V.14 Vertical variation (top vs bottom) in temperature and moisture content in reactors with different aeration and cooling 131 Table V. 15 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) 133 Table V. 16 Lab 5 Compost feedstock recipes 137 Table V.17 Cumulative amounts of leachate and condensate from reactors 142 Table V. 18 Loss of carbon over composting period as calculated via mass balance 143 Table V.19 Loss of nitrogen over composting period as calculated via mass balance... 144 Table V.20 Lab 6 Compost feedstock recipes .147 Table V.21 Loss of carbon over composting period as calculated via mass balance 150 Table V.22 Lab 7 Compost feedstock recipes 1521 Table V.23 Loss of carbon over composting period as calculated via mass balance... 1554 Table V.24 Lab 8 Compost feedstock recipes 155 Table V.25 Leachate and condensate from reactors 157 Table V.26 Loss of carbon over composting period as calculated via mass balance 157 Table V.27 Loss of nitrogen over composting period as calculated via mass balance... 159 Table V.28 Lab 9 Compost feedstock recipes , .....160 Table V.29 Leachate and condensate from reactors 162 Table V.30 Loss of carbon over composting period as calculated via mass balance 164 Table V.31 Loss of nitrogen over composting period as calculated via mass balance... 165 Table V.32 Pilot scale compost feedstock recipes 167 Table V.33 Pilot composting experimental details and results summary 168 Table V.34. Conventional disposal costs 184 Table V.35 Composting system component costs 185 Table V.36 Annual cost analysis example 189 Table VI. 1 Analysis from different lab tests 216 Table VI.2 Growing medium and amendment analysis (compost) 217 Table VI.3 Bioassay results from Lab 5 218 xm Table VI.4 Bioassay results from Lab 6 (t-test compared to distilled water) 220 Table VI.5 Bioassay results from Lab 7 (Cress seeds) 222 Table VI.6 Bioassay results from Lab 7 (Tomato seeds) ..223 Table VI.7 Bioassay results from Lab 8 225 Table VI.8 Bioassay results from Lab 9 .....226 Table VI.9 Analysis of wastewater from rotten wastes, leachate, and condensate ..228 Table VI.10 Compost leachate and condensate analysis (from Pilot 1, 2, and 3) 236 Table VII. 1 Growing media analysis methods 244 Table VII.2 Growing trial treatments - medium, nutrient solution & number of rows..245 Table VII.3 Growing trial nutrient solutions - EC, recipe and ammonia concentration 246 Table VII.4 Pre-season physical characteristics of the growing media 250 Table VII.5 Particle size of different media (Pre-season) 251 Table VII.6 Pre-season nutrient characteristics of the growing media materials 252 Table VII.7 Pre-season chemical characteristics of the growing media materials 253 Table VII.8 Microbial counts of compost amendment 253 Table VII.9 Post-season media density and porosity 254 Table VII. 10 Post-season media chemical characteristics 255 Table VII. 11 Tomato yield, size and culls for the trial growing season 256 Table VII.12 Tomato shelf life results 257 Table VIII. 1 Recipes and dates of compost amendment batches 267 Table VIII.2 Seedling test root scoring 269 Table VIII.3 Treatments and conditions for seedling test 3 - amendment as plug and covering 270 Table VIII.4 Treatments and conditions for seedling test 4 - amendment as plug and covering, second trial 271 Table VIII.5 Treatments and conditions for seedling test 6 - amendment comparison, sandwich covering method 272 Table VIII.6 Procedures - Disease suppression yield test..... 274 Table VIII.7 Yield test treatments 275 xiv Table VIII.8 Results summary for test 3 - amendment as plug and covering 278 Table VIII.9 Results summary for test 4 - amendment as plug and covering, second trial 279 Table VIII. 10 Results summary for test 6 - amendment comparison, sandwich covering method 280 Table VIII. 11 Diseased plants during yield test by treatment 282 Table VIII. 12 Tomato yield fruit size results by treatment 284 Table IX. 1 Wholesale and retail prices for premium soil conditioner products 294 Table IX.2 Bulk prices for selected growing media 306 X V LIST OF FIGURES Figure LI Dumping of plant debris beside greenhouses 2 Figure 1.2 Sustainable greenhouse production with bio-conversion of solid wastes 6 Figure II. 1 Monthly organic wastes production 15 Figure III. 1 Hammermill shredder - Bear Cat 30 Figure III.2 Teagle Tomahawk 100 Bale Shredder 32 Figure III.3 Modification of the Teagle Shredder - more blades added 33 Figure III.4 Modification of the Teagle Shredder -Baffles and Teflon Lining Added 34 Figure III.5 Modification of the Teagle Shredder- Larger and Shorter Hopper, and Propane Motor 34 Figure III.6 Greenhouse wastes handling process 37 Figure IV. 1. Chemical formula of PHB, PCL and aliphatic-aromatic copolyesters 44 Figure IV.2 Jute twine for plant support 55 Figure rV.3 Electronic tensile tester used in laboratory measurements 58 Figure IV.4 Linear density of alternative twines 66 Figure IV.5 Tensile strength of alternative twines 67 Figure 1V.6 The bio-degradable twines (EcoPLA®2000D)in a screen-net before being placed into the composter 71 Figure IV.7 Example of fragmented twines (EcoPLA®2000D)after being placed into a composter 71 Figure IV.8 Linear Density of bio-degradable twines 72 Figure IV.9 Tensile Strength of the bio-degradable twines.... 72 Figure V . l . Temperature and pH variation during natural composting process 94 Figure V.2 Lab scale composter (Scale 11.5:1) 100 Figure V.3 Lab scale reactor (Outside) 101 Figure V.4 Lab scale reactor (Inside) ..102 Figure V.5 Pilot scale bio-conversion reactor design 110 xvi Figure V.6 Pilot scale composter (outside) I l l Figure V.7 Pilot scale composter (inside - installing air piping) I l l Figure V.8 Temperature profiles of Lab 1 116 Figure V.9 Lab 2, 3,4 Reactor design (with and without air recirculation) 120 Figure V.10 Temperature profiles of Lab 2 121 Figure V . l 1 Temperature profiles of Lab 3 .124 Figure V.12 Temperature profiles of Lab 4 129 Figure V . l 3 Temperature profiles of Lab 5 .....138 Figure V.l4 Total carbon vs time 140 Figure V.l5 Total nitrogen vs time 140 Figure V. l6 C:N ratio vs time 141 Figure V.17 MC (wet basis) vs time 141 Figure V.l8 Ammonia vs time 141 Figure V. 19 Nitrate vs time 141 Figure V.20 Phosphate vs time 141 Figure V.21 pHvstime 141 Figure V.22 Temperature profiles of Lab 6 148 Figure V.23 Lab 6 Reactor A Oxygen concentration (%) vs. time 148 Figure V.24 Temperature profiles of Lab 7 151 Figure V.25 Lab 7 Reactor A Oxygen concentration vs. time 153 Figure V.26 Temperature profile of Lab 8 155 Figure V.27 Temperature profiles of Lab 9 160 Figure V.28 Pilot 1 Temperature profile 170 Figure V.29 Pilot 2 Temperature profiles 171 Figure V.30 Pilot 3 Temperature profiles 172 Figure V.31 Pilot 4 Temperature profiles 173 xvii Figure V.32 Pilot 4 Oxygen profiles 174 Figure V.33 Pilot 4 Temperature profile of curing 175 Figure V.34 Pilot 5 Temperature profiles , 176 Figure V.35 Pilot 6 Temperature profiles 177 Figure V.36 Pilot 6 Oxygen profile 178 Figure V.37 Pilot 7 Temperature profiles 179 Figure V.38 Pilot 8 Temperature profiles ; 180 Figure VI. 1 Bioassay on Lab 5 - 1-day cress seeds 219 Figure VI.2 Bioassay on Lab 6 - 1-day cress seeds 221 Figure VI.3 Bioassay on Lab 7 - 1-day cress seeds 223 Figure VI.4 Bioassay on Lab 7 - 6-day tomato seeds 224 Figure VI.5 Bioassay on Lab 8 - 1-day cress seeds 225 Figure VI.6 Bioassay on Lab 9 - 1-day cress seeds 227 Figure VI.7 Bioassay on rotten greenhouse waste - 1-day cress seeds 229 Figure VI.8 Bioassay on leachate, 1-day cress seeds. 230 Figure VI.9 Bioassay on leachate (10% -100%), 6-day tomato seeds 231 Figure VI. 10 Bioassay on leachate(10% - 20%), 6-day tomato seeds 231 Figure VI.l 1 Bioassay on 10% leachate from pilot 1-3 (1-day cress seeds) 232 Figure VI. 12 Bioassay on 10% leachate from pilot 1-3 (6-days cress seeds) 233 Figure VI. 13 Bioassay on condensate, 1-day cress seeds 234 Figure VI. 14 Bioassay on condensate, 6-day tomato seeds 235 Figure VII. 1 Test greenhouse layout 247 Figure VII.2 Test greenhouse for growth trial .....248 Figure VII.3 Particle size distribution of different growing media ; 251 Figure VII.4 Drain pH comparison of treatments 2 (sawdust) and 4 (sawdust plus amendment), both using nutrient solution N2 (higher ammonia) 259 Figure VIII.l Experimental layout of yield test 274 Figure VIII.2 Greenhouse site of disease suppression yield test 28075 Figure VIII.3 Sample seedlings with FORL (left: treatment 6, right: treatment 2) 278 xvi i i Figure VIII.4 Cumulative marketable tomato yield for sawdust medium vs. sawdust and amendment mixture, both inoculated with FORL 285 xix LIST OF EQUATIONS Equation V . l Composite C:N ratio 87 Equation V.2 Free air space of compost mass 90 Equation V.3 Porosity of compost mass ....90 Equation V .4 Composite moisture content of compost mass 92 Equation VI. 1 Germination index 210 Equation VI.2 Mest 211 XX LIST OF ABBREVIATIONS ASTM American Society for Testing and Materials BC Province of British Columbia BC MAFF British Columbia Ministry of Agriculture, Food and Fisheries BC MWLAP B.C. Ministry of Water, Land and Air Protection BC Reg British Columbia Provincial Government Regulation BTA Combination of 1,4-butanediol, terephthalic acid and adipic acid CCP Combined anaerobic/aerobic Composting Process CEC Cation Exchange Capacity C:N Ratio Carbon to Nitrogen Ratio DMS Dimethyl Sulfide Duratiori4o°c Duration of the temperature above 40°C Duration45°c Duration of the temperature above 45°C Duration55°c Duration of the temperature above 55°C EC Electrical Conductivity FAS Porosity or Free Air Space FORL Fusarium oxysporum f. sp. radicis. lycopersici GI Germination Index GVRD Greater Vancouver Regional District HDPE High Density Polyethylene MC Moisture Content OAP Organic Amendment Products P Porosity Pb Bulk Density p p Particle density PBT Poly-butylene terephthalate PCL Poly-e-caprolactone PET Ethylene terephthlate PFRP Process to further reduce pathogens PHB Polyhydroxybutyrate PLA Polylactic acids PP Polypropylene PS Polystyrene PVC Polyvinyl chloride SAR Systemic acquired resistance TC Thermocouple TC Total Carbon TN Total Nitrogen TEMP m a x Maximum temperature TIME55°c Time required for temperature to reach 55°C UBC University of British Columbia USDA U.S. Department of Agriculture USEPA U.S. Environmental Protection Agency VARP Vector Attraction Reduction Process XXII ACKNOWLEDGEMENTS I would like to thank the following people and agencies that contributed to the completion of this research project and dissertation. Dr. Victor Lo and Dr. Richard Branion of Chemical and Biological Engineering Department at the University of British Columbia (UBC) as my supervisors for the assistance to identify this important research topic, to share their extensive knowledge and experince on composting process, and to provide many technical resources related to this research project. Dr. Anthony Lau of Chemical and Biological Engineering Department (UBC) and Dr. Peter Jolliffe of Plant Science Department (UBC) as my committee members for their encouragement, timely guidance and unreserved support throughout my Ph.D. program. The financial support from the Natural Science and Enginneering Research Council (NSERC) of Canada (Industrial Postgraduate Scholarship), Science Council of British Columbia (G.R.E.A.T. Scholarship), BC Greenhouse Growers Association (the industrial sponsor), National Research Council (T.A.P. funding), Investment Agricultural Foundation (I.A.F. funding), and the University of British Columbia are sincerely acknowledged. The undertaking of my Ph.D. would not have been possible without the encouragement and patience of my wife, Amy. The confidence and encouragement I received from her are greatly appreciated. CHAPTER I OVERVIEW OF T H E THESIS PROJECT 1.1 INTRODUCTION 1.1.1 VEGETABLE GREENHOUSE INDUSTRY Every year, more and more vegetable supply comes from greenhouse vegetable production regardless of seasonal changes. In fact, the greenhouse vegetable industry has repeatedly achieved positive growth in sales in the last ten consecutive years. With the increasing number of greenhouse operations, environmental concerns regarding the industry are also on the rise. The greenhouse vegetable industry is one of the fastest growing sectors of agriculture in British Columbia. The crops produced by greenhouses include tomatoes, cucumbers, sweet peppers and lettuces. In 1986, there were only about 36 hectares of vegetable greenhouses in B.C. By the end of 2001, there were about 80 commercial greenhouses covering 200 hectares of agricultural land in B.C. According to Statistics Canada, income from the sales of greenhouse vegetables in BC increased from $43 million in 1993 to $200 million in 2001 (Agriculture Statistics, 2002). Tomato was the largest income earner, representing half of the greenhouse vegetable total income, while pepper and cucumber were the second and third in production area. In 2001, the industry directly employs over 1000 workers and spends approximately $20 million on labour costs. Unfortunately, the BC greenhouse industry also produces about 30,000 to 40,000 tonnes of wastes per year according to our survey in 1999. The bulk of the wastes come 2 from the residues of the plants that are uprooted at the end of the growing season in November. Other organics wastes are generated during the growing season which consist of damaged or rotten crops and primings. 1.1.2 SOLID WASTES PROBLEM Most of the solid wastes from greenhouses are either piled up on lands beside the greenhouses or trucked to landfills. Figure 1.1 shows an example of how those wastes were being stored at the backyard of the greenhouses. Piles of untreated organic wastes not only led to odour and pest problems, but also created a leachate problem, where high ammonia concentrations could pose a threat to watercourses (Robinson, 1990). Landfilling is also not considered a good waste management practice because it puts to waste two valuable resources, namely, the nutrient-rich plant wastes and the land it occupies. Besides, it costs the greenhouse operators millions in expenses in trucking and tipping fees each year. Incineration or local burning of greenhouse wastes is not suitable due to the high moisture content and generation of dioxin from plastics (Mitsuo, 2001). Figure 1.1 Dumping of plant debris beside greenhouses 3 On the other hand, bio-conversion of greenhouse wastes promises to be an environmentally sound practice. Firstly, bio-conversion would convert the biodegradable wastes into a useful product - compost. Secondly, it would make use of abundant local waste products such as sawdust and wood chips as bulking agents. Finally, the finished compost product would be a usable product with agricultural and horticultural value (Eitzer, 1997). Large commercial composting plants are already in operation in B.C. Unfortunately, the large operations usually accept wastes from greenhouses only if they have already been cleared of materials, such as plastic clips, twines, etc. This requirement has prompted many growers to simply truck their wastes to landfill areas where there is no such restriction. Centralized composting, like landfilling, also entails huge hauling and tipping fees. For a 4 ha greenhouse like Hazelmere greenhouse in our survey, the operator claimed that the expenses could accumulate to about $100,000 per year. For these reasons, greenhouse operators are looking into the possibility of operating their own bio-conversion facilities on site. Such an option would eliminate excessive hauling and tipping fees. Although the environmental impacts of several small composting sites remain to be assessed and compared to those of a single large operation, it is obvious that on-site bio-conversion would be a more environmentally friendly alternative than landfilling (Environmental Guidelines for Greenhouse Growers in B.C., 1994). The fact remains that, so far, no greenhouse operator has resorted to composting wastes on site. Such reluctance is founded mainly on the fear that the practice could give 4 rise to problems such as odour pollution, groundwater contamination, infestation of flies and mice, spread of pathogens, and outbreak of plant disease. To a lesser degree, the hesitation may be due to uncertainties over technological issues such as which bio-conversion method is to be used or what bio-conversion techniques are to be adopted. Moreover, there are strict regulations for organic matter recycling in B.C., especially for on-farm composting. For example, the composting process has to achieve 55 °C for a minimum of 3 days for pathogen reduction and waste materials derived from other non-agricultural operations are not allowed to be treated on the farm (B.C. Reg. 334/93). Compostable materials and recycling materials continue to be a waste until dealt with in accordance with the regulations; therefore, are not allowed to be sold as retail-grade organic matter or re-used in farm applications. This study represents the first concerted effort to address all of the above concerns. It hopes to show that bio-conversion on-site, if properly done, can be a cost-effective means of greenhouse waste management without becoming a threat to the safety and well-being of the workers, the crops, and the environment. 1.2 OBJECTIVES 1.2.1 SUSTAINABLE DEVELOPMENT Sustainable Development, as defined ten years ago by United Nations Commission on Environment and Development, means "meeting the basic needs of all the world's people today without compromising the ability of the future generations to 5 meet their needs" (Tebo, 1997). With the increased demands created by growing populations and rising incomes, sustainable agricultural development emphasises the need to enhance agricultural productivity in a manner that provides affordable, efficient and healthy diets to all at the lowest environmental cost (UNCED, 1992). The 2000s have emerged as a decade of the environment. An important dimension of this is a societal focus on reducing the volume of materials that are landfilled (Arent, 1996). The initial focus of the Canadian Federal government was to reduce the volume of solid wastes generated by large firms. The present focus is on local units of governments (e.g. GVRD) developing recycling and other programs to reduce the volume of household wastes. These efforts will be followed by greater governmental focus on wastes generated by small and medium businesses, including the greenhouse industry. In the last ten years, there has been an awakening of environmental consciousness in the agricultural and horticultural industry, such as the discharge of nutrients, pesticides, and waste materials. In Holland, a recent covenant between horticultural employers and the Dutch government for a sustainable development of greenhouse between 2000 and 2010 enables the growers to achieve the environmental targets in their own way, if they make a specific plan for their greenhouses (Erik, 1999). In Canada, we need a new era of agricultural growth in which the greenhouse industry plays a strong role, reaps the benefits and breaks the traditional industrial paradigm - creating wealth without regard to environmental consequences. The environmental consequences of existing greenhouse production - deterioration of agricultural lands, contamination of groundwater and spread of plant diseases - all contribute to reducing the ability of the environment to provide 6 resources for crop production; therefore, diminishing the potential well-being of future productions. The overall objective of the thesis project is to develop a sustainable growing practice in the vegetable greenhouse industry. It can be achieved by two steps (see Figure 1.1). First, an appropriate technology will be developed for the bio-conversion biodegradable wastes generated by the vegetable greenhouse. Second, the products from the process will be utilized in the greenhouse as organic growing media and disease suppression agent. By doing so, the greenhouse industry would not only achieve a sustainable growing practice with minimal wastes output, but also could benefit from cost savings in waste trucking, fertilizers and growing medium and increase in crop yield as well. Figure 1.2 Sustainable greenhouse production with bio-conversion of solid wastes 7 1.2.2 HYPOTHESIS a) With the constituents of the greenhouse wastes and proper process controls, the bio-conversion of the wastes can proceed with minimal addition of bulking agents, and no supplemental heat needed to achieve the pathogen reduction requirement - a minimum of at least 55 °C for three days (B.C. Reg 334/93). b) The bio-converted greenhouse wastes can be re-utilized in the greenhouse as growing medium or amendments, which will provide nutrients for the plants, including the macro-nutrients (N,P,K) and trace elements. c) The products can also be used as disease suppressive agents for certain plant diseases, and such disease suppression properties can be correlated to the bio-conversion process. 1.2.3 SPECIFIC OBJECTIVES 1.2.3.1 Bio-Conversion of Vegetable Greenhouse Wastes This part of the project concerns the development and demonstration of an appropriate technology for the bio-conversion of wastes generated by greenhouses in B.C. The specific objectives are: • Characterize the various types of wastes generated in greenhouses. • Test the effects of relevant parameters on bio-conversion efficiency. • Assess the quality of the finished compost product. • Assess the quality of some of the by-products from the process and the products, e.g. leachate, condensate and extracts. 8 • Investigate the technical possibility of using bio-degradable plastics for greenhouse production. • Investigate the feasibility of adopting box-type (in-vessel) bio-conversion technology for on-farm operation and to evaluate the economic feasibility of the bio-conversion technology. 1.2.3.2 Re-Utilization of the Bio-converted products This part of the project aims at evaluating the benefits of utilizing the bio-converted products as high-quality organic medium amendment for growing vegetable crops in greenhouses. The specific objectives are: • Analyze and compare the physical, chemical, and microbiological characteristics of the organic amendments with conventional sawdust growing medium. • Demonstrate the feasibility of using the organic amendments in a commercial greenhouse, and measure the effect on crop health and yield. • Investigate the disease-suppression abilities of the organic amendments on greenhouse vegetable seedlings. • Investigate the disease suppression ability of the organic amendments on greenhouse crop yield and health. 9 The following diagram shows the overall framework of the project. Waste: -> Materials Handling: -> Bio-Conversion: -> Tracking Shredding Lab Scale Process Existing Practice Mixing Pilot Scale Process Characteristics Handling Process (Chapter V) (Chapter II) (Chapter III) Biodegradable plastics(Chapter IV) -> Product: -> Compost Utilization : Compost Quality Analysis Physical and Chemical Characteristics Leachate Analysis Growth Test (Chapter VII) Bio-assay Disease Suppression Test (Chapter VIII) (Chapter VI) Marketing Studies (Chapter IX) 10 1.3 REFERENCES Agriculture Statistics. (2002) British Columbia Ministry of Agricultural, Food and Fisheries. # 22-202. Arent, G.L. (1996) The Greenhouse Wastestream. HortTechnology. 6(4): 365-366. B.C. Reg. 334/93. (2002) Organic Matter Recycling Regulation. Waste Management Act, s. 57. Eitzer, B.D., Iannucci-Berger, W.A., Mark, G., Zito, C. (1997) Fate of toxic compounds during composting. Bulletin of Environmental Contamination and Toxicology. 58(6):953-960. Erik, A. (1999) Closed soilless growing systems : a sustainable solution for Dutch greenhouse horticulture. Water Science and Technology. 39(5): 105-112. Environmental Guidelines for Greenhouse Growers in British Columbia. (1994) B.C. Ministry of Agriculture, Fisheries and Food and B.C. Federation of Agriculture and the Greenhouse Industry of BC. Government of British Columbia. Mitsuo, C. (2001) Status of Organic Recycling and Composting in Japan. Biocycle. 42(6) : 61-62. Robinson, H.D. (1990) On-site treatment of leachates from landfilled wastes. Journal of the Institution of Water and Environmental Management. 4(1): 78-89. Tebo, P.V., Rittenhouse, D.G. (1997) Sustainable Development : Creating Business Opportunities at DuPont. Corporate Environmental Strategy 4(3) : 5-12. 11 UNCED. (1992) United Nations Conference on Environment and Development, 1992. Agenda 21-An Action Plan for the Next Century, New York. 12 CHAPTER II WASTES TRACKING, WASTES CHARACTERISTICS AND GREENHOUSE SURVEYS II.l BACKGROUND This section deals with the wastes tracking throughout the growing season (February to December) and the characteristics of the wastes. The quantity of wastes produced per hectare of greenhouse per year was estimated. The waste materials include: • Daily prunings (from deleafing and truss breaking) -BIODEGRADABLE • Daily fruit rejects • Year-end plant wastes • Year-end used sawdust — • Plastic twine -> investigating biodegradable twine • Rock wool -> is attached and shredded with the year-end used sawdust • Plastic films and bags -> can be easily separated and trucked to landfill. Some greenhouses have been investigating the use of bio-degradable films and bags. 13 11.2 SPECIFIC OBJECTIVES • To understand the characteristics of the solid wastes produced from the vegetable greenhouse industry. • To understand the common waste management practices in the industry. 11.3 METHODOLOGY Generation of organic waste was studied in detail for Hazelmere Greenhouse (pepper and tomato), located at 17830-16th Avenue, Surrey, BC. Appendix A shows a waste tracking sheet that was used for monitoring the organic wastes from Hazelmere Greenhouses Ltd. (2 ha peppers and 2.4 ha tomatoes). As one of the objectives of this project is to come up with a feasible solution for treating the greenhouse wastes in B.C., we might need to know the existing practices of the operators in different greenhouses and their preference in terms of land use and cost. Therefore, different greenhouses were visited and surveyed, done either on site or by phone. The principal aim of the survey was to understand other greenhouses' wastes streams and current practices in wastes handling, so as to understand their concerns about plant disease and the possibility of incorporating "Composting" into their growing practice. Surveys were done for 8 other vegetable greenhouses: 4 tomato, 3 pepper, and 1 cucumber. A sample of the survey is shown in Appendix B. 14 II.4 R E S U L T S II.4.1 PLANT WASTES The annual amount of organic wastes generated from each hectare of greenhouse was about 150-200 tonnes according to the survey. Organic wastes from the vegetable greenhouses can be grouped into 4 major categories as follows (see Figure II. 1). a. Prunings • primarily leaves, small stem portions, and a limited amount of fruit • generated throughout the growing season • high moisture content, potentially rotting and producing odour if stored too long b. Rejects • fruit culls • generated throughout the production season, peaking during summer months • very high moisture content c. Year End • whole plants left at end of growing season • lower moisture content, contain more woody material (lignin and cellulose) • large, tough stems require size reduction • typically contaminated with plastic twine and clips 1 5 d. Sawdust Growing Medium • usually hemlock or yellow cedar • used by most but not all vegetable greenhouses • low moisture content, may be contaminated with plastic bags The tomato crop was found to have slightly higher volume of wastes than pepper. Table II.I summarizes this data (note the survey average includes Hazelmere). Table II.2 shows chemical characteristics of the wastes. Figure II.I Monthly organic wastes production (Top Left: Used Sawdust Medium, Top Right: Leaves and Prunings, Bottom Left: Year-end vines, Bottom Right: Fruit Rejects) 16 Table II. 1 Organic wastes generation Waste Annual Generation Annual Generation Annual Generation T/ha(Hazel mereGrn) T/ha (Average from T/ha(Average from Survey) Survey) Tomato Pepper Tomato& Pepper Cucumber Pruning 34.6 41.7 31.9 n/a Rejects1 22 4.4 13.2 n/a Year End 55.2 45 55.3 40 Sawdust medium 81.6 68 74.6 13.3 Total 193.4 159.1 175 n/a Table II.2 Greenhouse organic waste characteristics Moisture Content % TC % (wb) TN%(wb) C : N Leaves 87.15±2.38 4.28±0.86 0.75±0.08 5.7110.89 Vines 82.02±2.89 6.56±1.23 0.35±0.06 18.7412.59 Sawdust 73.13±1.58 15.41+2.34 0.2110.05 73.3815.66 Roots 86.64±1.71 4.19±0.57 0.4910.14 8.5710.64 Note: 1. vlean values of five determinations ± standard deviations. 2. The leaves, vines and roots were based on greenhouse tomato crop. II.4.2 NON-PLANT WASTES Apart from plant wastes, the generation of non-plant wastes from the greenhouses was also investigated in order to get a full picture of the waste stream. The non-plant waste materials generated in a vegetable greenhouse production can be grouped into four major categories as follows. 1 Rejects estimation: tomato - 4% * 55kg/m3 production = 22 T/ha; pepper - 2% * 22kg/m3 production = 4.4 T/ha; average = 13.2 T/ha 17 Plastic Films and Bags constitute about 26% of the total wastes low density polyethylene ground cover and sawdust bags a number of greenhouse materials, such as fertilizers, are packaged in plastics not much in the way of recycling or recovery options is available. Currently, they are either trucked to landfill or an incinerator Polystyrene Plastic (also known as Styrofoam) used in greenhouse operations include shipping containers, seedling trays and media components most of the growers have been re-using polystyrene shipping containers and seedling flats to the greatest extent possible Rockwool an inert, soilless medium manufactured from lava rock which is both non-polluting and non-degradable in the environment the material, used primarily in slab form, has to be disposed after each growing season as its water retention characteristics change the medium contains the concentrations and mix of plant nutrients found in the irrigation water there are currently no recycling or recovery options available to greenhouse growers in BC for the disposal of rockwool 18 d. Plastic Twines • plastic twines are used to tie the stems of the plants to the crop wires at 3.5 metres high • normally, they are removed and wasted after the end of the crop season • they create nuisance environmental concerns for the growers since they are difficult to separate from the plant debris 19 II.5 DISCUSSIONS II.5.1 CONVENTIONAL DISPOSAL A variety of organic waste disposal methods are currently used by growers, depending on the type of waste, greenhouse location, land availability, and grower preference. Based on the surveys and some site visits, typical disposal methods for the different types of wastes are as follows: • Prunings - land application (may be dried first), landfill, farm animal feed • Rejects - landfill, land application, farm animal feed • Year end - land application, landfill, burning • Sawdust - land application, pickup by landscapers or composting companies Land application may be on the grower's property, or a nearby farm or other private property. It may involve pre-drying pruning, spreading, or storing/composting in piles. Generally, growers will choose the least expensive disposal method; convenience is also an issue. In some cases, disposal methods appear to be not in accordance with recommended or legislated environmental practices (BC MAFF, 1994), in that the amount of waste applied or stored on the land is excessive, with no covering, potentially leaching into the surface or ground water. In many cases, it is likely that in the future, less waste can be applied to on-site land, due to waste build-up and potential leaching to surface or ground water. This would necessitate more expensive off-site disposal. Moreover, among the 8 greenhouses that were surveyed and visited, all of them had disease problems in their greenhouses (most 20 common ones were fusarium, pythium and boytrytis) and they were concerned about disease transmission from previous years' plant debris which had been stock piled beside the greenhouse. The collected information about the type of wastes and their daily/weekly volume is very important for designing the bib-conversion treatment system, which includes the sizing of the system, the waste handling, possible bulking agents required, storage, curing area, etc. Three terms used to describe reprocessing of waste are "re-use", "recycle", and "recover." Re-use means using the product again for its original purpose. Re-use of greenhouse materials wherever possible is always an important strategy to reduce the volume of wastes to be dealt with in greenhouse operations. However, not a lot of wastes can be re-used. Recycle refers to the process by which waste is remanufactured into a new product. Recover refers to any process that diverts wastes from landfills or incinerators. Greenhouse wastes consists of organic and inorganic materials, some of which can be recycled and recovered. The Resource Management division of BC's Ministry of Agricultural, Food and Fisheries (Environmental Guidelines, 1994) has been encouraging greenhouse growers to choose waste treatment options which reduce the volumes of waste generated and materials discarded by : • Composting organic debris • Substituting biodegradable materials for those that are not • Using recyclable materials • Using recoverable materials 21 II.6 CONCLUSIONS From the data and surveys, the following conclusions can be drawn: 1. Greenhouse wastes, when not handled properly, can leach out pollutants like ammonia and nitrate to the land and emit unpleasant rotten odors. Current practices are polluting the agricultural lands and groundwater. 2. The wastes may pose a disease transmission problem to the greenhouse which can account for crop losses as high as 30 to 50%, for example due to Fusarium crown rot in tomato. 3. The greenhouse industry has a continuous organic waste stream, and a consistent i waste characteristic. Almost 80 % of this waste is generated in a 3-month window near the end of the season. This suggests that plants and media from the remainder of the waste stream may be separated during this time. 4. Growers should be aware that disposal costs money. Waste-hauler tipping fees are increasing rapidly as the number of landfills declines and establishing new sites becomes a more politically difficult process. 22 II.7 REFERENCES BC Ministry of Agricultural, Food and Fisheries. (1994) Agricultural Waste Control Regulation and Code of Agricultural Practice for Waste Management. Environmental Guidelines for Greenhouse Growers in British Columbia. (1994) B.C. Ministry of Agriculture, Fisheries and Food and B.C. Federation of Agriculture and the Greenhouse Industry of BC. 23 CHAPTER III MATERIALS HANDLING III.l BACKGROUND AND LITERATURE RESEARCH Greenhouse year-end waste, mostly composed of stems, leaves, fruit rejects, prunings and twines, needs to be shredded to proper sizes prior to the bio-conversion process. The initial moisture content of the collected waste is about 75-85%. Such a high moisture content in the waste tends to pose difficulties for the ordinary hammer mill type of shredder. Besides polyethylene twines, which are attached to the plants for support, also create problems for high-speed hammer mill machines as the twines may wrap around the rotating shaft and stop the machine. The twines may also damage the bearings for some smaller machines. The wetter the organic material, the coarser should be the grind. Since finely ground materials give up moisture quickly, they become a soggy mass and cannot be properly aerated. Thus, it is necessary that the material be ground to the right degree of fineness, uniformity and degree of compression. Besides, it is also essential to know and understand the composting process to be employed so that the grinding operation can be tailored to it. Studies showed that depending on the type of raw materials to be processed, one or more of the following types of grinders and shredders could be chosen (Savage, 1981; Christopher, 1996; Gray, 1999; Glenn, 1999): a. garden chipper/hammermill; 24 b. granulator; c. tub grinders; d. high speed, high process volume shredders; e. shear shredders; f. medium-speed grinders. It has been suggested that material to be composted in a mechanical digester may be ground much more finely than that which would be windrowed. Composting employing forced air and occasional turning can also take advantage of finer grinds (Snell, 1991). Many garden chipper/hammermill have engine power rating ranges from 3.5 hp to 40 hp. Some machines are driven by the Power Take-Off (PTO) of a tractor while some smaller models are generally electric-powered and are best at shredding leafy green matter and small branches. Shredding is done by fixed-arm hammers. Such mechanisms spin at high speeds, pulverizing material that is fed into the machine (Ettlinger, 1994). Chipping, however, is accomplished by one or more fixed blades mounted on a rotary disc. The blades cut off material as it is fed through a chute. For the greenhouse's purpose, shredder machines are more suitable since they are better in handling soft materials such as leaves and twigs. However, the twines attached to the tomato plants may wrap around the shafts of those high-speed hammers of the shredders. A granulator uses sharp knives rather than blunt faced hammers, that produce a chopping and slicing effect resulting in a granular grind containing far less objectionable dust and fines. Although this machine does not utilize the exploding action that a hammer 25 mill does, it has a very serious disadvantage, which make this machine not suitable for the greenhouse's application. According to the chief operating officer of H.C. Davis Sons Manufacturing Company Inc.- Mr. Thomas McPherson, the maximum tolerance for moisture content of the input material is 12% while the greenhouse waste will easily exceed 80% (McPherson, 1999). Tub grinders are a high volume organic or wood waste processor. Their dump height can be as high as a house, and the width of the tub can range from 2 to 4 meters. In general, material is fed into a round hopper by either a conveyor system or a machine such as a wheel loader. Large tub grinders have their own knuckle boom grappler/loaders, and the hopper usually rotates to prevent clogging. Material passes through a hammermill and is discharged into a conveyor (Aquino, 1996). High speed, high process volume shredders can do several jobs simultaneously. Such a shredder can grind and shred different materials, mix them up and blow the material into windrows up to 7.5 meters high. This high speed shredder can process a wide array of materials like biosolids, sawdust, recycled compost, brush, pallets, logs, wet grass or even Christmas trees. This kind of machine can be powered by a 150 to 200 hp engine and can process up to 200 cubic yard of sludge per hour (Anonymous, 1991) . Although such a machine has a very high throughput and can handle both big branches or high moisture content waste (sludge for example), its large volume and high price hinder greenhouse operators from considering it a practical option. The shear shredder - properly described as a high torque, low speed industrial shear shredder is a newcomer to reduction technology. It has some advantages over the 26 hammermill, grinder and granulator. The industrial shear shredder has two or more counter-rotating shafts equipped with hooked knives capable of reducing a wide variety of materials from municipal solid waste to baled metals. There are couple of factors to be considered. First, one should not equate horsepower and shredder performance. Factors such as maximum knife tip force and shredder shaft speed are much more reliable indicators of a shredder's capability and capacity than mere horsepower. Usually, the knife tip force determines the type of materials such a shredder can handle while the shaft speed indicates the throughput of the shredder, provided it can shred that particular waste. Besides, the drive type is also an important issue to be considered. Usually, shredders under 200 hp have a single drive motor and shredders over 200 hp have a dual drive motor (Glass, 1997). Besides the knife tip force and shaft speed, one should also consider other design features like knife design, configuration of knives on the shredder shafts and method of feed. In knife design, engineers need to consider parameters such as width of knife, the number of hooks per knife, and the length and angle of the hooks to suit a specific application. A particular product must easily fit within the shredder feed opening to allow the shredder knives to grasp items to be shredded. Bulk loading using overhead cranes or front end loaders can increase labor savings and decreases the risk of injury caused by heavy lifting. Although shear shredders are easier and less costly to maintain than hammermills and others, it is also important to examine specific features provided by manufacturers. Features like hexagonal shredder shafts, adequate bearing and seal protection, stack 27 tightening systems and shredder drive configurations will help to keep maintenance and operating costs to a minimum. Medium-speed grinders are usually a single-rotor design and operate between 100 rpm and 400 rpm. Unlike the shear shredders which utilize shafts, the grinder consists of a round drum with multiple replaceable cutter inserts which cut against a fixed bed knife. Material exits through a screen on the discharge size and thus product sizes are usually consistent. A typical machine consists of a solid, machined-steel rotor with relatively small cutter tips, a charge hopper, a hydraulic feed ram and an electric motor with speed reducer. These grinders can operate with or without screens, and are used to shred plastics, electronic scrap, wood waste and non-ferrous metals. A common application of this kind of machine is for shredding plastic bottles and tires. It can also grind scrap wood into a fuel-quality product. The power rating of such macahines may go up to 400 hp and the rotor weights of up to 20,000 lbs (Newell, 1997). They rely on vertical feed, and, much like a hammermill, the rotor inertia is used to produce the work. But unlike a hammermill, a true cutting process occurs within the machine. 28 111.2 SPECIFIC OBJECTIVES • To find specific types of shredders or grinders that can handle high moisture content greenhouse waste and are able to shred twines. • To modify some existing machines to suit our purposes. • To design the waste feeding mechanism from greenhouse decropping to the bio-conversion reactor, e.g. whether shredding the waste in the greenhouse aisle or outside when the waste is piled up. 111.3 METHODOLOGY In order to find a suitable shredder and/or grinder for the greenhouse application, a few different suppliers were contacted. Firstly, the greenhouse operators were not in the composting business, and were not prepared to spend a lot for a shredder to process their waste. So, a machine of below $10,000 Cdn was targeted. Secondly, the machine had to be small enough to be brought into the centre pathway (4 m wide) of the greenhouse in case the operator wanted to shred the waste right after removal. Thirdly, the shredder needed to be able to handle all three kinds of greenhouse crop wastes which were unique in their characteristics: leaves and fruit prunings (wet and soft), vines (long and woody), plastics twines (thin and rigid). Based on the above three criteria, two shredders were bought for the project and tested. The machines were tested for their flow rates, shredded product size, the handling by workers and the ability to cut twines. Mechanical modifications could be made, if necessary, to improve particle size distribution, flow rate, 29 sharpness, etc. The chosen shredder would be tested in a greenhouse at the end of the crop year to monitor the complete shredding process, including move-in, shredding and move-out. Time would be recorded for different wastes as reference. 30 III.4 RESULTS AND DISCUSSIONS III.4.1 SHREDDER TEST 1 - Bear Cat Full Size Chipper/hammermill (Model 70080) Figure III.l Hammermill Shredder - Bear Cat Mechanism : Hammermill - staggered pattern of free swinging, reversible, self-sharpening shredding knives. Horsepower: 8 hp Hopper Size : 22" x 35" (55 cm x 87.5 cm) Overall Size : 40" x 22" x 39.5" (100 cm x 55 cm x 99 cm) Figure III.l shows the "Bear Cat". This particular shredder would be too small for commercial operation, but would be good enough for testing the mechanism of hammermill shredding for greenhouse wastes application. At the end of a crop season, the Bear Cat was used to shred the year-end plants, mainly vines with the twines attached, for the lab scale and pilot scale composting. This machine could shred on average 10 plants 31 within 2 V2 minutes, provided that the flow was smooth and the machine did not need to be stopped as a result of plugging. In the beginning of the project, it was thought that the plant residues would be very easy to shred or chop. It was said that any machine could chop plant debris from a commercial greenhouse and the only factor we would have to consider was the horsepower of the machine. However, after this hammermill shredder was used, two major problems were realized with the greenhouse year-end waste: 1. High moisture content (over 80%); 2. The twines would spin with the shaft eventually forming a big wool-like mass which would plug the machine. The high moisture wastes after being shredded would become a soggy mass and could not be properly aerated in the composting process. The twines attached to the plants could wrap around the rotating shaft and eventually damage the bearings. In addition, the noise level was quite high because of hammering mechanism. Therefore, this kind of hammermill shredder was considered not suitable for processing greenhouse plant debris. III.4.2 SHREDDER TEST 2 - Teagle Tomahawk 100 Bale Shredder (shear shredder) Since twine presents a problem for most size reduction equipment such as hammer mills, other suitable shredders for greenhouse application were investigated. After searching a few manufacturers, the Teagle bale shredder was bought for testing (Figure III.2). 32 Figure 111.2 Teagle Tomahawk 100 Bale Shredder Mechanism : rotating disk within a drum, mounted with steel cutter sections Horsepower : PTO-driven /18 hp (separately built engine) Hopper Size (WxL): 32" x 20" (80 cm x 50 cm) Overall Size (WxLxH): 62" x 72" x 49" (155 cm x 180 cm x 122.5 cm) A PTO-driven straw bale shredder was purchased (Teagle Tomahawk 100) after the Bear Cat Chipper proved to be unsuitable. In the first trial, the speed was a bit slow as the wastes were wet and accumulated inside the drum. There was not enough of an air blowing effect to throw the shredded wastes out from the drum and the twines were not cut efficiently. Therefore, the following modifications were made at different times after trial and error for our application (see Figure III.3,4, 5): a. Extra blades were added to enhance the cutting mechanism; b. Baffles were added onto the rotating drum to increase the air blowing speed, thus enhancing the throughput rate; 33 c. As it was found that the wet shredded wastes always stuck on the inside wall of the shredder, a teflon sheet was used to line the inside of the drum to reduce the friction and thus avoid plugging by the chopped materials; d. As it was found out that the hopper and the shredding blades were to far apart and the opening was too small to push the leaves into the shredder, the hopper was shortened and widened in order to make feeding easier; e. An 18 hp propane-driven motor was added to the machine so that it would be more mobile for shredding inside the greenhouse, instead of being driven by the PTO of a tractor. Figure III.3 Modification of the Teagle Shredder - More Blades Added 34 Figure III.4 Modification of the Teagle Shredder -Baffles and Teflon Lining Added 35 It was found that this type of shredder mechanism (a rotating disk within a drum, mounted with steel cutter sections) successfully shredded the year-end waste (tomato and pepper) including twines. What makes this shredder different than other size reduction machines is that it operates at a low speed with high torque. Opposing rows of cutting rings with hook-like protrusions gradually pull materials in between the knives at constant speed rather than chopping them with high speed rotating knives. This difference accounts for lower noise and heat generation (Knights, 1995). ON-SITE WASTE HANDLING PROCESS Once the machine was tested successfully, it was then used in the real clean-up situation in the greenhouse at the end of the year. The machine was used in both tomato and pepper greenhouses. The whole shredding process, which is described in Fig. III.6, was done inside the greenhouse with minimal interference with the original clean-up procedure. The time, labour required and size reduction were recorded for both crops. The distribution of particle sizes was measured by simply sieving the shredded wastes through 3 screens with mesh sizes of 4.8 mm, 2.4 mm, and 1.0 mm. According to Haug (1993), the desired range of particles for composting should be between 0.5 mm to 10 mm. During the test, one phenomenon was observed: pepper wastes had a faster through-put rate than tomato wastes, but less compression (volume reduction) and smaller particle sizes after shredding. See Table III. 1. This was due to two reasons. First, the pepper vines were much shorter than tomato vines (3 m compared to 10 m), and thus much easier to handle and put through the hopper of the shredder. Second, the pepper vines were more woody (lower moisture content). For the pepper crop cleanup, approximately 55 m 3 of 36 waste was shredded resulting in 11 m 3 of shredded waste, in 4 hrs and 15 mins (approx. 10 plants in 30 seconds). For tomato crop cleanup, approximately 80 m 3 of wastes was shredded resulting in 13 m 3 of shredded wastes, in 7 hrs and 20 mins. The results showed that the year-end plants, including the twines and clips could be shredded into uniform pieces using the modified machine. A shredder like this would be able to handle the wastes for a 1 to 3-ha size greenhouse without delaying the cleanup process. However, there were also maintenance problems with the unit, e.g. broken blades, due to its light construction and small size; a heavier, more reliable unit would be required for commercial operation. A trial was also done to investigate whether separating the twines before shredding was possible. Workers were then asked to unwind the twines along the tomato vines before shredding. This was found to slow down the process by at least 3 to 4 times. Table III.l Comparison of shredding pepper and tomato year-end wastes Pepper Year-end Wastes Tomato Year-end Wastes Volume before shredding 55 m 3 80 m 3 Volume after shredding 11m3 13 m 3 Volume reduction 5 times 6.15 times Time required 4 hr 15 min 7 hr 20 min Shredding Rate (based on 12.9m3/hr 10.9 m3/hr intial volume) No. of Labour required 4 (2 cutting crop, 1 putting cut crop into the hopper, 1 5 (2 cutting crop, 2 putting cut crop into the hopper operating the shredder) since tomato vines are longer, 1 operating the shredder) Particle size: >4.8 mm 17.5% 23% 2.4-4.8 mm 35% 40% 7.5% 46% 1.0-2.4 mm <1.0mm 25% 6% 37 Figure III.6 Greenhouse Wastes Handling Process : 1. Cutting down the plants; 2. Pushing the plants into the hopper; 3. Shredded wastes collected into a bin; 4. Forklift taking full bin out of the greenhouse. 38 III.5 CONCLUSIONS Waste processing and handling are mainly about using space and time efficiently to produce good quality substrates for composting. The less space waste occupies; the more one can haul in one load or hold in a composting facility. The less time it takes to process waste; the more efficient the operation (Siegel, 1999). Generally, only the year-end waste requires size reduction, due to the long, tough plant stems. The desired range of particle size of the shredded material should be large enough to provide some physical structure, but small enough to maximize breakdown over the composting and curing period. As the desired range for composting should be between 0.5 mm to 10 mm, from the results, more than 75% of pepper crop and more than 71% of tomato crop shredded by the modified Teagle machine fell into this range. Contamination of year-end waste, primarily with plastic twines, presents a significant problem for both size reduction of the waste, and compost quality in terms of handling and appearance. There are several methods that can be used to deal with this. From a composting point of view, the ideal option is to source separate the organic plant waste from non-plant contaminants at the time of waste collection, for example removing twines from pepper and tomato vines during cleanup. However preliminary studies of this method showed a significant increase in cleanup time by 3 to 4 times and thus increased the labour costs. Shredding the conventional plastic twines along with the organic plant waste is another option which has been investigated. In this case the shredded plastics pass through the composting process unaltered, and should be later screened out (as much as 39 possible). Note however that although the shredded twine is highly visible after composting, the amount of plastic contamination is typically well below the 1% foreign matter content limit for Class A horticultural compost (B.C. Reg. 334/93). Utilization of bio-degradable twine products is another alternative and will be investigated in the next chapter. In this study it was found that a shredder with a rotating bladed drum would be more suitable for greenhouse applications. Not only because it could handle twines, but also it was economical (about $ 10,000 Cdn, including modifications) and mobile inside a greenhouse, whereas most of other commercial composting shredders were too big and too expensive. The ones used in Holland for greenhouse operation cost about $150,000 Cdn and questionnaires showed local B.C. growers could not afford them. It was also found out that the vines should be shredded as soon as they were taken out from the rows before they are piled up and get tangled with each other. Therefore, it may be more desirable to have the shredding inside the greenhouse when the plants are pulled from the sawdust bags. 40 III.6 REFERENCES Anonymous. (1994) New Equipment Take Composting Process to a Higher Level. Public Works. 124(5) : 46-47. Aquino, J.T. (1996) The Tale of the Tub Grinder. Waste Age. April : 115-120. B.C. Reg. 334/93. (2002) Organic Matter Recycling Regulation. Waste Management Act, s. 57. Christopher, D., Sid, H., Geoff, H. (1996) Commercial and residential organic pilot. BioCycle. 37(10): 51-53. Ettlinger, S. (1994) The Chipper/Shredder. Horticulture. V72. 52-55. Glass, R. (1997) Shredder Guide - Shear Shredders. Recycling Today. V35. 4-12. Glenn, J. (1999) Municipal composter expands as wood processor. BioCycle. 40(7) : 46-47. Gray, K. (1999) Tire chips bolster canal levee. BioCycle. 40(8) : 40-41. Haug, Roger T. (1993) The Practical Guide to Compost Engineerings Boca Raton: Lewis. Knights, M . (1995) A shredder is better. Plastics Technology. V41(10). 36-39. McPherson, T. H.C. (1998) Davis Sons Manufacturing Company Inc. Interview on June 11. 41 Newell, S. (1997) Shredder Guide - Medium Speed Grinders. Recycling Today. V35. 43-72. Savage, G.M., Diaz, L.F., Trezek, G.J., Goulueke, C.G., Wiles, C., Oberacker, D. (1981) On-site evaluation of municipal solid waste shredders. Resource Recovery and Conservation. 5(4) : 343-362. Siegel, H. (1999) Shredding Your Costs. Waste Age. 30(5). 258-263. Snell, J.R. (1991) Proper Grinding for Efficient Composting. BioCycle. 54-55. April. 42 CHAPTER IV UTILIZATION OF BIO-DEGRADABLE TWINES I V . l B A C K G R O U N D A N D L I T E R A T U R E R E S E A R C H IV.1.1 TRENDS IN BIODEGRADABLE PLASTICS Plastics have been used in many applications due to their durability, light weight and processability. However most plastics remain undegraded after discard, which pollutes the environment and disturbs the ecosystem. As plastic recycling is quite limited from an economic view point, biodegradable plastics could be a good substitute for plastics used for many industrial, commercial and agricultural usages (Jang, 2002). Initiated by increasing problems with plastics wastes during the last decade, new polymers have been developed, which can undergo a controlled biological degradation, i.e. composting (Amass, 1998). In most cases, the primary attack is an enzymatically catalyzed hydrolysis of ester, amide or urethane bonds in the polymers. However, in many cases, the term "biodegradation" is also used if the primary degradation step is caused by a hydrolysis which is not catalyzed by enzymes, but the depolymerization intermediates are finally metabolized by micro-organisms or resorbed by the body, in the case of medical applications (Muller, 2001). Since 1970s, the production and processing of polyhydroxybutyrate (PHB) as biodegradable plastics materials were developed. PHB is a natural aliphatic polyester and belongs to the group of polyhydroxyalkanoates, which are produced and intracellularly 43 accumulated by various micro-organisms. It was available on the market under the trade name "Biopol". Beside this natural polyester, a number of synthetic aliphatic polyesters have been shown to be also enzymatically hydrolyzable. The aliphatic polyester currently most important for commercial biodegradable plastics is poly-e-caprolactone (PCL), which is predominantly used as component in starch-blends. However, PCL exhibits a significant disadvantage - its low melting temperature of about 60 °C - excluding it from many applications (Muller, 2001). Polymers like PHB and PCL have good biodegradability, but lack structure and durability. Polymers like PET (ethylene terephthlate) and PBT (poly-butylene terephthalate) provide excellent material properties but are considered non-biodegradable. In order to combine both biodegradability and good material properties, copolyesters containing aliphatic and aromatic monomers were tested as biodegradable materials. By investigating the material properties, biodegradability and price, the combination of terephthalic acid, adipic acid and 1,4-butanediol (BTA) turned out to be the most appropriate combination (Witt, 1996). A number of materials, which are probably modifications of this basic BTA-structure are ready at the market fEcoflex' from BASF AG/Germany or "Eastar' from Eastman/US). Figure IV. 1 shows the chemical structure of PHB, PCL, and BTA-copolyester. 44 PHB —CH-CHj-C-O— — n PCL ? — C H J - C H I - C H J - C H J - C H J - C - O — n BTA - copolyester (random) 0 O O . — . o / /-(CHjj4-0-C-(CHa)«-C-0-/ m-(CH^--0-l~aJ)-C--O-/ — i n Figure IV. 1. Chemical Formula of PHB, PCL and aliphatic-aromatic copolyesters Bioploymers (e.g. starch, protein and cellulose) from various agricultural sources have also been investigated for manufacturing biodegradable plastics (Jane, 1994). Cross-linking of starch and zein mixtures and of soy proteins have shown that the tensile strength and water resistance of the plastics can be increased. Cereal flours are economically competitive materials for bio-plastics. Soy isolate and soy concentrate, with proper processing, showed positive properties for extrusion and injection molding of bio-plastics. Cellulose fibers can be used as extenders in the plastics to improve tensile strength and water resistance. These polyesters can be degraded by a variety of microorganisms, including some fungal phytopathogens. Many phytopathogens, for example Fusarium (a common pathogen in greenhouse crops) secrete cutinase, a serine hydrolase that degrades cutin, the structural polymer of plant cuticle (Murphy, 1998). The main constraint on the use of biodegradable polymers is the difference in the price of these polymers compared with that of bulk-produced, oil based plastics. The cost 45 of Biopol® was approximately £8,000 per tonne, compared with prices of commodity polymers of between £500 per tonne (PVC and PP) and £600 per tonne (HDPE and high impact PS) (Amass, 1998). The blending of biodegradable polymers is a method of reducing the overall cost of the material and offers a method of modifying both the properties and the degradation rates of the materials. These kind of materials were also tested in this study. The blends can be miscible or immiscible with each others. The advantages of producing miscible blends include single-phase morphology in the melt and reproducible mechanical properties. However, forming a miscible blend, particularly with a non-biodegradable polymer, can reduce or even inhibit the degradation of the biodegradable component. Immiscible blends have the disadvantage of having properties that are dependent on the blend morphology produced by processing and these are often not reproducible. However, some can show higher biodegradation rates than the unblended biodegradable homopolymer(s) (Amass, 1998). IV.1.2 USE OF BIODEGRADABLE PLASTICS IN HORTICULTURE Plants, in particular, tomatoes, peppers, and cucumbers, have soft stems. They rely heavily on the strength of twines, linking them to a support wire, to support their weight as they grow. For a tomato plant, the vines grow approximately 30 centimeters (12 inches) per week and can reach 12 meters (40 feet) in length by the end of the production season (BC Hothouse, 1999). Twines are also used to help greenhouse workers to move the plants along crop wires before the plants reach the ceiling of the 46 greenhouse. According to Hazelmere Greenhouses Ltd. (the surveyed greenhouse), for a 4 ha tomato greenhouse, it is estimated that approximately 2,000 kilograms of twines are needed per year. Twines also impose a burden on both the environment and the greenhouse growers. After the plants have grown around the twines throughout the season, they become entangled. Therefore, they are mixed together with the plants when the plants are taken down at the end of the production season. The large volume of organic plant residues could potentially be regenerated into a soil amendment by a bio-conversion process. However, the presence of the plastic twines in plant residues significantly reduces the value of the organic waste as the twine content interferes with the composting process during mixing making the waste undesirable for alternative solid waste disposals. Commercial composters generally do not accept organic wastes with plastic because producing and marketing compost with plastic - even small amounts - hurts the compost industry as a whole (Purman, 1998 ; Croteau, 1998). This is part of the reason why a large volume of greenhouse solid wastes ends up in the already overflowing landfills, as mentioned in Chapter II. Significant effort has been made to promote the use of alternative renewable resources to reduce the reliance on petroleum and its chemical derivatives. For example, in the United States, the Plant/Crop-based Renewable Resources 2020 is one such strategic vision supported by the U.S. Department of Energy, USDA and U.S. agricultural, chemical, and forestry communities (Faulkner, 1999). Reducing the use of 47 plastic twines in the greenhouse vegetable industry would contribute to the recognition of this vision. The use of biodegradable substitutes to replace the use of plastic twines could offer additional waste management options for the greenhouse industry. It would not only reduce the generation of plastic waste but also greatly increase the commercial value of the organic wastes as the absence of impurities and consistent quality of the organic wastes would make them more attractive to commercial composters (Purman, 1998). This could permanently remove a large fraction of the municipal solid waste stream from landfills (Khan, 1996). A process with lots of waste, as long as it's "wanted waste", may be better than one with a small amount of waste that must be landfilled or burned (Benyus, 1997). Furthermore, it could avoid the complications induced by separating twines from plants. In short, it was believed that the problems associated with the plastic twines could be minimized or even eliminated by using biodegradable twines that would possess similar, desirable physical qualities as traditional plastic twines. 48 IV.2 SPECIFIC OBJECTIVES • To identify an environmentally compatible substitute(s) that could replace the current use of petrochemically-derived plastic twines in greenhouse vegetable production; • To test the substitutes' durability under the greenhouse growing environment; • To test the substitutes' compostability in an in-vessel composter. IV.3 METHODOLOGY The purpose of this study is to investigate the use of alternative bio-degradable twines for the greenhouse vegetable industry in order to replace the current non-biodegradable plastic twines. There were four stages in this study : First stage : Select twine candidates for the study, which involved literature and industry-based research to understand the properties of twines, their physical conditions as well as the market needs. Second stage : Utilize the selected twines as support for plants in a commercial greenhouse located in Surrey, BC. Third stage : Measure the changes in physical properties of the twines collected from the greenhouse in a laboratory during the growing season. 49 Fourth stage : Conduct a compostability study on the alternative twines in a pilot scale composter. IV.3.1 S E L E C T I O N O F T W I N E C A N D I D A T E S The first stage of the experiment was to select twine candidates for the experiment. They had to meet the following criteria: 1. Availability They must be commercially available in large quantities to meet the market demand. Twines are usually demanded in large quantities in the beginning of the year when a new production season begins. 2. Tensile Strength It must have sufficient physical strength to support the plants during the production period under typical greenhouse conditions. For greenhouse crops, which generally have a production period of 11 months or less, the twines must not only sustain the weight of the plants throughout this period but also withstand the physical stress imposed when twines are being move along the crop wire. 3. Handling Twines that tend to easily absorb and retain moisture would be more likely to host plant pathogens that are prone to causing a disease outbreak. Also, materials that 5 0 have a rough texture might potentially injure greenhouse workers as they could cause scratches or cuts. 4 . Cost They must be relatively economical compared to traditional plastic twines. This practical aspect of the product must enable the twines to compete with other similar products available in the market in order to attract the greenhouse growers. 5 . Biodegradability They must be readily biodegradable. "Readily biodegradable", in this case, is defined as the breakdown of the material into inorganic components within a reasonable time frame, between 4 to 8 weeks of active composting and 4 to 8 weeks of curing. This range of time frames was chosen based on the operation schemes commonly observed in commercial composting facilities. 6. Colour It is desired that the twine be white or transparent in colour to maximize the reflection of sunlight inside the greenhouse. In a greenhouse environment, sunlight plays a crucial role in both energy consumption and crop yield. Therefore, any amount of reduction in sunlight would be undesirable. A lot of materials inside a greenhouse, for example plastic ground cover, bags, lamp covers, glutters, trusses, are specially made in white. A brief outline of the six criteria is provided in Table IV. 1. The twines selected for this study included Vertomil®, Ecolan®, Cargill EcoPLA (EcoPLA® 2000D), cotton, 51 and jute. Vertomil® and Ecolan® were supplied by Lankhorst Touwhfabrieken bv. based in Holland. EcoPLA®2000D was supplied by Cargill Plastic Company based in the United States. Cotton and jute were made available by Westgro Co. Ltd., a local agricultural product supplier. Table IV.2 summaries the compositions of the selected twines. The following section provides further information on the five types of twines and the reasons why they were chosen in this study. Table IV. 1 Criteria for Twines Criteria 1. Readily available 2. Sufficient strength throughout a production period 3. Pose no or very little harm to both plants and humans 4. Relatively economical 5. Readily biodegradable 6. Light in colour Cotton Cotton is a naturally occurring fiber made of seed-hair fibers obtained from Gossypium sp. At present, the chief cotton-growing countries of the world are China (23%), the United States (17%), the former USSR (15%), India (11%), Pakistan (8%), Brazil (4%), Turkey (3%), and Egypt (2%). It is one of the three most important fibers in the textile industry. Chemically, cotton is the purest, containing over 90%> of cellulose with little or no lignin. Established techniques have been developed to measure soil microbial activity by cotton strip assay (CSA). The technique involves burying cotton strips and measuring their tensile strength after a certain time. This gives a measure of the rotting rate, R, of the cotton strips. R is then a measure of soil microbial activity (Corella, 1997). 52 Jute Jute is a naturally occurring fiber made from two herbaceous annual plants, Corchorus capsularis (linden family, Tiliaceae) originating from Asia, and C. olitorius originating from Africa. It contains high contents of lignin, a natural phenolic polymer frequently present in cell walls of fibers, which contributes to their stiffness. Compared to cotton, jute has a higher tensile strength and modulus of elasticity but a lower extensibility (elongation). Among all fibers, jute has the highest moisture regain of 14%. The colour ranges from white to reddish brown, but usually has a golden luster. It has traditionally been one of the principal bast fibers (tonnage basis) sold on the world market; however, the precipitous decline in jute exports by India indicates a decreasing market demand for this fiber that is vitally important to the economies of India (West Bengal), Bangladesh, and Pakistan. Nowadays, in addition to India, jute is grown Bangladesh, Thailand, Nepal and Brazil (Young, 1994). EcoPLA® 2000D EcoPLA®2000D is a biodegradable plastic made from polyesters, namely, EcoPLA resins. EcoPLA resins are thermoplastic starch with starch/polylactic acid composites. This thermoplastic starch is used as a strengthening agent. It also increases the surface area that can be directly attacked by microorganisms. These products are much more affected by moisture than the starch/polyethylene/catalyst composites. The primary bonds in these biodegradable synthetics are readily broken apart by hydrolysis. Heat and UV light will also cause deterioration, which can be a problem for greenhouse usage. Two 53 major PLA resin producers are Cargill Plastics Ltd. and Dow Chemicals Ltd., based in the U.S.A. Since polylactic acids (PLA) are made from starches, such as potato wastes, the resources required are renewable and sustainable. Since the raw materials are biologically based, they undergo complete degradation usually within several weeks. Furthermore, the by-products produced during the composting process would likely be benign compared to nonhydrolyzable polymers, such as, polyethylene (Karlsson, 1995). Cotton/Viscose & Jute/Viscose These two types of twines are made of both natural fibers and a regenerated cellulosic fiber, more commonly known as rayon. The terms, viscose and rayon, are used somewhat interchangably. The regenerated cellulosic fibers are formed when a natural polymer, or its chemical derivative, is dissolved and extruded as a continuous filament, and the chemical nature of the natural polymer is either retained or regenerated after the fiber formation process. The details of production process can be found in various literature (BeMiller, 1992 and Woodings, 1994). The main raw material required for the production of viscose is cellulose, a natural polymer of D-glucose. An estimated 10 billion tonnes is produced annually by natural processes. In general, regenerated cellulosics are highly hydrophilic with a moisture regain from 11% for the polynosics to 13% for regular rayon at 65% relative humidity. The fiber structure swells as fluid is imbibed, and fiber strength and stiffness fall. The fibers degrade hydrolytically when contacted with hot dilute or cold concentrated mineral acids. Regarding environmental issues, cellulosic fibers are the only mass-produced, man-made fibers made directly from 54 a natural polymer (cellulose), unlike polyesters, nylons, polyolefins, and acrylics, which come from nonrenewable reserves of fossil fuels. These natural polymers can undergo complete biodegradation or incineration resulting in final breakdown products of carbon dioxide and water (Woodings, 1994). Table IV.2 Selected Twine Candidates and Their Compositions Names / Trade names ( (imposition Remarks Jute Jute Natural fiber Vertomil® Viscose / Jute Regenerated cellulosic fiber and natural fiber Ecolan® Viscose / Cotton Regenerated cellulosic fiber and natural fiber E c o P L A ® 2000D Biodegradable plastic Made from polylactic acids; Underwent biodegradability studies. Cotton Cotton Natural fiber IV.3.2 UTILIZATION OF SELECTED TWINES After the five twine candidates were selected, they were tested in a commercial greenhouse located in Surrey, B.C. from January 1998 to November 1998 (one full growing season). A total of 250 sample twines, with 50 replications for each material, were individually tied to pepper plants as supports. Figure IV.2 shows jute being used in the greenhouse. The twines were exposed to the typical greenhouse conditions, such as high humidity, high temperature and sunlight exposure, throughout the production period between January and November. Table IV.3 represents the range of climatic conditions found in the greenhouse. 55 Table IV.3 Typical Greenhouse Conditions Daytime Temperature 20-24 °C Night Temperature 17-18 °C Humidity 60-80% Plant Density 250 - 280 plants per 100m2 Crop Season Approximately 11 months These environmental conditions are believed to play a crucial role in the physical strength of the twines since the initiation of material degradation generally proceeds via thermal, hydrolytic, photo, oxidative, and/or mechanochemical chemical pathways (Griffin, 1994). Increased exposure to sunlight, high temperatures, and high humidity could trigger the onset of degradation prematurely and result in an accelerated reduction in the physical strength of twines that are in use. On the other hand, these conditions tend to favour biodegradation of the twines since biodegradation is usually the secondary process following one or more of the chemical reactions. Figure IV.2 Jute Twine for Plant Support 56 During the field test, the 250 twines were treated as if they were part of the commercial production. The plants were pruned, and mature peppers were harvested. The twines were manually maneuvered along the main lines as the plants grew. Periodically, they were examined, and the number of nearly broken twines was recorded. They were then replaced. In most cases, the twines were replaced before breaking to avoid damaging the plants. In addition, a sample of each type of the twines was collected for laboratory measurement at the end of each month until the pepper production ceased. IV.3.3 MEASUREMENT OF CHANGES IN PHYSICAL PROPERTIES OF TWINES Laboratory measurements were conducted in the Pulp and Paper Research Building on the UBC campus. A total of 11 sets of twine samples, each set containing all five types of the twines, were collected from the commercial greenhouse for laboratory testing from January to November. Continuous monitoring of the twines began in January 1998 and observations were recorded. The laboratory measurements were designed to measure two parameters of the twines: linear density and tensile strength. These properties were believed to provide sufficient evidence in determining the physical properties and durability of the twines. When the twines arrived in the laboratory, they were allowed at least two days to acclimatize to the laboratory conditions - a controlled environment having a relative humidity of 51% and a temperature of 21 °C - to minimize effects induced by variations in moisture content and temperature of the twines. 57 Linear Density The size of the twines was first determined in order to put a quantitative measurement on their thickness. Linear density of the twines was measured for it is a common practice in the fiber industry to designate the thickness of fibers. Linear density has a unit of mass per unit length. This particular method of measurement was chosen because it would be difficult to determine the cross-sectional area by other methods, such as direct microscopic observation and measurement. Direct observation and measurement of a cross section under a microscope is considered the most accurate method. However, it is a destructive test that does not allow subsequent study of other mechanical properties. It is also slow and tedious. In addition, it does not take into account any variations in the cross-sectional area along the fiber length. Measurement of fiber diameters from microscope observation of longitudinal views is slightly easier, but the ellipticity of the cross section in certain fibers could lead to serious errors (Akhtar, 1988). For the determination of linear density, each month five twine samples were sectioned into lengths of 25cm. The five twine samples were weighed, and the mean and standard deviation were recorded. The linear density of the twines was calculated in the unit of grams per centimeter. Tensile Strength The tensile strength and elongation of the five candidates were determined using an electronic tensile tester Model QC II from Thwing-Albert Instrument Company shown in Figure rv.3. The tensile strength test was a destructive test where the material was not recoverable after the test. Table IV.4 summarizes the settings of the electronic tensile 58 tester. The setting "peak" indicated that the twine was subjected to a load until the tensile tester sensed its maximum strength. This load-and-release cycle was repeated until the breakage of the twine sample occurred. Each time the load was exerted on the twine, readings of tensile strength and elongation were recorded. The American Standardization and Material Testing (ASTM D2433-93,1998) was used as a reference for the procedures of the test. Samples obtained over a period of eleven months from January to November were tested. Twines with a length of 25 centimeters were used in the measurement. The results were recorded in British units, pounds and one thousand of an inch, respectively. These units were selected because the industry still utilizes the British system. However, the analysis of the data was performed in SI units. Figure IV.3 Electronic Tensile Tester Used in Laboratory Measurements 59 Table IV.4 Settings of Tensile Tester Variables Setting Unit of tensile strength British (pounds) Sensitivity Moderate Speed B Testing mode Peak Unit of elongation British (fraction of an inch) IV.3.4 COMPOSTABILITY A compostability test was conducted in the middle of November 1998 after the production period was completed and plant residues were collected. Data collection began on December 1, 1998 and carried through till April, 1999. The three materials that believed to be bio-degradable were selected for this test, i.e. cotton, jute and EcoPLA®2000D. Determination of biodegradability had been a controversial topic until ASTM's Committee D20 issued a new standard - D6400, Standard Specification for Compostable Plastics in May 1999. This marked the culmination of eight years of research and testing by leading producers and users of bio-degradable plastics. Prior to May 1999, no scientifically based standard existed for determining the biodegradability and compostability of plastic products. Before this point, manufacturers were free to use any test method they felt most appropriate (Mojo, 2001). To measure biodegradability, most producers used tests such as those found in ASTM D822, which had historically been used to assess the physical properties of plastics films. The compostability test in this study was designed by the author based on this information since at the time of the study the ASTM D6400 had not been published yet. 60 Samples of these three candidates were composted in a pilot scale composter described in Chapter V.3.2. The main ingredients of the compost included the plant residues collected from the greenhouse, and alder bark chips. Sections of new twines with a length of approximately 76 cm were individually enclosed in a mesh bag with a mesh size of 1.5 mm x 1.5 mm. They were placed in the composter containing approximately 10 tons of organic waste, an estimated 8 to 10 cubic meters in volume. In order to achieve maximal homogeneity, periodical mixing and turning was done. For ease of recovery, the twines were chained to a fishing line whose end was tied to the side of the composter. Composted twine samples were collected periodically. A total of eight twine samples were recovered for each type of the twines. According to ASTM, a synthetic material is biodegradable when it retains no more than 10% of its original weight after 12 weeks in a highly controlled compost environment at optimal conditions of aerobic/anaerobic activity, nitrogen levels, etc (McDonald, 1998). Therefore, a reduction in weight was used to determine biodegradability. The composted twines were removed from the mesh and then washed in distilled water. They were let dry for at least six days in the laboratory before weighing. To obtain the weight of some of the very fragmented twines, the mesh and the twine were weighed together. The mesh was then weighed again after the content was removed completely. Wherever possible, the tensile strength of the composted twines was measured to reflect the deterioration in strength. 61 IV.3.5 STATISTICAL ANALYSIS For the linear density and tensile strength tests, 5 samples from each of the five twine candidates were selected randomly in the test area of the greenhouse each month, and taken out and replaced by conventional plastic twines with a different color to avoid confusion in next month's test. Means and standard deviations were calculated and presented as error bars in the graphs. For the compostability test, 3 samples from each of EcoPLA, Cotton and Jute were dug out from the compost mass each time for linear density and tensile strenght tests. Tests were done more frequently, i.e. about once a week, in the active composting period and were slowed down to once a month during the curing period. 62 IV.4 RESULTS AND DISCUSSIONS This section focuses on the presentation and discussion of the findings. The first section accounts for the observations made during the use of the twines in the commercial greenhouse. The second section describes the physical testing of twines in the laboratory. It also contains data analyses and discussion of the laboratory measurements, which are presented both graphically and in tables. The last section embodies the results from the compostability study. IV.4.1 ON-SITE OBSERVATIONS AND SURVEYS During the. application of the twines in the commercial greenhouse, it was observed that 49 out of all 250 biodegradable twines broke or showed signs of breakage. Among those 49 twines, cotton and cotton/viscose represented the majority of them, 14 and 18 twines, respectively, translated into 29% and 37%, respectively. Twines that were broken or showing sign of breakage twines were replaced to avoid damage of the crop. Table IV.5 listed the time in service of each twine. 63 Table IV.5 Time in Service of each type of twine (50 samples each) Type of twine Bre: kage by time (non-cumulative) 2 months 4 months 6 months 8 months 10 months Total Jute 0 1 2 3 2 8 (16%) Cotton 1 3 3 6 1 14 (28 %) Jute/Viscose 0 1 2 3 3 9 (18%) Cotton/Viscose 1 3 6 5 3 . 18 (36%) EcoPLA 0 0 0 0 0 0 (0 %) Plastics (control) 0 0 0 0 0 0 (0 %) The details of the results and observations are as follows : Jute. - A few loosened jute twines were observed, although none of them were completely broken. Greenhouse workers did not particularly like jute because it was rough and difficult to tie and cut. It was noted that jute became soft and slowly deteriorated when it remained wetted for a long time. Cotton - A few loosened cotton twines were observed and some of them were partially broken and therefore were replaced before the plants fell down. Softness of cotton appealed to the greenhouse workers. It was also easy to cut during the year-end removal. However, due to its softness, it was difficult to tie. It had a tendency to absorb moisture and could be potential trap of fungi and microorganisms. 64 Jute/Viscose - For all the broken jute/viscose twines, the broken parts were from the jute portion. Since it is composed of two separate thin twines winding around each other, the jute component was found to be more vulnerable. It appeared that most of the strength was sustained by the viscose component. The workers preferred jute/viscose to cotton and jute because it was thinner and easier to handle in comparison. Cotton/Viscose - For all the broken cotton/viscose twines, the broken parts were from the cotton portion. Since it is composed of two separate thin twines winding around each other, the cotton component was found to be more vulnerable. It appeared that the strength of the twine lay in the viscose component. The greenhouse workers preferred cotton/viscose because it was thinner and easier to tie and cut compared to cotton and jute alone. EcoPLA - None of the 50 EcoPLA twines showed signs of breakage. It seemed, in this regard, that it was the only one comparable to the traditional plastic twines. They were strong and did not show any sign of deterioration at the end of the season. However, the workers found that the EcoPLA twines were hard to tie and cut since its texture was rather tough and inflexible, and was quite slippery. Based on the above results, the biodegradable plastic twine EcoPLA®2000D was found to be most durable among the five biodegradable candidates in the eleven months field trial. 65 IV.4.2 LINEAR DENSITY Linear density is one way of representing the thickness (weight/length) of the twines. Figure IV.4 depicts the linear density of the five types of twines over the period of time between January and November. Cotton, jute/viscose and cotton viscose retained their original linear densities throughout the test period. Jute, which had the highest linear density, appeared to experience an increase in linear density over the test period. EcoPLA®2000 showed the greatest decrease in linear density. This occurred in the spring and summer. In those months, the temperature inside the greenhouse in the daytime was usually higher than the desired set point temperature because greenhouses in B.C. usually have heating capability but no cooling capability. Therefore, this phenomenon could be caused by the accelerated degradation in those months, which had higher temperature, humidity and sunlight levels (particularly UV). The change in linear density was characterized by the negative slopes in Figure IV.4. The linear density of jute, on the other hand, varied over time primarily due to the inconsistency in thickness. The result indicated a gain in linear density for jute, demonstrated by a positive slope in linear density. This might also be due to the absorption of moisture and growth of microorganisms on the jute. During the laboratory testing, several significant observations were made. It was observed that there was a considerable variation in the linear density of jute from each batch of testing (see the error bars from Figure IV.4). The variation was likely induced 66 during the manufacturing process. Also, the texture of jute tended to be the roughest among all five candidates. Linear Density Profile of Alternative Twines used in Pepper Production 0.0350 0.0300 § 0.0250 jfr 0.0200 N c ,§ 0.0150 2 0.0100 0.0050 0.0000 -Jute -EcoPLA Cotton Jute/Viscose - Cotton/Viscose ^ — * — — * — x — * g -^rr jL___^ - J L ^ - >*- T X * " — ^ . — * * ^ k w • * 1 ~* * — — • » *• - » _ Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Time Figure IV.4 Linear Density of Alternative Twines 67 IV.4.3 TENSILE STRENGTH TEST Tensile strengths of all five twine candidates over a time period between January and November are graphically presented in Figure IV.5. Jute, cotton and cotton/viscose tended to retain their original tensile strengths throughout the test period. Jute was stronger than cotton which, in turn, was stronger than cotton/viscose. EcoPLA had a high Tensile Strength Profiles of Alternative Twines used in Pepper Production Jan Feb Mar Apr May Jun Time Jul Aug Sep Oct Nov Figure IV.5 Tensile Strength of Alternative Twines initial tensile strength which was retained for about four months, after that its tensile strength began to decline becoming lower than that of jute and about the same as cotton. This is at least partially attributable to the observed decline in linear density which meant that there was less substance to resist the load in the tensile tester. Jute/viscose retained its original tensile strength for about four months then becoming weaker. At the end of the test period it had the lowest strength. 68 Change in tensile strength with time was characterized by the slope of a plot of tensile strength vs time. It was anticipated that the strength of the twines would decline over time due to physical degradation and increasing loading as the plants grew. EcoPLA, jute/viscose showed declining tensile strengths. Especially EcoPLA came up with a relatively declining slope. Jute showed fluctuation throughout the study. This might be due to the uneven manufacturing and its moisture absorbing properties. Although some of the twines showed declining tensile strengths, especially the EcoPLA, none of the twines showed breakage under 80 lb, i.e. they would all be good enough to sustain the weight of the plant (20 to 40 lb) for the whole growing season under the greenhouse climatic conditions. Compared to others, jute was able to withstand the most load-and-release cycles before it broke. With cotton/viscose and jute/viscose, the cotton and the jute components often broke first. The viscose component sustained the load solely after three to four cycles of load-and-release. The texture of EcoPLA®2000D tended to be more rigid compared to its counterparts. 69 IV.4.4 COMPOSTABILITY STUDY In this bio-degradability study, only EcoPLA®2000D, cotton and jute were used. Cotton/viscose and jute/viscose were not used, because both had a twisted combination of two twines and viscose was not readily degradable in an ordinary composting environment. The purpose of this study was to find out the biodegradability properties of cotton, jute and EcoPLA. Figure IV.6 shows how the twine samples were placed inside a screen-net which was put inside the composter. The "net" and its contents got mixed and composted together with the other substrates. The "nets" were taken out from time to time to examine the biodegradability. Figure IV.7 shows an example of a broken-down EcoPLA twine after a few days of composting. It was taken out from the net, all the dirt attached was washed off, and then it was dried before weighing. The composting results showed that, in general, there was a significant decline in the linear densities of the three biodegradable twine materials - EcoPLA®2000D, cotton and jute (see Figure IV.8). EcoPLA, jute and cotton had an 80.5%, 85.6% and 89.4% weight loss after 136 days of composting, respectively. However, since all the twines were very "dirty" when they were taken out, they needed to be washed carefully in order to take out any foreign materials. In this process some weight loss would be expected since there were always some small fragments of the broken twines being washed out as well, especially when the twines were degraded into very small pieces at the end of the curing process. This meant that there were some unavoidable errors in this experiment. 70 Figure IV.8 shows the loss in linear density over time in the composter. The small linear density gains for jute on day 28 and cotton on day 21 might be explained by the lack of uniformity in the composting environment. These twines were most probably located at the outer edge of the compost where composting was less active. Nevertheless the overall result from the compostability study of the twines suggested that the materials could indeed degrade readily in a composting environment within a reasonable time frame - they all began to break into fragments in the active composting process where the temperature was maintained at around 55-65 °C. Al l of the twines experienced most of their loss in linear density in about 100 days. Jute degraded faster than cotton which lost linear density more rapidly than EcoPLA. Figure IV.9 shows the effects of composting pn tensile strength. The arrows indicate the occurrence of fragmentation (breakage) of the any twine sample at that time i.e. they broke apart with tension. Jute first fragmented, then cotton, and finally EcoPLA. Even though they showed breakage at different times, their breakages were all before the end of the active composting period. Tensile strength of all of the twines was virtually destroyed in 30 days of exposure to composting. EcoPLA was more resistant than cotton which in turn was more resistant than jute. Figure IV.6 The bio-degradable twines (EcoPLA®2000D) in a screen-net before being placed into the composter Figure IV.7 Example of fragmented twines (EcoPLA®2000D) after being placed into a composter 72 Figure IV.9 Tensile Strength of the bio-degradable twines 73 IV.5 CONCLUSIONS A decision matrix (see Table IV.6) was constructed to evaluate the five twine candidates based on the criteria listed in Table IV. 1. In the evaluation matrix, a scale of 1 to 5 was used - 1 being the least satisfactory and 5 being the most satisfactory among the five candidates. For the purpose of comparison, traditional plastic twine was included in the evaluation matrix. Cost was based on the estimated price per unit length (100 m) of the products available in the fiber markets, However, since EcoPLA®2000D, Vertomil® and Ecolan® were still in their experimental stages, the price may be cheaper in the future when more greenhouses try to use them. It was also noted that the disposal cost for the conventional plastics was not accounted for since it might vary between different disposal methods. Should this cost be included, the mark for plastics would be lower. Tensile strength of the candidates was evaluated based on the information obtained from both the field study and the laboratory measurement of tensile strength. The assessment of handling was based on interviews with greenhouse workers, while biodegradability was evaluated based on the physical properties of the materials and the compostability study. Lastly, a white colour was preferable due to more sunlight reflection in the greenhouse. The resulting evaluation matrix concluded that EcoPLA®2000D had the highest score among the five potential substitutes of the traditional plastic twines. In short, this study found that EcoPLA®2000D was the most satisfying alternative twine. Nevertheless it was not as good as the plastic twine currently in widespread use. Upon completion, the evaluation 74 representatives from the greenhouse industry as well as academic professionals to ensure that scores were properly assigned. Table IV.6 Evaluation Matrix Factors/ Candidates Jute Cotton E c o P L A ® 2000 Vertomil® Eeolan® Plastic Cost (1.0) 3 2 1 3 3 5 Tensile Strength (1.0) 4 2 5 3 3 5 Handling (0.8) 3 3 5 4 4 5 Biodegradability (0.8) 5 5 5 3 3 1 Colour (0.7) 2 5 5 3 5 5 Full Score 21.5 21.5 21.5 21.5 21.5 21.5 Score 14.8 13.9 16.7 13.7 15.1 18.3 Percent Satisfaction (%) 69 65 78 64 70 85 From the results obtained from the experimental in- greenhouse tests, laboratory measurements, and the compostability study, it was concluded that it was feasible to utilize alternative twines in greenhouse vegetable production. Furthermore, among the five types of twines, EcoPLA®2000D was found to be the most suitable substitute for the traditional plastic twines currently used in greenhouses. 75 It was hoped that the conclusion drawn from this study would encourage the use of biodegradable twines in the greenhouse vegetable production. The replacement of plastic twines with biodegradable alternatives could drastically increase the potential values of the organic waste generated from the greenhouse vegetable industry. It not only promotes the use of alternative renewable resources, but also makes the compost, generated from the greenhouse organic wastes, freer of impurities and more consistent in quality. 76 IV.6 REFERENCES Akhtar, A. (1998) Localized Intrinsic Strengthening Approach (LISA): A Practical Method for Determining the Tensile Strength of Multistrand Cable. Journal of Testing and Eval uation 16:124-133. Amass, W., Amass, A. and Tighe, B. (1998) A review of biodegradable polymers : use, current developments in synthesis and characterization of biodegradable polyesters, blends of biodegradable polymers and recent advances in biodegradation studies. Polymer International. V47. 89-144. ASTM (American Standards for Testing and Materials). (1998) D2433-93. Standard Test Methods for Rubber Thread. BC Hothouse. (1999) www.bchothouse.corn/tomdex.htm. April 6. BeMiller, J. N . (1992) Carbohydrates. Encyclopedia of Chemical Technology. 4 t h edition, Vol. 4, New York: John Wiley & Sons. 911 -948. Benyus, J. M . (1997) Biomimicry. New York: William Morrow and Company. Corrella, R.L., Harcha, B.D., Kirkbyb, C.A., O'Brienc, K., Pankhurstb, C E . (1997) Statistical analysis of reduction in tensile strength of cotton strips as a measure of soil microbial activity. Journal of Microbiological Methods. 31(1-2): 9-17. Croteau,G. (1998) Assessing the Degradability of Polymeric Materials. BioCycle. 39(3). 71-75. Faulkner, D.L., McLaren, J.S. and Mustell, B. (1999) Renewable Resources 2020. Resource 6:11-12. 77 Griffin, G.J.L. (1994) Chemistry and Technology of Biodegradable Polymers. London: Blackie Academic & Professional. Jane, J. Lim, S., Paetau, I., Spence, K., Wang, S. (1994) Biodegradable plastics made from agricultural biopolymers. Polymers from agricultural coproducts. American Chemical Society Meeting. 92-100. Jang, J.C., Shin, P.K., Yoon, J.S. (2002) Glucose effect on the biodegradation of plastics by compost from food garbage. Polymer Degradation and Stability. 76(1). 155-159. Karlsson, S., Albertsson, A.C. (1995) Degradable Polymers, Recycling, and Plastics Waste Management. New York: Marcel Dekker. Khan, M.R. (1996) Conversion and Utilization of Waste Materials. Washington: Taylor & Francis. McDonald, P. (1998) Industry Report, Biodegradable Twine Report. June. Mojo, S. (2001) What's Next for Biodegradable and Compostable Plastics. ASTM Standardization News. April. Muller, R.J., Kleeberg, I., Decker, W.D. (2001) Biodegradation of polyesters containing aromatic constituents. Journal of Biotechnology. 86(2) 87-95. Murphy, C.A., Cameron, J.A., Huang, S.J., Vinopal, R.T. (1998) A second ploycaprolactone depolymerase from Fusarium, a lipase distinct from cutinase. Applied Microbiological biotechnology. V50(6). 692-696. Noelie, B., Cotton, R. (1993) Encyclopedia of Chemical Technology. 4 t h edition, Vol. 7, New York: John Wiley & Sons. 620-647. 78 Purman, J. (1998) Plastics and Compost End Use. BioCycle, 39(3). 62-63. Witt, U. , Muller, R.J. and Decker, W.D. (1996) Evaluation of the biodegradability of copolyesters containing aromatic compounds by investigations of model oligomers. Journal of Environment Polymer Degradation. V.4. 9-20. Woodings, C.R. (1994) Fibers (Regenerated Cellulosics). Encyclopedia of Chemical Technology. 4 t h edition, Vol. 10, New York: John Wiley & Sons. 696-726. Young, R. A. (1994) Fibers (Vegetable). Encyclopedia of Chemical Technology. 4 t h edition, Vol. 10, New York: John Wiley & Sons. 727-744. Special Acknowledgement: This part of the project is the joint effort of the author and Ms. Coby Wong, an undergraduate student who worked under the supervision of the author. This project could not have been successful without the cooperative work of Ms. Coby Wong, and the workers in Hazelmere Greenhouses Ltd. 79 CHAPTER V BIO-CONVERSION PROCESS V.l BACKGROUND AND LITERATURE RESEARCH V.l . l COMPOSTING METHODS Composting can be done using one of three following methods: aerobic, anaerobic, or vermiculture. They are briefly reviewed, separately in the following sections. a) Aerobic Composting Aerobic decomposition is a biological, exothermic, oxidation of organic matter, carried out by a dynamic and rapid succession of microbial populations. The organic matter is ultimately transformed into a final stable humus type product (compost) through its mineralization and humification, by which approximately 20 to 30% of the volatile solids are converted into CO2 and H2O (Hoyos, 2002). Thermophilic temperatures between 55°C and 70°C are commonly achieved, which results in pathogen kill and weed seed inactivation. Due to the oxidation that occurs, odor creation is limited compared to anaerobic systems. b) Anaerobic Composting Anaerobic decomposition occurs in the absence of oxygen, and temperatures of less than 55°C are achieved. The process yields only partially reduced and oxidized compounds that may continue to breakdown after treatment. During anaerobic digestion, 80 organic compounds can be transformed to an extent of almost 80% into biogas, for example methane gas. The anaerobic degradation of organic matter is a multi-phase process comprising acidogenesis and subsequent methanogenesis (Held, 2002). The main disadvantage of anaerobic decomposition is that the by-products of the process (such as fatty acids, aldehydes, hydrogen sulfide and ammonia) often create offensive septic odors, c) Vermicomposting Vermicomposting is a variation of aerobic composting in which earthworms are added to promote rapid decomposition of organic materials. During vermicomposting, earthworms eat, grind, and digest organic wastes with the help of aerobic and some anaerobic microflora, converting it into a much finer, humified, microbially active material (Maboeta, 2003). In addition to feeding on the microorganisms, the earthworms fragment particle agglomerations which results in a more homogeneous texture. After a period of 10 days or more, the earthworms are removed from the compost, the material is draught dried, filtered through a sieve, and then packaged and marketed. Modern compost systems are mostly aerobic. One of the more important reasons for choosing the aerobic process is that aerobic processes are not characterized by objectionable odors. Some promoters of aerobic composting are prone to characterize it as being odorless, but by the very nature of the raw material and of the intermediates formed, some odor production is bound to occur. Even the claim of producing no objectionable odors may be inaccurate. A technology of Combined anaerobic/aerobic Composting Process (CCP) has been considered as an attractive alternative for aerobic 81 biowaste composting because of its 17 times lower overall emission of volatile compounds compared to single phase aerobic composting (Smet, 1999). A second and perhaps more important reason for aerobic composting pertains to public health and crop production. Public health and crop safety come from the high temperatures that are the natural concomitants of a properly conducted aerobic compost operation. The temperature in an aerobic pile usually reaches levels above the thermal death point of most plant and animal pathogens and parasites. These elevated temperatures also are lethal for weed seeds. A third reason for aerobic composting is that it is more rapid than anaerobic fermentation. In recent years great progress has been made in accelerating the anaerobic fermentation process through careful design of equipment and operational procedures. However, many organisms that occur in composting and that can rapidly break down refractory compounds are obligate aerobes and therefore obviously could not survive in an anaerobic environment. Moreover, all anaerobic processes need an aerobic stage eventually to stabilize (i.e., oxidize) the compost. During waste treatment, mass losses of carbon were highest in the aerobic treatment and lowest during anaerobic treatment. Following their application to soil, the amount of CO2 -C evolved from wastes was highest from anaerobically-treated material, intermediate from non-decomposed material and lowest from aerobically-treated material. The effect on carbon stabilization efficiencies of various waste treatment was ranked as follows : aerobically-treated and composted > non-decomposed > anaerobically-treated (Kirchmann, 1997). For reclaiming 82 degraded soils, aerobically composted wastes are better than anaerobically degraded wastes. Naturally, aerobic composting is not without its drawbacks. For example, the maintenance of aerobic conditions involves more handling and greater spatial requirements than would be the case with anaerobic composting. Another downside is the inevitable loss of at least some nitrogen in aerobic composting. The loss is the accompaniment of the high temperature and eventual alkaline conditions reached in an aerobic pile. These two conditions promote the volatilization of ammonia during composting. Gaseous nitrogen losses during composting occur mainly as ammonia, but may also occur as nitrogen and N O x . Nitrogen losses can be as much as 33% of the initial nitrogen during composting of poultry manure (Hansen, 1989). Biological Stabilization • Biological stabilization by aerobic composting can be carried out using one of three systems, as described below. A comparison of these 3 systems is made in Table V . l . i) Windrow The most commonly used composting method is the windrow system which involves stacking organic wastes into elongated piles, windrows, that are periodically turned using mechanical equipment. Windrow operations are generally quite simple and require few control measures other than monitoring temperature and moisture. Aeration and moisture are two very important factors and therefore intensive management of the composting process by turning and moisture addition is likely to affect the nitrogen 83 fertilizer value of the finished compost (Shi, 1998). The major disadvantages associated with windrow operations include excess odor generation, susceptibility to upset from adverse weather, substantial land requirement, and a relatively long time for composting and curing for a given volume of material (Hay, 1990). In order to enhance the bio-conversion process, windrow piles are recommended to be turned 6-7 times during the whole thermophilic phase while turnings may not be necessary during the mesophilic phase (Diaz, 2002). The turning not only provides compost aeration but, also homogenization of the compost mass can be achieved, ii) Aerated Static Pile Aerated static pile composting involves piling the material in a static windrow over a gallery of pipes which either introduce air into or the windrow compost pile by blowing air through it (active aeration) or by drawing air through the pile as a replacement for the heated air that rises out of the windrow. Forced aeration eliminates frequent turning but requires a compressor/blower and pipe network to distribute the air. Passive aeration occurs when air heated in the compost pile by the release of from the composting process, rises up and out of the pile and is replaced by cooler air drawn into the pile from outside of it. Passive aeration can save energy costs while being just as efficient as forced or active aeration. Passive aeration requires the proper design of aeration ducts, and thus, the proper prediction of the convective airflow rates created by the temperature differential between the compost and the ambient air (Barrington, 2003). Sartaj et al. (1997) reported that passive aeration had a higher composting rate than active aeration, and did not produce adverse cooling effects and high nitrogen losses, as did 84 active aeration. Aerated static pile and windrow are the most usually employed composting methods for agricultural wastes (Real, 1996). iii) In-Vessel The in-vessel system uses a specially designed, mechanized, enclosed unit that controls the temperature and flow of air. Following preparation, the material is slowly fed into the bays via wheel loaders, gravity-fed hoppers, or conveyor belts. Aeration within the vessel is done by mechanically turning the composting material, by forced draft compressors or blowers, or a combination of these two. Mechanical mixing or turning usually can be performed daily to ensure uniform decomposition. Two stages are distinguished in the in-vessel composting system: a high-rate phase and a curing phase. The first stage is performed in the bioreactors and the second one often in an exterior composting pile. No precise distinction exists between these two stages, but high bio-reaction rates (i.e. rapid biodegradation), high oxygen-uptake rates, high temperatures, and high potential for odour production are essential for the first step (Stelmachowski, 2003). The required decomposition time of an in-vessel system is relatively short — 21 days. The main disadvantages of in-vessel methods are their high capital costs and the requirement for intensive and skilful management (Haug, 1993). For greenhouse wastes application, capital and operating cost may not be an important issue since the greenhouses are required to pay high tipping fees to truck their wastes to landfill. However, space requirement, odour and the quality of the finished compost would be more important. Therefore, in-vessel composting is probably preferable for greenhouses. 85 V.1.2 CRITERIA AND SPECIFICATIONS OF COMPOSTING Under natural conditions, the decomposition process of organic material can extend over a period of months or even years, depending on climatic conditions. However, the natural process can be accelerated by controlling the process factors. Each of these factors has the potential to significantly affect the composting process. Some of the more important factors in the composting operation are: carbon to nitrogen ratio (C:N), surface area and particle size, aeration, porosity, moisture content, temperature, pH of materials and nutrients. Table V.l Comparison of the Three Biological Stabilization Systems Windrow Static Pile In-Vessel General Active systems, most common on farms Effective for farm and municipal use Large-scale systems for commercial applications Labor Increases with aeration frequency and poor planning System design and planning important Monitoring needed Requires consistent level of management / product flow to be cost efficient Capital Cost Minimal Medium High Site Can require large land areas Less land required given faster rates and effective pile volumes Very limited land, due to rapid rates and continuous operations Bulking Agent Flexible Less flexible Must be porous Flexible Active Period > 9 months Ranges 21-40 days Ranges 14-30 days Curing 30+ days 30+ days 30+ days Size Height Width Length 1-2.8 m 3-6m Variable 3-4.5m Variable Variable Dependent on bay design Variable Aeration System Turning the windrow and natural convection Forced blowing or suction air flow through pile (passive aeration) Mechanical systems turn and aerate material within compost bay Process Control Initial mix Turning (6-7 times during thermophilic stage) Initial mix Aeration Temperature or time control Initial mix, Aeration, Temperature or time control, turning Odor Factors From surface area of windrow Turning can be odorous during initial weeks Odor can occur, but controls can be used such as pile insulation and filters on air system Odor can occur, often due to equipment failure or system design limitations 86 Carbon: Nitrogen Ratio (C:N) Carbon and nitrogen compounds are the components most likely to seriously limit the composting process if present in either excessive or insufficient amounts, or, in other words, when the C:N ratio is incorrect. Micro-organisms in compost digest (oxidize) carbon compounds as an energy source, and ingest nitrogen for protein synthesis. The optimum proportion of these two elements should approximate 30 parts carbon to 1 part nitrogen by weight (Hamoda, 1998). C:N ratios within the range of 25:1 to 40:1 should result in an efficient process. Given a steady ratio, micro-organisms can decompose organic material quickly. When the C:N ratio is too high, there is too little nitrogen and decomposition slows. When the C:N ratio is too low, there is too much nitrogen and it will likely be lost to the atmosphere in the form of ammonia gas. This can lead to odour problems. Most materials available for composting do not fit this ideal 30: 1 ratio. Table V.2 lists the C:N ratio of some of the wastes. Different materials must be blended to meet the proper ratio. 87 Table V.2 C/N Ratio of Various Wastes (Haug, 1993) Material Nitrogen As N | % d\v) C/N ratio Night soil 5.5-5.6 6-10 Urine 15-18 0.8 Blood 10-14 3.0 Animal tankage 4.1 Cow manure 1.7 18 Poultry manure 6.3 15 Sheep manure 3.8 Pig manure 3.8 Horse manure 2.3 25 Raw sewage sludge 4-7 11 Digested sewage sludge 2-4 Activated sludge 5 6 Grass clippings 3-6 12-15 Nonlegume vegetable wastes 2.5-4 11-12 Mixed grasses 214 19 Potato tops 4.5 25 Straw, wheat 0.3-0.5 128-150 Straw, oats 1.1 48 Sawdust 0.1 200-500 C:N ratio of a substrate can be calculated according to the following formula (B.C. Ministry of Agriculture, Fisheries and Food, 1991): Equation V.l Composite C:N ratio C:N ratio = weight of C in ingredient a + weight of C in b + weight of C in C + . . . weight of N in a + weight of N in b + weight of N in c + . . . = f% Ca*a*(l-Ma)1 + \% Ca*b* (1-Mb)l+ \% Cc*c* (l-Mc)l + . . . [% Na*a*(l-Ma)] + [% Na*b* (1-Mb)] + [% Nc*c* (1-Mc)] + . . . where a = total weight of ingredient a b = total weight of ingredient b c = total weight of ingredient c Ma, Mb, Mc,... = moisture content of ingredients a, b, c,... % Ca, Cb, Cc,... = % carbon of ingredients a, b, c,... (% of dry weight) % Na, Nb, Nc, ...= % nitrogen of ingredients a, b, c,... (% of dry weight) 88 Surface Area and Particle Size Microbial activity occurs at the interface of particle surfaces and air. The surface area of materials to be composted can be increased by breaking them into smaller pieces. Increased surface area allows the micro-organisms to digest more material, multiply faster and generate more heat. Generally, the smaller the size and more fragile the particle, the greater the biological activity and rate of composting are. However, Hamoda et al. (1998) reported a higher organic decomposition with 40 mm particle size than with 30 mm and 20 mm. This was because with larger particle sizes, the voids between the waste particles were bigger than with smaller sizes, thus oxygen had better access to the particles and degradation proceeded faster. This suggests that even though small particles have increased surface area for microbial activity, oxygen transfer can be limited when the sizes are too small. Haug (1993) developed a model showing that particles with 0.1 cm diameter and lower appeared to be small enough so that oxygen transfer limitations existed. In Table III. 1 of Chapter III, both pepper and tomato year-end wastes after shredding showed less than 10% of particles smaller than 0.1 cm before they were processed in the composter; therefore, it should be able to fulfil the partcile size requirement for oxygen transfer. Aeration Aeration is necessary for a composting process for three basic reasons. First, air must be supplied to satisfy the oxygen demand from organic decomposition (stoichiometric demand). Aeration replaces oxygen-deficient air in the center of the 89 compost pile with fresh air. Rapid aerobic decomposition can only occur in the presence of sufficient oxygen. Second, aeration is necessary for removing water from wet substrates to provide drying (drying demand). Third, aeration is used to remove excess heat generated by organic decomposition to control optimal process temperatures (heat removal demand). Appendix C shows a sample of the aeration requirement calculation used in this study. Aeration occurs naturally when air warmed by the compost rises through the pile, drawing in fresh air from the surroundings. Wind also stimulates aeration. Initial mixing of materials usually introduces enough oxygen to start composting. Oxygen requirements are greatest during the initial weeks of most vigorous activity. Air movement through the compost pile is affected by porosity and moisture content. Regular mixing of the pile, referred to as turning, enhances aeration in a compost pile. Porosity or Free Air Space (FAS) The concept of free air space was first adopted from soil science by Shulze (1962) to establish the relationship between moisture content and the physical structure of composting materials. Porosity refers to the spaces between particles in the compost pile and is calculated as the volume of spaces, or pores, divided by the total volume of the pile. If the material is not saturated with water, these spaces are partially filled with air that can supply oxygen to decomposers and provide a path for air circulation. As the material becomes water saturated, the space available for air decreases. FAS(%) and porosity (P) % are related through the following equation (Haug, 1993) : 90 Equation V.2 Free air space of compost mass FAS = Vg/Vt or (Vt - V s - Vw) / V t where V g is the gas volume, V s is the solid volume, V w is the water volume and V t is the total volume. Equation V.3 Porosity of compost mass P = (V,-V 8 ) /V t or 100(1 -pb/pp). where pb is the bulk density (gem"3) and p p is the particle density (gem"3). Compacting the compost pile reduces its porosity. Excessive shredding can also impede air circulation by creating smaller particles and pores. Turning can fluff up the material and increase its porosity. Adding coarse materials, such as straw or woodchips, can increase the pile porosity, although some coarse materials are slow to decompose. As the compost process proceeds, the porosity decreases restricting aeration. By estimating the consumption of oxygen required for the composting of a wide variety of residues from municipal refuse with different MCs, a 30-36% FAS was recommended to obtain optimum composting (Jeris, 1973 and Madejon, 2002). For agricultural wastes, bulk densities of the mixture less than 550 to 640 kg/m3 are usually adequate (BC Ministry of Agriculture, Fisheries, and Food, 1991). For greenhouse wastes, since shredding and addition of bulking agents are usually required, the compost operators could usually blend the substrates to an optimum bulk density. Moisture Content Moisture plays an essential role in the metabolism of micro-organisms and indirectly in the supply of oxygen. Micro-organisms tend to utilize only those organic 91 molecules that are dissolved in water. Although the maximum rate of transfer of nutrients and waste products takes place in a liquid environment (100 % MC), it is not possible to operate an aerobic composting system where solid substrates are used in this condition. Therefore, the optimum MC is a trade-off between moisture requirements of microorganisms and their simultaneous need for an adequate oxygen supply (Madejon, 2002). By conducting a set of well-controlled incubation experiments, Liang (2003) proved that moisture content was a more dominant factor in impacting aerobic microbial activity of composting than temperature. Even at lower temperatures such as 22 °C, higher moisture content treatments (50%, 60% or 70%) exhibited higher microbial activity (oxygen uptake rate) than those at below 40%. In contrast, at low moisture contents such as 30-40 %, microbial activity was consistently low, even under higher temperature regimes (43, 50 and 57 °C). Particularly, the enhancement of composting activities induced by temperature increment could be realized by increasing moisure content alone. To ensure an adequate composting process, 50-60%o is the recommended range for moisture content (Mckinley, 1985). Hamoda (1998) reported that the optimum moisture content for the composting process is 60%, as the largest decrease in TOC took place under this condition. If the moisture content fells below 40%, the bacterial activity would slow down and cease entirely below 15%. When the moisture content exceeds 60%, nutrients are leached, air volume is reduced, odors are produced (due to anaerobic conditions) and decomposition is slowed. If the pile becomes too wet, it should be turned and restacked. This allows air to circulate back into the pile and loosens the materials for 92 better draining and air drying. Adding dry material, such as straw, sawdust or finished compost, can also remedy an excess moisture problem. If the pile is too dry, water can be added and the pile can be remixed to re-start the composting process. Moisture content of the composting mix can be calculated using the following formula (B.C. Ministry of Agriculture Fisheries and Food, 1993): Equation V.4 Composite moisture content of compost mass (wet basis) Moisture Content = Weight of water in ingredient a + water in b + water in c + ... total weight of all ingredients =(a * MaHfb * MbVKc * M c H .... a+b+c... where a, b, c, Ma, Mb, Mc are the same as in the C:N ratio calculating formula. Temperature Heat generated by micro-organisms decomposing organic material increases the compost pile temperature. There is a direct relation between temperature and rate of oxygen consumption. Higher temperature is associated with a greater rate of oxygen uptake and a faster rate of decomposition. A large variety of mesophilic, thermo-tolerant and thermophilic aerobic microorganisms, including bacteria, actinomycetes, yeasts and fungi have been extensively reported to be present in composting studies. Under aerobic conditions, temperature is the major factor that determines the types of microorganisms, species diversity, and the rate of metabolic activities. For instance, pathogenic bacteria, like Escherichia coli and faecal Streptococci populations decreased, respectively, from 93 2xl0 7 to 3.1xl03 and 107 to 1.5xl03 cells/g waste dry weight at the" end of the thermophilic composting cycle (Hassen, 2001). Under optimal conditions, composting proceeds through three phases : (1) the mesophilic phase, which can last for several days, (2) the thermophilic phase, which can last from a few days to several months, and (3) the cooling and maturation phase which lasts for several months (Figure V. l ) . The length of the composting phases depends on the raw materials being composted and the efficiency of the process. At the beginning of the composting process, mesophilic bacteria predominate, but after the temperature increases to over 40 °C, thermophilic bacteria take over and thermophilic fungi also appear in the compost. When the temperature exceeds 60 °C, microbial activity decreases dramatically, but after the compost has cooled, mesophilic bacteria and actinomycetes dominate again (Tuomela, 2000). Mari et al. (2003) further determined the temperature requirements of the compost microflora at different stages by measuring the respiration rate in compost over a wide range of temperatures during the composting process (before and after each turning). At the start of composting the mean respiration rate at lower temperatures (17-42 °C) was more than two times greater than the mean respiration rate at higher temperatures (42-63 °C), indicating that mesophilic microorganisms initially predominated. However, the ratio of lower temperature respiration rate-to-higher temperature respiration rate dropped below 1 within 48 hours, indicating that a thermophilic microflora were established. The ratio was maintained between 0.5 and 1 for the rest of the active composting period, 94 indicating no further changes occurred in the temperature requirements of the compost microflora. Temperature (*C) P H 60 • i 50 w / 40 30 -20 -10 • mesophilic thermophilic cooling flnci maturation 0 -Time . Temperature •pH Figure V.l . Temperature and pH variation during natural composting process. (Golueke, 1991) The typical composting temperature range is between 32°C and 60°C (Haug, 1993). Hamoda (1998) stated that the optimum temperature for decomposition, as judged by the amount of TOC reduction, was 40 °C. According to the Agricultural Composting Factsheets (B.C. Ministry of Agriculture, Fisheries and Food, 1991), a temperature of 55°C or above has to be achieved and maintained for three consecutive days for effective weed seed destruction. Animal health materials such as drugs and hormones are destroyed if the composting process remains thermophilic for a minimum of three days. Specific time and temperature requirements to eliminate pathogens can be found in Production and Use of Compost Regulations, B.C. Ministry of Environment, Lands and Parks. 95 The variability of temperature associated with composting can limit the ability to detect statistically significant differences between treatments. The use of an inoculum from a single source, e.g. biosolids from wastewater treatment plants, might decrease experimental temperature variations and thus improve process dynamics (Schloss, 2000). pH of Materials Composting may proceed effectively over a range of pH without seriously limiting the process. The optimum pH for micro-organisms involved in composting lies between 6.5 and 7.5. Composting itself leads to major changes in materials and their pH as decomposition occurs. For example (see Figure V.l ) , at the start of composting the mass is at ambient temperature and usually slightly acidic. Soluble and easily degradable carbon sources, such as monosaccharides, starch and lipids, are utilized by microorganisms in the early stage of composting. The pH decreases because organic acids are formed from these compounds during degradation. In the next stage, microorganisms start to degrade proteins, resulting in the release of ammonium with a resulting increase in pH (Tuomela, 2000). Studies have indicated that the activities in the initial phase of composting were reduced when the temperature was raised too quickly under low pH conditions (Smars, 2002). Figure V . l shows how pH will typically change over the composting process. Whatever the pH measured in the starting materials, composting will always yield an end product with a stable pH usually near neutrality. Lime or other buffering agents have sometimes been added to prevent the pH from decreasing below 7, especially at the early stage of composting. The degradation rate of organic matter in pH 96 controlled environments was faster than in those without pH control. The optimum pH for the growth rate and degradation activity of proteins of the microorganisms is in the range of 7-8, while the decomposition of glucose proceeded rapidly at an early stage of composting in a pH range from 6-9 (Nakasaki, 1993). Nutrients Two factors enter into the determination of a suitable nutrient balance: the elemental composition of the microbial cell mass, and microbial metabolism. For reproduction and consequent decomposition to take place, all microorganisms must have access to a supply of the elements of which their cellular matter is composed. Other elements are required which enter into the metabolic activities of the organisms by serving as an energy source or as an enzyme constituent. Many elements are utilized to some extent by microbes, and some are essential to their survival. The relative amount required of each element varies. Those needed in large amounts are termed macronutrients; those in minute amounts are designated as micronutrients or trace elements. The principal macronutrients are carbon(C), hydrogen (H), oxygen (O), nitrogen (N), phosphorus (P), and potassium (K). About 50% of the cell mass consists of carbon, and from 2 to 8 % is nitrogen. Potassium is present in only a fraction of a percent. Hydrogen and oxygen constitute a large percentage of the cellular mass in the form of water and as a part of the cellular material. Trace elements may even be toxic when present in above trace quantities. Generally, micronutrients are present in 97 most wastes in an abundance sufficient to permit satisfactory composting without the need for further additions. Cation Exchange Capacity The cation exchange capacity (CEC) is a measure of the nutrient holding (adsorptive) power of the material and is related to potential use of the compost as a fertilizer. It also is related to the state of decomposition, tending to increase as the material is progressively decomposed and is a useful humification index during composting (Sanchez-Monedero, 1999). However, it is not a measure of maturity or stability. Many other factors influence CEC, but the data provide a basis for comparative purposes and to estimate the potential use of the compost as a fertilizer of a compost-amended mixture. A field trial conducted by Ouedraogo (2001) showed that application of mature compost increased soil CEC from 4 to 6 cmol/kg (centimoles of + charge per kg soil), and improved soil properties and crop productivity. 98 V.2 SPECIFIC OBJECTIVES The overall objective of this part of the project was to design and develop an appropriate technology for the composting of biodegradable wastes generated by greenhouses in the province of British Columbia. The specific objectives were: • To test whether the greenhouse organic wastes could be composted in a well-controlled in-vessel bioreactor with or without the addition of inoculum and bulking agents. • To evaluate the effects of relevant parameters and process controls on composting efficiency. • To evaluate the effects of different substrate recipes, inoculum, bulking agents on the greenhouse wastes composting process with regard to heat generation and degradation of organic matter. • To assess the quality of the finished compost product. • To investigate the feasibility of adopting box-type (in-vessel) composting technology for on-farm operation and to evaluate the economic feasibility of the composting technology. 99 V.3 M E T H O D O L O G Y V.3.1 LAB SCALE EXPERIMENT SYSTEM DESIGN Four lab scale, experimental compost reactors were built and set up in the laboratories of the Department of Bio-Resource Engineering Department, UBC and were later (Lab 5 to Lab 9) moved to the warehouse of Hazelmere Greenhouses for the ease of feedstock loading. Their layout is presented in Figure V.2. Figure V.3 and Figure V.4 are photos of the lab scale reactors. They were constructed from 120 litre plastic barrels, with a perforated acrylic plenum which allowed upward, forced airflow, a removable lid with seal, and headspace exhaust and sample ports. The reactors were thermally insulated with an integral controlled heater to reduce heat loss from the compost mass. They were equipped with air compressors and air diffusers at the bottom. The required aeration rate for the supply of oxygen and cooling was determined in each run by estimating the oxygen uptake from empirical data and calculating the heat output of the compost (Haug, 1993). Sample calculations are shown in Appendix C. An air re-circulation system was set up to provide additional air cycle with or without cooling, inside the reactor, without introducing extra oxygen; a similar configuration was used in the industrial Herhof-Rottebox composting system (Weppen, 2001; Smars, 2001). The condensate from the recirculation system would be pumped to the bottom of the the reactor and mixed with the leachate. 100 f>— Computer Feedback Control (computereonimied) Figure V.2 Lab Scale Composter (Scale 11.5:1) Leachate was collected at the bottom of each reactor and removed via a valve. Evaporated water was trapped as condensate from the inverted cone-shaped aluminum lid and the cooler outside (optional in different runs). This setup enhanced the collection and monitoring of the nutrient and pH changes of the leachate and condensate. Some studies indicated that pH in the condensate and leachate, at least during the pH-shift from below to above 7 in the material, related to the pH in the material (Smars, 2002). The leachate 101 and condensate could be re-circulated internally to increase the moisture content of the compost if necessary. Three thermocouples were place at the top, middle and bottom of the reactors to record the compost temperature during the process. Oxygen levels could be monitored by putting an oxygen probe into one of the openings on the side of the reactors. In the first part of the lab experiments, different process control strategies were used to test their efficiencies on composting greenhouse organic wastes. The purpose of the first set of experiments was to find out the most efficient and cost-effective control strategy. The following aeration control strategies were used in the study : Figure V.3 Lab Scale Reactor (Outside) 102 Figure V.4 Lab Scale Reactor (Inside) Temperature Feedback Aeration Control (Rutgers method) The Rutgers method is probably the most commonly used of all composting process control strategies (Monteoliva-Sanchez, 1996; Bernal, 1998; Paredes, 2002). The control objective of the Rutgers process is to maximize microbial activity by regulating temperature via controlled ventilation of the compost (Finstein, 1983). In the initial stage, rapid temperature rise is encouraged by timer-controlled aeration, usually on a 20 to 30% duty cycle. Once compost temperature reaches the desired setpoint, air is supplied at a sufficient rate to keep the temperature below the set point. Appendix D shows the flowchart and program code used in this study for temperature feedback aeration. 1 0 3 Recirculation Air recirculation capability was added to the temperature feedback setup. This allowed aeration for cooling purposes to be controlled in a closed loop, separate from aeration for the supply of oxygen. The internal air recirculation also resulted in achieving a more uniform temperature distribution throughout the composting mass as compared to no recirculation, and thereby accelerated degradation of the organic matter (Bari, 2001). This system potentially also offers better management of exhaust gases and odour by reducing the volume of exhaust air and extending the residence time of exhaust gases in the compost (Mathur, 1994). Linear Temperature Feedback Aeration Control Linear temperature feedback (Fraser, 1997) was developed as an improvement to ordinary temperature feedback control. To help keep compost oxygen in the desired range, aeration through the pump at the bottom of the reactor was increased linearly based on compost temperature and a model coefficient. Once temperature reached a set point, it was kept below that limit by temperature feedback controlled aeration. As temperature decreased at the end of the process, aeration was decreased based on temperature and another model coefficient. Appendix E shows the flowchart and program code used in this study for linear temperature feedback aeration. In the second part of the lab experiments, the focus was changed to the effects of different substrates on the process performance. Linear temperature feedback was used for aeration control. 104 Each of the lab-scale experiments consisted of an active composting phase, lasting approximately 2-4 weeks, followed by curing, in fabric bags, for 1-3 months. Capacity of each of the reactors was approximately 20 kg of feedstock. Greenhouse plant wastes were collected from Hazelmere Greenhouse (pepper and tomato) directly. Before the start of each experiment, the proper ratio of substrate mix was calculated where C:N ratio and bulking density were used as references. The wastes were weighed and mixed thoroughtly by hand on a concrete pad. Shredding and addition of bulking agents were done whenever necessary. The feedstocks were then fed into the composter carefully by hand. Feedstocks were sampled before and after active composting, and occasionally during curing; measurements were focused on the active composting phase since the lab system was primarily developed for this. hi total, there were 9 lab scale composting experiments investigated in a period of about 18 months. They were designed to investigate different parameters of composting and their effects on the process and compost quality. Table V.3 lists all the lab experiments with their purposes. There were 2 reactors built for Lab 1 and Lab 2, and 4 reactors built for Lab 3 to Lab 9. Even though all the reactors were built under the same configuations at the same time, minor variations were unavoidable, e.g. factory variations from the heating blanket, thickness of barrel, insulation, tubings, thermocouple wires, aeration pumps, etc. In any case, when one parameter was tested more than one time, different reactors were used in different runs to avoid variations coming from the reactors. For example, alder bark with non-shredded greenhouse waste was tested in 105 Reactor C in Lab 5 and was re-tested in Reactor A in Lab 6 in order to confirm the results were caused by the differences in substrates but not by differences in reactors. Table V.3 List of Lab tests Lab Test Purpose Sub Sets Remarks Testing of composting process strategies Test 1 Find out whether manure had to be added to start the greenhouse wastes composting; test composters With manure vs without manure Monitor odor problems with manure Test 2 Investigate the effect of air recirculation With air recirculation vs without air recirculation Test 3 Investigate the effect of air recirculation; investigate the efficiency of using curing compost as biofilter With air recirculation vs without air recirculation Track odor emissions Test 4 Investigate the effect of air recirculation on cooling and aeration Fresh air vs recirculated air aeration; Fresh air cooling vs recirculation cooling Testing of effects of various substrates Test 5 Find out whether shredding is needed; Investigate different bulking agents Shredded vs non-shredded prunings; Hemlock bark vs alder bark Track studies (chemical analysis) Test 6 Find out whether shredding is needed; Investigate different bulking agents Shredded vs non-shredded prunings; Hemlock bark vs alder bark Test 7 Find out whether shredding is needed; Investigate different bulking agents Shredded vs non-shredded pruning; Fresh alder bark vs composted alder bark Test 8 Find out the best mixing ratio of wastes and bulking agents 4 different mixing ratios Test 9 Finding out the best mixing ratio of wastes, bulking agents and used sawdust 4 different mixing ratios In the course of the experiments, some parameters were re-tested a couple of times, for example, air-recirculation was tested in Lab 2, 3 and 4, and alder bark as a bulking agent was tested in Lab 5, 6, and 7. However, those experiments could not be regarded as identical replicates based on two reasons : First, the characteristics of the plant wastes changed during the season. For example, in the beginning of the year, there were only leaves in the wastes; in the middle 106 of the year, there was a lot of small cull fruit in the wastes, which caused higher moisture content and higher sugar content (readily biodegradable carbon); near the end of the year, there were a lot of vines, and therefore a higher C:N ratio; at the end of the year, the wastes were mixed with leaves, vines, cull fruit, and sawdust. Since each composting run might last for a month, storing the same batch of wastes for the next run might cause deterioration of the wastes, whereas using another batch could not guarantee exactly the same composition (C:N ratio, moisture content, bulking density, etc) of wastes. Secondly, there were uncontrollable external factors, such as outside temperature and humidity, that could cause experimental variations in different runs. Even though each reactor was well insulated, there was still heat loss due to occasional low outside temperature. The temperature of the aeration from ambient air might affect the temperature profile of the composting process. This was especially significant when the reactors were moved to the warehouse of Hazelmere Greenhouses (no effective heating in winter months) for the ease of feedstock loading. Due to the above reasons, each set of experiments was carefully planned. First of all, there were two sets of experiments: 1. Testing of composting process strategies; 2. Testing of effects of various substrates. The purpose was to find out the differences in control strategies and substrates, and to see whether similar results were able to be repeated. Secondly, in each run, when there were two design parameters (e.g. X,Y), with two variables for each parameter, one of them would be kept constant (e.g. X) for two reactors so that the second parameter (e.g. Y) could be varied for comparing 107 independently, i.e. {(Xi,Yi), (Xi,Y 2), (X 2 ,Yi), (X 2,Y 2)}. By doing so, each variable could be tested twice, achieving 4 comparisons, i.e. test of parameter X : (Xi,Yi) vs (X 2 ,Yi) and (Xi,Y 2) vs (X 2 ,Y 2 ) ; test of parameter Y : (X,,Y,) vs (Xi,Y 2) and (X 2 ,Y,) vs (X 2 ,Y 2 ) as shown below: Parameter X i x 2 Y i Reactor A X , Y , ^ f Reactor B - * X 2 Y , f Y 2 I X,Y 2 «-Reactor C J - • X 2 Y 2 Reactor D Lab 4, 5, 6, 7 followed this design to compare different aeration and cooling methods, different bulking agents under shredded or non-shredded substrates conditions, etc. By doing so, a parameter could be compared in the same run and also re-tested in different lab runs. 108 V.3.2 PILOT SCALE EXPERIMENT SYSTEM DESIGN A scaled-up compost experiment was designed based on the findings of the lab-scale studies and the commercial Herhof-Rottebox composting system (Weppen, 2001; Smars, 2001). A used, standard 20', steel, refrigeration, shipping container (Cratex Container Sales Ltd., Coquitlam, BC) was modified to become the reactor, because it was already well-insulated for minimizing heat loss. First the refrigeration unit was removed and an epoxy painted, steel grating was installed on the floor. Thermocouple probe fittings and drain fittings were installed on the sidewall and bottom-end of the container. The bottom was lined with a leakproof PVC liner. An epoxy painted steel grating was installed on top of the PVC liner, suspended on top of 4" x 4" cedar beams running longitudinally. Nine sections of perforated 1" schedule 40 PVC piping were installed between the beams, with gravel on top for even air distribution under the steel grating. These 1" pipes were connected to the main aeration pipe, a 4" PVC schedule 80 pipe. The aeration floor was sloped down and in from its outer edges to encourage liquid formed during composting to collect under the perforated centre portion of the aeration floor. The perforated centre portion of the aeration floor was arched or peaked to direct more air toward the centre of the composting mass while minimizing the amount of air following the path of least resistance up the side wall (Brown, 1997). A computer with data acquisition and control card (Advantech Co. Ltd -Model 711), running Labtech Control software was installed, with connections to thermocouples for temperature monitoring. A 110 VAC 1 HP centrifugal blower (Dayton Co. Ltd) was connected to the main aeration pipe for air supply. Leachate was collected in a 1200 L 109 plastic tank through a diaphragm pump. Temperature feedback control (Rutgers method) was used for aeration control. Prior to each experiment, feedstocks were mixed by a Bobcat loader on an asphalt pad at the parking lot of Hazelmere Greenhouse. After mixing, the feedstock was loaded into the container, to a depth of approximately 4 feet. Any unused portion of the perforated floor was covered in plastic and heavy sawdust bags were put on top to avoid short circuiting of air flow. The container was closed, the aeration control program started and temperature monitored. At the end of the active phase, the material was unloaded using the Bobcat, and piled in the curing area to a height of approximately 5 feet. The pile was then covered with Compostex fabric (Texel Inc., Saint-Elzear-de-Beauce, PQ) to shed water and allow airflow, and allowed to cure. Periodic temperature and oxygen measurements were taken in the pile. Figure V.5 provides the dimensions of the compost container. Figures V.6 and V.7 are photographs of it. 110 Notes: Outside height 2.6 m Usable inside floor dimensions 599 cm x 218 cm Usable inside height (approx.) 1.5 m Usable volume 18 m 3 (23.5 yd3) Figure V.5 Pilot Scale Bio-Conversion Reactor Design Figure V.7 Pilot Scale Composter (inside - installing air piping) 112 Physico-chemical monitoring and analyses (lab and pilot experiments) Temperature was continuously monitored inside every composter. Due to a limited budget, only one oxygen probe was available and therefore only one reactor could be monitored for the oxygen profile in each run. Chemical analysis of compost is a frequently used method for establishing the quality of a compost (Bidlingmaier, 1999). At the beginning and end of each lab composting test, the intial waste mixture and compost were analyzed with regard to the most relevant physico-chemical parameters, such as C/N ratio, moisture content, porosity (lab 6 only), ammonia, nitrates, pH. In each case, 3 to 5 grab samples were taken and mixed as a composite sample for analysis (Johnson, 1993). Gravimetric moisture content was determined by evaporation at 105 °C for 24 hours (Amer. Sco. Agron, 1982). The pH was measured using distilled water dilution as Standard Methods (Amer. Soc. Agron., 1982). Total organic carbon (TOC) was measured by combustion at 680 °C and CO2 measurement, using a Shimadzu TOC-5050 Total Organic Carbon Analyzer with SSM-5000 Solid Sampling Module. Water soluble ammonium, nitrate and phosphorus were measured using water extraction and the Technicon Autoanalyser II industrial method. Total nitrogen was measured using ignition at 950 °C in a Leco FP228 Nitrogen Determinator. Statistical Analysis The lab experiments were set up to compare different treatments of process controls and mixing substrates by the 4 lab reactors (except Lab 1 and Lab 2 had only two 113 reactors) as described in the methodolgy section. For physico-chemical analysis, means and standard deviations were calculated. For the pilot scale experiments, since there was only one reactor, each run was analyzed and investigated separately without direct replication. 114 V.4 RESULTS AND DISCUSSION V.4.1 LAB SCALE COMPOSTING DATA TESTING OF COMPOSTING PROCESS STRATEGIES V.4.1.1Lab 1 Test This was the first composting run, and was intended to test the compostabilty of greenhouse wastes in general. More specifically, it was designed to test whether an inoculation of microorganisms (chicken manure) was needed. Through this test, a preliminary idea of proper C:N ratio, bulking agent ratio, moisture content, temperature and aeration control could be established. In this case, the major components were vines, leaves and sawdust (see Table V.4). The vines are the stems or branches of the tomato plants. They smelt like pickles and were very fibrous. The moisture content was about 60%. The leaves were less fibrous and looked like grass clippings. They were very wet (about 80% MC). The sawdust had been used as growing media, which thus contained a lot of roots, and was about 50% MC. The chicken manure was from the UBC farm and was used as an inoculum here. Chicken manure was found to support Trichoderma (Gilocladium) virens growth and viridiol production at levels capable of controlling weed emergence and growth in greenhouse trials (Hutchinson, 1999) and therefore has been one of the most common ingredients for compost. The wood shavings were dry and used to provide structure and porosity to the process. 115 The regulatory requirements for composting to be classified as a process to further reduce pathogens (PFRP), mandate that the minimum operating temperatures must be maintained at 55 °C or above for three days (USEPA, 1994). The provincial government of British Columbia (B.C.MWLAP, 2002) requires compost to be treated aerobically for 14 days or longer at 40 °C or above in order to classify the composting process as a vector attraction reduction process (VARP). These two requirements were only considered as guidelines in this study, but not a "success-determination" factor for the process since the lab scale composter showed limitations in achieving long-duration, high temperatures due to the small amount of organic mass and the high heat loss (Kaiser, 1996; Vinneras,1 2003). The compost mix recipe for Lab 1 is presented in Table V.4. Figure V.8 plots the temperature profiles recorded over the composting period. Feachem et al. (1983) used 45 °C as a benchmark temperature for thermal inactivation of several kinds of common pathogens. By using the formulae given in Table V.5, the degree of inactivation could be calculated at temperatures above 45 °C, which can be easily attainable in aerobic digestion. For example, if the temperature of the compost mass was 50 °C, the hours required to inactivate Salmonella would be 13.94 hours, using the equation : t = 7 5 . 4 x l 0 - ° - , 4 6 6 ( T - 4 5 ) where t is the time required to attain no viable Salmonella (hour) T is the temperature above 45 °C 116 Table V.4 Lab 1 Compost Feedstock Recipes Hatch Mater ia l s VVt(lb) \vt% Moisture Content % Composite MC (wet basis) C:N Ratio Composite C:N Ratio L a b i A Shredded vines 13 37% 82.0 78.9 18.8 24.0 Shredded leaves and cul l fruit 6 17% 87.2 5.2 W o o d shavings 2 5.7% 32.8 99.4 Used sawdust 9 2.6% 75.0 73.4 Chicken manure 5 14% 65.4 23.0 Lab I B Shredded vines 14 40.0% 82.0 78.4 18.8 25.4 Shredded leaves and cul l fruit 7 20.0% 87.2 5.2 W o o d shavings 2 5.7% 32.8 99.4 Used sawdust 12 34.3% 75.0 73.4 Chicken manure 0 0.0% 65.4 23.0 Salmonella is a very common pathogen, and is often used as an indicator for health hazardous compost (Watanabe, 1997; Tiquia, 1998; Sidhu, 2001). Therefore, the three 117 days at 45 °C (Duratiori45°c), for inactivation of Salmonella, was also used as one parameter to compare compost processes in this study. Table V.5 Equations for the time in hours (t) required to attain no viable organisms (equal to 12log i 0 )* of different pathogens at different temperatures (T) above 45 °C. Organism 39! n Eq nation V\ lira l=45"C Enteroviruses Virus t ^ 5 5 T 7 l F ™ ^ Time (t) = 55.9 hr or 2.3 days Salmonella Bacteria t = 7 5 . 4 x l O - 6 l ^ T - ^ Time (t) = 75.4 hr or 3.1 days Vibrio cholera Bacteria t = 0 . 8 9 x l O ^ , , 7 < T - 4 5 > Time (t)=0.89 hr or 0.04 days Shigella Bacteria t = 1 3 . 8 x l O * l M f r - 4 S ) Time (t)=13.8 hr or 0.58 days Ent. Hystolica Protozoa t = 2 1 . 3 x l O - " " * f f - ' U ) Time (t)=21.3 hr or 0.89 days Ancylostoma Helminth t = 9 . 3 1 x l 0 4 U 3 4 W ' - ^ Time (t) = 9.3 hr or 0.39 days Ascaris Helminth t = 1 7 7 x l O - * l w ' T - ^ Time (t) = 177 hr or 7.4 days Schistosoma Helminth t = 1 0 . 0 x l 0 ^ 1 8 4 4 ( T - 4 5 > Time (t) = 10 hr or 0.42 days Taenia Helminth t = 6 . 6 x 1 0 - ™ " ^ Time (t) = 6.6 hr or 0.28 days Source : Feachem et al (1983) •Note : The values with no viable organisms was interpreted as an inactivation of approximately 121ogi0, which was considered to be a level that determines sterilization by the Parental Drug Association. The parameters involved in the temperature record, that were used for discussion in this study, were : (1) maximum temperature (TEMPm a x); (2) time required for temperature to reach 55°C (TIME55°c); (3) duration of the temperature above 55°C (Duration55°c); (4) duration of the temperature above 45°C (Duration45°c); (5) duration of the temperature above 40°C (DuratiorLwc). When the temperature reached over 55 °C for both reactors, the oxygen contents inside the reactor were checked. Reactor A and Reactor B recorded oxygen levels of 18.2 % at 56.3 °C and 18.1 % at 55.5 °C respectively. This showed that the aeration rate was adequate to provide an oxygen level of above 16 %, which is said to enhance microbial activities and prolong the mesophilic phase in composting (Beck-Friis, 2003). At the peak 118 temperature, Reactor A (with chicken manure) gave some offensive ammonia odour, while Reactor B (without chicken manure) gave minimal odour. The reactors were opened for remix and observations at Day 14. Reactor A still had some ammonia smell while Reactor B already showed a pleasant earthy "compost-like" smell. This might plausibly be caused by some undegraded chicken manure. Extensive fungal growth was noted inside the compost in both reactors which showed that the temperatures were not high enough to kill or suppress fungal growth. As shown in Figure V.8, Reactor A heated up first and achieved an overall higher temperature in the thermophilic stage of the process (higher T E M P m a x and TIME55°c). PFRP was achieved in Reactor A (Duration55°c =3 days) and was not achieved in Reactor B ( D u r a t i o n 5 5 ° c < 1 day). This showed the chicken manure inoculation increased TEMPmax,TIME55<>c and Duration55c. Even though Reactor B did not heat up as quickly nor attain as high a temperature in the thermophilic stage and did not achieve PFRP, it still had a very good temperature profile and maintained a temperature of about 40 °C for the mesophilic stage of the process. The Duratiori4o°c and Duratiori45°c were almost the same for Reactor A and Reactor B. In fact, the temperature of Reactor B spiked up again after Day 8 and surpassed that of Reactor A from Day 9 to 11. This implied a higher degradation rate of Reactor B during the mesophilic period since some readily degradable materials were not fully degraded in the relatively short thermophilic period of Reactor B. The overall degree of degradation can be indicated by the loss of carbon in the substrates, though a small amount might be lost in the leachate, as shown in Table V.6. Based on the 119 loss of carbon, both reactors achieved a similar degree of degradation at the end of the process (31.3% vs 32.0%). Table V.6 Loss of carbon over composting period as calculated via mass balance (with manure vs without manure) Reactor Day % M C M M w M s % C M , A M , A M , / M C A (with manure) 1 78.89 ±1.13 35 27.60 7.40 40.82 ±0.58 3.02 16 77.00 ±1.11 23 17.71 5.29 39.22 ±0.93 2.07 -0.95 -31.32 % B (without manure) 1 78.43 ±0.95 35 27.44 7.56 36.52 ±0.76 2.77 16 80.29 ±1.17 25 20.07 4.93 38.10 ±1.08 1.87 -0.88 -32.01 % Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. V.4.1.2Lab 2 Test After the first run, a potential odour problem resulting from using manure in the feedstock was recognized. Therefore, in this run, 2 modifications were made: no manure was added, and an air re-circulation system was added to one of the reactors. By doing air re-circulation, odour can be minimized, especially during the thermophilic stage. Internal recirculation of air has been said to achieve a more uniform temperature and moisture distribution and thereby an accelerated degradation of the organic matter (Bari, 2001). The overall aeration requirement should be reduced by recirculating air in composting (Das, 1997). Figure V.9 illustrates how the air recirculation system compares to the system without air recirculation in the following lab runs (Lab 2, 3 and 4). In Lab 2, there 120 were only two reactors (with and without recirculation). In Lab 3, two more reactors were added as biofilters. In Lab 4, all four reactors were used as composters. EXHAUST VALVE AIRFLOW INLET METER VALVE FRESH AIR PUMP AIRFLOW NEEDLE METER VALVE RECIRCULATING AIR PUMP n EXHAUST R E A C T O R D AIR RECIRCULATION REACTOR SETUP (Lab 2, 3 & 4) AIRFLOW INLET METER VALVE F H X } FRESH AIR PUMP EXHAUST R E A C T O R C NON-RECIRCULATION REACTOR SETUP Figure V.9 Lab 2,3,4 Reactor Design (with and without air recirculation) The air recirculation system was designed to provide both aeration and cooling. The aeration was first done by the fresh air pump and was evenly distributed from the diffuser at the bottom of the composter. Based on previous composting tests (Fraser, 121 1997) with the same size of reactor, the aeration control was set at a maximum of 7 min/L, instantaneous based on Linear Temperature Feedback Control as mentioned in Section V.3.1. The spent air from top valve of the reactor then either passed through a metal coil (for cooling) or by-passed (for aeration) back to the reactor. Table V.7 presents the feedstock recipes for composting in Lab 2. Figure V.10 plots the temperature profiles. Table V.7 Lab 2 Compost Feedstock Recipes Batch Materials Wt Wt% (lb) Moistun Content Composi MC C:N Rat Composi C:N Rat Lab2A Shredded vines 20 50% 75.0 66.25 18.75 29.19 Without air re-circulation Shredded leaves 10 25% 82.02 5.22 Hog fuel 10 25% 32.79 74.02 Lab2B Shredded vines 20 50% 75 66.25 18.75 29.19 With air-recirculation Shredded leaves 10 25% 82.02 5.22 Hog fuel 10 25% 32.79 74.02 REACTOR A REACTOR B / / 0 2 4 6 8 10 12 14 16 Ambient Temp=25.S±1.0 (avg±SD) TIME (Days) Figure V.10 Temperature Profiles of Lab 2 122 TEMPmax and TIMEs5oC were about the same for both reactors (approximately 56 °C in 2 days). However, Duration5s°c was longer in Reactor A than in Reactor B. Both reactors finished active composting in about 14 days. Reactor A remained longer in the thermophilic stage while Reactor B sustained longer in the mesophilic stage. The composts were dark brown and very wet in both reactors. Both reactors had a very strong odour of ammonia which was an unpleasant smell (even worse than the Lab 1 with manure). Filamentous fungi were observed on the surface of Reactor B. Table V.8 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) Reactor Day % M C M M , M s % C M, A M , z M V l c / M c ' A (without air recirculation) 1 66.25 ±0 .57 45.0 29.81 15.19 33.99 ±1 .32 5.16 15 74.60 ±0.98 24.3 18.13 6.17 41.38 ±2.01 2.55 -2.61 -51 % B (with air recirculation) 1 66.25 ±0 .57 45.0 29.81 15.19 33.99 ±1 .32 4.77 15 73.78 ±1 .02 20.3 14.94 5.31 41.23 ±1 .78 2.19 -2.97 -58 % Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. The loss of carbon was high in both reactors, as shown in Table V.8, with Reactor B having a higher loss. This could have been a result of air-recirculation (Bari, 2001). Odour problems, especially of ammonia, were still present even without adding manure. Ammonia is the greatest nuisance odor compound among the exhaust gases that evolve during composting processes. It is generated by facultative and strict anaerobes, in raw materials with high concentrations of ammonium and nitrate (Nakasaki, 2000; 123 Ingham, 1998); in this case the vegetable leaves. The odour problem could have been improved by turning the compost more frequently in the beginning of the process (Illmer, 1997) or adding a biofilter made from compost (Liang, 2000; Delhomenie et al, 2002) to treat the off-gas. Better process control to avoid anaerobic conditions in the reactor can ) largely reduce odour. Air-recirculation helped in maintaining a more constant and higher temperature since hot air from the reactor was recycled as the aeration gas. V.4.1.3Lab 3 Test Lab 3 was conducted to verify the advantages of air-recirculation for cooling and to test the use of re-used compost in a biofilter. Refer back to Figure V.9 for the reactor setup. In order to avoid odour in the material handling and composting process, composts from Lab 2 was used as inoculum instead of manure. Some studies have shown that the composting process could be enhanced by adding bacterial inoculum or compost which contained a diversified bacterial population (Faure, 1991; Lei, 2000; Ichida, 2001). The possibility of using finished compost (from Lab 2 and Lab 1) in external biofilters, to control odours was also investigated. The composition of the substrates is listed in Table V.9. 124 Table V.9 Lab 3 Compost Feedstock Recipes Batch Materials YVt (lb) Wt% Moisture Content % Composite MC C:N Ratio Composite C:N Ratio Reactor A Shredded vines 25 62.5% 82.02 82.83 9.60 9.13 Without air re-circulation Shredded prunings 10 25% 87.77 5.58 Compost from Lab 2A 5 12.5% 77.00 13.86 Reactor B Shredded vines 25 62.5% 82.02 82.83 9.60 9.13 With air-recirculation Shredded prunings 10 25% 87.77 5.58 Compost from Lab 2A 5 12.5% 77.00 13.86 Reactor C (Curing reactor, also used as biofilter) Compost from Lab 2B 20 80% 73.78 75.08 16.28 17.20 Compost from Lab 1B 5 20% 80.29 20.86 Reactor D Compost from Lab 2B 20 80% 73.78 75.08 16.28 17.20 (Curing reactor, also used as biofilter) Compost from Lab IB 5 20% 80.29 20.86 In this run ammonia that caused the odour problems was measured at different stages of the process. Some bio-degradable twines were tested for bio-degradability in this test. The results have been analyzed in Chapter IV. ii r A I j — K£ — Ri ;acto jacto r A -r B r ixr i W\ *, k I 1 11 i • 1 U . it flh II \ / If! M fl M 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Ambient Temp=24.7±1.0 (avg±SD) Time (day) Figure V. l l Temperature Profiles of Lab 3 125 TIME55°c of both reactors were about the same, since the air-recirculation function did not start until the temperature had reached the set point temperature. Both reactors achieved PFRP (Reactor A Duration55<>c = 6 days; Reactor B Duration55°c = 5 days). The T E M P m a x of Reactor A (without recirculation) overshot the set point temperature (60 °C) temperature even with a 100% aeration cycle (7 L/min) through Linear Temperature Control. On the other hand, TEMP m a x of Reactor B (with air recirculation at 7 L/min) could be effectively maintained at or below the set point. Resulting in a prolonged mesophilic degradation period, i.e. longer Duration45°c , which resulted in a higher degree of carbon loss, i.e. 51% vs 36% for reactor A without air recirculation (see Table V.10). Smars et al (2002) have demonstrated that the degradation time could be reduced by preventing the temperature from rising until the pH had reached a certain value after the initial phase of the process, because microbes, active in the beginning of the process, were hampered by high temperature. Another advantage found for air recirculation in composting was that the temperature fluctuations were much less in each aeration cycle. With recirculation, the temperature was controlled to within 1 °C whereas without recirculation, temperature fluctuated up and down by 5 to 8 °C. This was due to the recirculation of hot air in Reactor B. The effect of temperature variations may cause qualitative shifts in microbial species composition and special tools for incubation of samples in a gradient of stable temperatures have a long record of use in the studies of microbiology (Thamdrup, 1998; Elsgaard, 2002). Recirculation of hot air inside the 126 reactor should also enhance the homogeneity of compost temperature and compost quality (Tiquia, 2000). Table V .10 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) - Lab 3 Reactor l)a\ %\1C Mi M„ M, n m \M, \M,/M(. A (without recirculation) air 1 82 .83± 1.39 40.0 33.13 6.87 44.24 ±1 .56 3.04 25 78 .52± 0.81 25.7 20.20 5.52 35.19 ±0.65 1.94 -1.09 -36% B (with recirculation) air 1 82 .83± 1.39 40.0 33.13 6.87 44.24 ±1 .56 3.04 25 77 .06± 0.92 24.0 18.49 5.51 26.90 ±0 .34 1.48 -1.56 -51% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. Another purpose of this test was to verify whether odour problems could be mitigated by applying air-recirculation and a biofilter. Ammonia samples were taken at Day 3, Day 5 and Day 10. The concentrations of ammonia and air outflow rates were recorded and emissions were then calculated. All results are shown in Table V . l 1. The biofilters were arranged as shown in Figure V.9. Table V. l l Comparison of Ammonia Emissions Inlet NH_, Avg. Air out Emission Kate Outlet N H 3 % Removal (ppm) - before rate (L/min) (ug/min) (ppm) - after biofilter biofilter Reactor A 240 6.0 1034 40 83% Day 3 Reactor B 250 2.3 413 40 84% Day 3 Reactor A 270 3.9 756 30 89% Day 5 Reactor B 360 2.3 595 80 78% Day 5 Reactor A 230 1.8 297 0 100% Day 10 Reactor B 270 1.8 349 0 100% Day 10 • 127 The results presented in Table V . l 1 show that the emission rate of NH3 was much higher from reactor A (no air recirculation) than from reactor B (with air recirculation) after 3 days of composting. Emissions from reactor A continued to be higher after 5 days of composting although the differences in emission trates between the two reactors became smaller after 5 days. Emissions from reactor A became slightly less than reactor B after 10 days during the mesophilic period. The high emission levels of NH3 at day 3 in reactor A are attributable to the high air flow rate in the beginning of the process when reactor A used a lot of fresh air (full cycle of 7 L/min) to cool down the compost mass while reactor B used air recirculation cooling. At day 5 the temperature was almost at the setpoint (60 °C), and therefore the difference in air flow rate between the two reactors became smaller. At day 10, the air flow rates became the same in both reactors since no cooling was needed. The emission rate was slightly higher in reactor B this time because the ammonia concentration was slightly higher in reactor B probably due to the recirculation of ammonia gas back into the reactor in the cooling periods. Since the emission rates from reactor B with air recirculation were less in the thermophilic period when cooling was needed, it would seem that air recirculation was effective in reducing ammonia emissions in the thermophilic period and hence in reducing at least part of the odour emissions. The use of recycled compost (from Lab 2 and Lab 1) as media in the biofilters also proved effective in removing ammonia (odour) from the process. In most of the cases, especially when the ammonia concentrations were high, the removals ranged from 78% to 100%. Half of the compost from Lab 2 was "cured" in the biofilter and the other 128 half was cured in ambient air as a control. The ammonia and nitrate concentrations in the compost cured in the biofilters were higher (averaged 25 mg/L vs 7 mg/L and 480 mg/L vs 285 mg/L respectively). This showed that by using compost as a media for biofilter, the ammonia-N was absorbed in the compost media. This was a way to conserve nitrogen in the process. V.4.1.4Lab 4 Test As a continuation of Lab 2 and Lab 3, Lab 4 test was designed to test the differences between cooling and aeration methods. Lab 4 was performed during the clean-up time of Hazelmere Greenhouses. Therefore, shredded year-end pepper wastes, which had about 75% vines and leaves, and 25% used sawdust medium, were used. Al l 4 reactors were filled with 40 lb (18.2 kg) of shredded year-end pepper wastes and 5 lb of compost from Lab 3 as inoculum. Table V.12 shows the different aeration methods used in each reactor. Table V.12 Different aeration methods for Lab 4 Aeration Fresh air Cooling method Recirculation IF method instantaneous How and How control method and cooling control method instantaneous air flow rate set point Reactor A Fresh air 7.0 L/min L T F * * Fresh air N/A 60 °C Reactor B Fresh air 7.0 L/min L T F Recirculation T F 7 L/min 60 °C Reactor C Fresh air + non-cooling recirculation* 7.0 L/min L T F Recirculation T F 7 L/min 60 °C Reactor D Fresh air 1.5 L/min (continuous) None N / A N / A *Air recirculated without passing through heat exchanger, to minimize cooling effect * * L T F - Linear Temperature Feedback (Fraser, 1997) *** T F - Rutger's Temperature Feedback (Bernal, 1998) 129 A few comparisons could be achieved using the above matrix : Reactor A vs Reactor B - Cooling method (fresh air vs recirculation); Reactor B vs Reactor C - Aeration source (fresh air vs fresh air+recirculation); Reactor D vs Reactor A,B,C - Aeration control (no control vs control of aeration and cooling). 80 -70 -Fresh Air Cooling-Reactor A ^ — Recirc Cooling-Reactor B — Recirc Cooling plus Periodic Recirc-Reactor C - No Cooling-Reactor D » 60 -50 -O d 40 -Tem 3 20 -10 0 -0 1 2 3 4 ! Ambient Temp=15.3±2.2 (avgtSD) Tim. (d) 1 f 1 i ( Figure V . 1 2 Temperature Profiles of Lab 4 All 4 reactors began to heat up right away, and had reached 60 °C by around Day 2, while Reactor D (no cooling) reached almost 70 °C (see Figure V.12). Reactor A, with fresh air cooling (100% cycle at 7 L/min), over-shot the set point temperature (60 °C) for 12 hours and reached at maximum (TEMPmax) of 65 °C. Reactors B and C were well controlled below the maximum temperature set point (TEMP m a x = 60 °C) by air-recirculation cooling. This further confirmed the results from Lab 3 that air recirculation was an effective method for controlling the process temperature below the set point. Duration55°c of Reactors A, B, C and D were 2.2, 2.4, 0.9 and 3.4 days respectively, and therefore only Reactor D could pass the PFRP requirement. Duration4s°c of Reactors A, 130 B, C and D were 3.7, 3.8, 2.0 and 4.6 days respectively, and therefore Reactors A, B and D passed the pasteurization requirement for Salmonella. Not much odour was released from these reactors. Reactor C seemed to finish the process at Day 6, while the rest of the reactors did not drop back to ambient temperature until Day 8. The amounts of leachate and condensate were different in the different reactors (see Table V.l3). Most composting processes try to minimize the volume of leachate and condensate since they are generally regarded as wastewaters, unless they are diluted to a non-toxic level (see Chapter VI.4.3). Studies have shown that the amount of leachate and condensate produced was dependent on the percentage of evaporative heat loss, the moisture content of the composting mass, the airflow applied, and the temperature decrease of the air used in the air conveying system prior to discharge to the atmosphere (Bari, 2001). Since the air recirculation was an upward air flow, it helped in keeping the moisture in the compost mass rather than dripping to the bottom since moist air kept recycling from bottom to top through the compost mass. In this case, Reactor C out-performed all the other reactors with only 400 ml of leachate (less than half of Reactor A and B, and only one-third of Reactor D) plus zero condensate. This obviously was one big advantage of air recirculation all the time which retains the moisture in the compost (79% MC with Compost C, 76% with Compost A, 77% with Compost B and 77% with Compost C - see Table V.l3) and minimized the amount of wastewater products (leachate and condensate). Condensate was low and leachate was high in Reactor B when compared to Reactor A. It could plausibly because part of the condensate of Reactor B 131 was collected as leachate during the air recirculation periods, i.e. over 60 C. All reactors were emptied at Day 8 since they all had dropped to ambient temperature at that point. Table V.13 Leachate and Condensate from Reactors Reactor A Reactor Ii Reactor C Reactor D Leachate (ml) 750 850 400 1200 Condensate (ml) 200 15 0 210 Moisture content of finished compost 76.06+1.32 77 .10±1.03 79.13+1.11 77 .22±0 .97 Many studies in composting have reported vertically non-homogenous temperature and moisture distributions in the composting mass (Koenig, 1998). Another purpose of this run was to investigate various aeration and cooling modes to try to improve the vertical distribution of temperature and moisture throughout the depth of the composting mass, and thereby increase the rate of organic degradation. Therefore the temperature differences between the upper (10 cm below the top surface) and lower portions (10 cm above the bottom) of the composting mass were monitored and compared throughout the process. As well the moisture contents were compared at the end of the process (Table V. 14). Table V.14 Vertical variation (top vs bottom) in temperature and moisture content in reactors with different aeration and cooling Reactor A Reactor B Reactor C Reactor D Aeration / cooling method Fresh air/ Fresh air Fresh air/ Recirculation Fresh air + non-cooling recirculation/ Recirculation Fresh air (fixed) Average Temperature Variance 4.75 °C ± 1.85 (189)* 4.19 °C ± 1.97(189) 2.86 °C ± 1.35(189) 5.82 °C ± 2 . 3 0 ( 1 8 9 ) Average Moisture Content Variation 3.7 % ± 1.24 (3) 3.2% ± 1.13(3) 1.4% ± 0.63 % (3) 2.3 % ± 0.94 (3) * ±Standard Deviation (Number of samples, n), n for temperature : # of measurements in 8 days 132 Since Reactor D was under continuous aeration from the bottom, the compost temperature was always cooled by the ambient air (15-20 °C) at the bottom layer. Therefore, the vertical temperature variance in reactor D was the highest. Reactor A was under intermittent aeration, and when there was no aeration, air was mixed by natural convection inside the reactor. Therefore, the vertical temperature variance was less than Reactor D. Reactor B had an air recirculation period for cooling when the temperature was above 60 °C and that explained why the variance was a little less than Reactor A. Reactor C, with full-time air recirculation, achieved the lowest vertical temperature variance, and thus led to a more homogeneous temperature distribution. In all the cases, moisture contents were higher at the bottom of the reactors since leachate and condensate were moving downward through the compost. Reactor D had a continuous airflow, and therefore moist air was removed from the reactor continuously. This explained why it had a lower variance than Reactors A and B. Reactor B, with limited air recirculation, had a lower variance than Reactor A, Reactor C, with full-time air recirculation, had the lowest variance. Table V.l5 shows the degree of C degradation in the different reactors. The ranking was Reactor C (highest), Reactor B, Reactor A and Reactor D. This coincided with the degree of homogeneity of temperature and moisture content in the different reactors. In conclusion, internal recirculation of air led to a more homogeneous temperature and moisture content distribution throughout the compost mass as compared to no recirculation, resulting in improved organic matter degradation. 133 Table V.15 Loss of carbon over composting period as calculated via mass balance (without air recirculation vs with air recirculation) Reactor Day % M C M M s % C M , A M , A M C / M C A (Fresh air aeration & Fresh air cooling) 1 71.75 ±0 .86 45.0 32.29 12.71 33.58' ±0 .78 4.27 8 76.06 ±1 .32 40.0 30.42 9.58 35.23 ±1.53 3.37 -0.90 -21% B (Fresh air aeration & air-recir. cooling) 1 71.75 ±0 .86 45.0 32.29 12.71 33.58 ±0.78 4.27 8 77.10 ±1.03 36.0 27.76 8.24 37.42 ±1.37 3.08 -1.18 -28% C (Fresh air + recirculation aeration 1 71.75 ±0 .86 45.0 32.29 12.71 33.58 ±0.78 4.27 & air-recirculation cooling 8 79.13 ±1.11 37.5 29.67 7.83 36.89 ±1 .19 2.89 -1.38 -32% D (Fixed aeration) 1 71.75 ±0.86 45.0 32.29 12.71 33.58 ±0 .78 4.27 8 77.22 ±0 .97 37.0 28.57 8.43 42.57 ±1 .34 3.59 -0.68 -16% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. Conclusions from Part 1 of Lab test results Bacterial inoculation with manure is not necessary to start up the composting process (Lab 1). Not having to use manure as an inoculum resolves a lot of potential hygienic and transportation problems for greenhouse operators. With good control of the aeration algorithm and heat loss, composting of greenhouse wastes could satisfy' the requirement for a Process to Further Reduce Pathogens (PFRP - 55 °C for 3 days) and the requirement for inactivation of Salmonella (45 °C for 3 days). 134 Ammonia concentration could be over 300 ppm and might cause a serious odor problem. Ammonia emissions could be significantly reduced by air-recirculation which had a lower overall air outflow rate. Also, it was proven that the ammonia could be removed by a downstream biofilter with compost as the medium. Less leachate and condensate were found in the reactors with air recirculation control. This could be a big advantage for greenhouse operators using on-site composting with the in-vessel system since both leachate and condensate were high in ammonia concentration and were required to be treated before land application. The parameter of air recirculation control was tested in Lab 2, 3 and 4. Even though identical replicates were not done due to the variation in greenhouse waste streams at different times and the unpredictability of the external climate (see V.3.1 Methodology section), recirculation control was found to be a more effective method for maintaining process temperature below a set point than any kind of temperature feedback control in all the three lab runs. Biological agents typically do not survive the high temperature phase of the composting process (Bollen, 1993; Hoitink, 1999) and high temperature (e.g. above 60 °C) might hamper the microbial activity (Tuomela, 2000). Morever, in the last three lab runs, systems with air recirculation for cooling and aeration showed higher degradation of C levels, and in Lab 4 particularly, showed more consistent temperatures and moisture contents within the compost mass (upper and lower section), which would plausibly maintain a more consistent compost quality throughout different sections of the compost mass (Tiquia, 2000) , e.g. degree of degradation, stability, achieving PFRP, etc. 135 Based on higher degradation of C levels (Lab 2, 3 and 4), effective control of maximum set point temperature (Lab 3, and 4), less ammonia emission (Lab 3), less leachate and condensate amount (Lab 4), less vertical variation of temperatures and moisture contents (Lab 4), the incorporation of air recirculation into the fresh air aeration was shown to be a better aeration strategy. Even though the lab-scale air-recirculated reactor has a lot of advantages, it would require a lot more capital investment (e.g. air pumps, pipings, condensers, insulation, etc) and operating costs, i.e. electricity. For budget reasons, air recirculation devices were not installed in all the experimental setup for further lab tests nor in the pilot scale reactor. In Part 2 of the lab scale tests, the substrate recipe (C:N ratio, moisture content, porosity, etc) will be the focus of investigation. 136 TESTING OF EFFECTS OF VARIOUS SUBSTRATES V.4.1.5Lab 5 Test After trying different reactor configurations and aeration strategies, the next target was to find out the best substrate recipes for composting greenhouse waste, especially the prunings that were generated everyday. How to maximize the usage of an existing plant waste stream and to minimize the addition of foreign bulking agents and carbon sources (since greenhouse wastes proved to have relatively low C:N ratios) were considered. At the same time, the effects of differences between bulking agents on the process and on the final compost quality were investigated. Amendments have been used to condition wet substrates and can serve two purposes. First, they can be added to reduce bulk weight and increase void volume allowing for proper aeration. Second, they can be added to increase the quantity of biodegradable organics in the mixture and, thereby, increase the energy content of the mixture (Haug, 1993). A time series study of Carbon, Nitrogen, Moisture Content, Ammonia, Nitrate, Phosphate and pH was conducted during the composting process. After an extensive search in local gardening and landscaping markets, two bulking agents were found to be most common and readily available in B.C.: alder bark (alder hog fuel) and hemlock bark mulch. So, these two materials were used in the following lab and pilot scale tests as bulking agents. For cost and practicality purposes, the necessity of shredding the prunings before composting was investigated as well. That posed a big concern to the greenhouse operator because shredding 10 acres of daily greenhouse prunings involves a lot of time and labour. In order to minimize errors and costs, all four 137 reactors used linear temperature feedback for aeration control (intermittent) and air flows were set at 7 L/min with a 100% cycle. The compost formulations for the various reactors used in Lab 5 are shown in Table V.l6. Figure V.l3 plots the temperature profiles. Compost samples (50 g) from each reactor were taken every day for the tracking study. Bio-degradable twines samples were tested in Lab 5 for the rate of bio-degradability (discussed in Chapter rV). All the track study results are shown in Figures V.l4 to V.21. Table V.l6 Lab 5 Compost Feedstock Recipes Batch Materials Wt(lb) Wt% Reactor A Non-shredded prunings 22 47.6% Non-shredded with Hemlock Hemlock bark 17.6 38.1% Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor B Shredded prunings 22 47.6% Shredded with Hemlock Hemlock Bark 17.6 38.1% Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor C Non-shredded prunings 22 47.6% Non-shredded with Alder bark 17.6 38.1% fresh alder Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor D Shredded prunings 22 47.6% Shredded with fresh alder Alder bark 17.6 38.1% Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% 138 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Time (d) Ambient Temp=20.3±1.7 °C (avg±SD) Figure V.13 Temperature Profiles of Lab 5 Both reactors with shredded wastes (reactors B and D) heated up faster (shorter TIMEmax) and reached TEMP m a x (57 °C and 60 °C respectively) at Day 1, while both reactors with non-shredded wastes (reactors A and C) did not reach T E M P m a x (51 °C and 57 °C respectively) until Day 4 (see Figure V.13). This indicated that microbes could degrade carbon faster when the wastes were shredded to increase the surface area available for reaction. None of the reactors passed the PFRP requirement (Duration55°c >3 days). However Duration^ for reactors A, B, C, D were 4.0, 5.3, 24.0 and 18.2 days, which passed the Salmonella pasteurization requirement. All the reactors were opened for re-mixing on Day 12, and since reactor A and B (with hemlock) did not re-heat, they were emptied on Day 14. During the mixing, it was found that the thermocouple in reactor D was not totally submerged into the compost mass; so, another thermocouple was used and 139 re-positioned properly. Both reactors with alder bark as bulking agent had a prolonged mesophilic stage; reactor C, with non-shredded wastes did not finish until Day 34 while reactor D, with shredded wastes, finished at Day 28. Composting with alder bark gave better temperature performance than composting with hemlock bark. In Figure V.17 it can be seen that the moisture content in the reactors oscillated around somewhat but the trend was generally down as composting continued. Initially all 4 reactors had a moisture content of 70%. Reactors A and B, which as noted above, were emptied at day 14 had final moisture contents of about 68%, thus not much moisture loss ocurrred. Reactor C's final moisture content was 60% and reactor D's was 57%. Final moisture contents are summarized in Table V.18. Figure V.14 plots % total carbon vs. composting time. There are no significant differences among the 4 reactor treatments. The final C losses were computed and are summarized in Table V.18. Again no significant differences were observed. Thus unshredded wastes performed as well as shredded wastes in terms of degradation of C containing compounds. Figure V.15 is a plot of % total nitrogen vs composting time. Table V.19 compares the initial and final amounts of N . Figure V.15 shows an overall increase in total N over the course of the time of composting with some oscillations along the way. In the beginning of the composting process N was released in the form of ammonium-nitrate and some of it was taken up by the compost as can be seen in Figure V.18. The concentration of nitrate in the compost also built up as seen in Figure V.19. Thus near the start of the composting process proteins began to decompose early in the thermophilic 140 phase. Some of the resulting NH3 can be oxidized to NO3 in a nitrification process mediated by microorganisms. Figure V.l6 contains a plot of the C:N ratio over the duration of the composting period. Initially values of this ratio decreased for all 4 reactors. From Figure V.l5 and Figure V.l8, there were initial increases in total N and NH3, and from Figure V.l4, there was an initial decrease in total C. Therefore, overall there was an initial decrease in the C:N ratio. The C:N ratio of all 4 reactors had dropped to between 25 to 30 (from 30 to 35) at the end. Figure V.20 shows the data for P concentration. Figure V.21 plots pH vs composting time. At the end of the active composting time (14 days for reactors A and B and 28-33 days for reactors D and C) the pH had dropped a little from its initial value but was always within 0.5 pH units of neutrality (pH 7.0). Lower levels of pH near the end of composting probably occurred because organic acids were generated during the process. For reactors C and D phosphate levels tended to go down during composting. For reactors A and B they more or less remained the same although there were oscillations. Possibly the PO4 concentration was associated with PO4 solubility as a function of pH. 60.00 50.00 40.00 30.00 20.00 10.00 0.00 Total Carbon (dryba»|«) | H H H f l ~*~ Reactor A H aPJ^l ~*— Reactor B 5f Reactor Cl_—__ | —*— Reactor D - — Date Figure V.14 Total Carbon Vs Time Figure V.15 Total Nitrogen vs Time 141 Figure V.16 C:N Ratio vs Time Figure V.17 MC (wet basis) vs Time Nitrate N03 (mg/L) - Reactor A - Reactor B Reactor C - Reactor D * ^ T ^ - X - » W i , X , * l > < | * , . 10 15 20 Date 30 Figure V.18 Ammonia vs Time Figure V.19 Nitrate vs Time 180.00 160.00 140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00 P h o s p h a t e P 0 4 Reactor A | Reactor B 1 Reactor C _ Reactor D 1 (mg/L) k 15 Date 20 25 8.0 7.6 7 0 p H 6.5 6.0 5 5 5.0 p H 10 15 20 Date - Reactor A - Reactor B Reactor C - Reactor D 25 30 Figure V.20 Phosphate vs Time Figure V.21 pH vs Time 142 Table V.17 provides data on the amounts of condensate and leachate collected from the 4 reactors over the total active composting period. Comparing the amount of leachate from reactor A (non shredded) with that of reactor B showed that shredding definitely resulted in more leachate. No doubt this occurred because the shredding process damaged the plant structure allowing its liquid contents to more readily leak out. This can also be seen by comparing reactors C and D. The amount of leachate produced did not seem to depend on whether the bulking agent was alder or hemlock bark. Comparing the amounts of condensate generated between reactors A and B and reactors C and D again showed that the shredding process resulted in more condensate. This probably occurred because the shredding process released more plant liquids so more of it was evaporated leading to condensate. The greater amounts of condensate collected from reactors C and D merely reflect that they ran at a high temperature for more or less twice as long as reactors A and B. From the wastewater standpoint, the non-shredded wastes were preferable since less wastewater (leachate and condensate) would need to be treated. Table V.17 Cumulative amounts of Leachate and Condensate from Reactors Reactor A Hemlock/No.n-Shredded Reactor B Hemlock/ Shredded Reactor C A Ider/Non-sh redded Reactor D Alder/ Shredded Leachate (ml) 45 180 75 140 Condensate (ml) 410 830 1800 2710 143 Table V.18 Loss of carbon over composting period as calculated via mass balance Read or Bail M \1„ % C WI, \ M K M , A (Non-shredded with hemlock) 1 71.07 ±0.81 46.2 32.83 13.37 39.51 ±1.15 5.28 14 67.95 ±1.01 32.6 22.15 10.45 40.32 ±2.01 4.21 -1.07 -20% B (Shredded with hemlock) 1 70.77 ±0.94 46.2 32.70 13.50 40.31 ±1.41 5.44 14 66.76 ±1.32 33.0 22.03 10.97 39.98 ±0.99 4.39 -1.06 -19% C (Non-shredded with alder) 1 70.19 ±0.73 46.2 32.43 13.77 42.75 ±1.16 5.89 34 60.72 ±0.69 28.6 17.37 11.23 40.93 ±0.87 4.60 -1.29 -22% D (Shredded with alder) 1 68.95 ±1.26 46.2 31.85 14.35 42.61 ±1.18 6.11 28 57.80 ±0.95 26.4 15.26 11.14 41.58 ±1.20 4.63 -1.48 -24% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. Table V.18 shows the loss of carbon for each reactor. Even though the total carbon concentration did not decrease (see Figure V.l6), the total mass of carbon in the compost mass was reduced about 20% because of an overall weight loss (mass balance). Unshredded wastes seemed to be as good as shredded wastes in terms of degradation. The reactors using alder bark seemed to have a more complete bio-conversion than the ones using hemlock bark, considering the reaction temperature and the loss of carbon. The longer period at high temperature and a higher degree of degradation suggested a higher content of biodegradable organic matter in the alder bark. This is consistent with a typically higher biodegradable fraction in hardwood bark (alder) compared to softwood bark (hemlock) (Haug, 1993). 144 Table V.l9 Loss of Nitrogen over composting period as calculated via mass balance Reactor Pais "„MC VI M« IV1, %N M„ \M„ \M„/M„ A (Non-shredded with hemlock) 1 71.07 ±0.81 46.2 32.83 13.37 1.31 ±0 .18 0.18 14 67.95 ±1.01 32.6 22.15 10.45 1.53 ±0.21 0.16 -0.015 -9% B (Shredded with hemlock) 1 70.77 ±0 .94 46.2 32.70 13.50 1.19 ±0 .09 0.16 14 66.76 ±1 .32 33.0 22.03 10.97 1.36 ±0 .15 0.15 -0.012 -7% C (Non-shredded with alder) 1 70.19 ±0.73 46.2 32.43 13.77 1.31 ±0.11 0.18 34 60.72 ±0 .69 28.6 17.37 11.23 1.65 ±0 .14 0.19 0.005 3% D (Shredded with alder) 1 68.95 ±1 .26 46.2 31.85 14.35 1.23 ±0 .08 0.18 28 57.80 ±0.95 26.4 15.26 11.14 1.67 ±0.13 0.19 0.010 5% Notations : MC=Moisture Content; M=Compost Mass; Mw=Water Mass; M5=Solid Mass ; %N=Mean Percentage of Nitrogen ± Standard Deviation; M„=Nitrogen Mass; AM„=Change in Nitrogen Mass. The actual availability and the release potential of nitrogen are important factors in determining plant growth rate in the presence of compost. During aerobic composting, some of the initial nitrogen content in the feedstocks can be lost. Nitrogen loss can be reduced by using higher C:N mixtures that enhance N immobilization or by lowering the pH of the compost to increase ammonium solubility during the thermophilic stage (Raviv, 2002). Sawdust and bark were found to be efficient bulking agents to reduce the loss of nitrogen due to volatilization of ammonia during composting (Morisaki, 1989). Table V.l9 shows the nitrogen balance in each reactor. The reactors with alder bark as bulking agent showed a net increase in total nitrogen mass, compared to losses in the ones with hemlock bark. The reason for a net gain in nitrogen could be due to 145 nitrogen fixation, though much of the fixed nitrogen would have to be converted to organic form, i.e. microbial biomass, since the higher ammonium- and nitrate-nitrogen could only account for a portion of the increase. Since the experiments with alder bark contained more readily degradable carbon, as shown in Table V.18, it helped to immobilize nitrogen. One the other hand, the hemlock bark, containing less readily degradable carbon, allowed more nitrogen to be lost (Hansen, 1989). The relatively low pH (below 7) in the thermophilic stage, as shown in Figure V.21, would also help in conserving nitrogen by reducing ammonia volatilization. V.4.1.6Lab 6 Test Lab 6 was an attempted repeat of Lab 5 in order to re-confirm that alder bark would be a better choice than hemlock bark as the bulking agent. In order to minimize any experimental errors caused by variation in reactors, the alder and hemlock barks were put in different reactor orders this time (see Table V.20). The porosities of shredded and non-shredded wastes were also compared. According to the BC Provincial composting guidelines, bulk densities of the mixture less than 550 to 640 kg/m3 are usually adequate (Ministry of Agriculture, Fisheries, and Food, 1991). This was only achieved when the substrates were not shredded (reactor A and B) (see Table V.20). The oxygen profile was monitored in one of the reactors (reactor A, in this case) in order to see the oxygen consumption at different stages of the process (see Figure V.23). 146 During this testing period, it was extremely cold inside the warehouse of the greenhouse since the greenhouse production had ceased during that period, and thus there were substantial heat losses from the reactors. Therefore the reactors, in spite of being insulated, were unable to reach the desirable high temperature as they did in the Lab 5 tests (see Figure V.22). Table V.20 lists the feedstock recipes used in this set of experiments as well as the measurements of initial bulk density and porosity. From the data in Table V.20, it can be noted that shredding the prunings resulted in higher bulk density compared with not shredding them. Compare reactor A with reactor C and reactor B with reactor D. As would be expected from the bulk density observations, shredding (reactors B and D) resulted in lower porosities when compared to the non-shredded reactors (A and C). Use of alder bark as bulking agent produced greater bulk density than hemlock bark. Compare reactors A and B, and C and D. With the non-shredded prunings hemlock bark gave a higher porosity (reactor B vs reactor A), but with the shredded prunings alder bark resulted in a slightly higher porosity. 147 Table V.20 Lab 6 Compost Feedstock Recipes Batch Materials Wt (lb) Wt% Bulk Density (kg/m3) Porosity (%) Reactor A Non-shredded Non-shredded prunings 22 47.6% 5 8 8 ± 2 45 .6±2 .3 with alder Alder bark 17.6 38.1% 4(3) %(3) Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor B Non-shredded Non-shredded prunings 22 47.6% 4 1 9 ± 2 5 7 ± 2 . 1 % with hemlock Hemlock bark mulch 17.6 38.1% 1(3) (3) Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor C Shredded prunings 22 47.6% 7 7 8 ± 3 34 .7±1 .8 Shredded with alder Alder bark 17.6 38.1% 5(3) % (3) Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor D Shredded prunings 22 47.6% 7 0 4 ± 2 30 .4±1.5 Shredded with hemlock Hemlock bark mulch 17.6 38.1% 9(3) % Used sawdust 4.4 9.5% (3) Recycled compost (Pilot 2) 2.2 4.7% Figure V.22 Temperature Profiles of Lab 6 148 Both reactors with hemlock bark as bulking agent (reactors B and D) heated up faster than those with alder (A and C) (See Figure V.22) (shorter T I M E m a x ) but only reached T E M P m a x of 35 °C and 31 °C for the non-shredded and shredded wastes respectively. On the other hand, in the other two reactors, with alder bark as bulking agent, non-shredded waste (reactors A and C) heated up more slowly but reached a higher TEMPmax (47 °C and 49 °C respectively) at Day 11 and Day 12. None of the reactors passed the PFRP requirement (Duration55°c >3 days) nor the Salmonella pasteurization requirement (Duration45°c >3 days), due to the heat losses during the process. In reactor A, the oxygen level (Figure V.23) was at about 21 % over the beginning 8 days, when there was no apparent heating up of the process. It then began to deplete when the process temperature went up from Day 8 to Day 14. Oxygen consumption was higher 149 during the high temperature period of the process. However, in that high rate of degradation period, the oxygen level still maintained above 14% throughout the process which is within the ideal range for optimal aerobic degradation (VanderGheynst, 1997; Bari, 2001). Al l the reactors were emptied on Day 18 since no re-heating had occurred after mixing. After the active composting process, the substrates were cured in a covered and ventilated shed. It was interesting to discover that seedlings were found in the curing piles from reactors B and D, both of which did not achieve high enough temperatures to kill seeds. This showed that when the composting temperature was not high enough, seeds in the greenhouse wastes would not be killed and would germinate during curing. This would not be desirable for any compost users. Therefore this would be an important factor to consider in deciding whether the process was able to make good quality compost. The results of Lab 6, with the same substrate recipe as Lab 5, are consistent with those presented in Lab 5. First of all, they showed that shredding was not necessary since the reaction times and temperatures were about the same. Secondly, it showed that with alder bark as the bulking agent, the microbial activities were higher (higher temperature) and led to a higher degradations of organic C, as shown in Table V.21. : 26 % and 29 % with alder bark versus 9 % and 10 % with hemlock bark, even though this lab run was conducted in a extremely cold environment (average 5°C ambient temperature). 150 Table V.21 Loss of carbon over composting period as calculated via mass balance Reactor Day % M C M M w M s % C M , A M , A M C / M , . A (Non-shredded with alder) - 1 70.05 ±1.38 46.2 32.36 13.84 42.59 ±1 .86 5.89 18 76.06 ±1 .57 39.5 30.04 9.46 46.10 ±1.45 4.36 -1.53 -26% B (Non-shredded with hemlock) 1 74.30 ±1 .12 46.2 34.33 11.87 38.92 ±0.93 4.62 18 77.10 ±0.98 38.4 29.61 8.79 47.73 ±0 .86 4.20 -0.42 -9% C (Shredded with alder) 1 70.05 ±1 .54 46.2 32.36 13.84 41.32 ±1.15 5.72 18 79.13 ±1.63 41.2 32.60 8.60 47.21 ±1 .36 4.06 -1.66 -29% D (Shredded with hemlock) 1 74.30 ±1 .29 46.2 34.33 11.87 39.14 ±1.28 4.65 18 77.22 ±1.41 38.1 29.44 8.68 48.22 ±1.13 4.19 -0.46 -10% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. V.4.1.7Lab 7 Test In Lab 5 and Lab 6, it was determined that alder bark was a better choice of bulking agent than hemlock bark. There are 2 kinds of alder bark available on the market: fresh alder bark and composted alder bark with the latter being more expensive. In this lab run, these two different alder barks were tested using the same mixing ratios as in the previous lab runs. Table V.22 shows the mixing ratios for reactors A to D. 151 Table V.22 Lab 7 Compost Feedstock Recipes Hatch Materials VVt(ll)) Wt% Reactor A Shredded prunings 22 47.6% shredded with Fresh alder bark 17.6 38.1% fresh alder Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor B Non-shredded prunings 22 47.6% non-shredded Fresh alder bark 17.6 38.1% with fresh alder Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor C Shredded prunings 22 47.6% shredded with Composted alder bark 17.6 38.1% composted alder Used sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% Reactor D Non-shredded prunings 22 47.6% Non-shredded with composted alder Composted alder bark 17.6 38.1% Used Sawdust 4.4 9.5% Recycled compost (Pilot 2) 2.2 4.7% A few comparisons could be achieved using the above matrix : Reactor A vs Reactor B and Reactor C vs Reactor D - Unshredded vs Shredded waste. Reactor A vs Reactor C and Reactor B vs Reactor D - Fresh Alder vs Composted Alder Bark as bulking agents Ambient Temp=16.7±3.1 °C (avg+SD) 8 , a ' Figure V.24 Temperature Profile of Lab 7 152 V , - r 1 1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 T I M E ( d ) Figure V.25 Lab 7 Reactor A Oxygen Concentration vs. Time Both reactors with shredded wastes began to heat up before those containing non-shredded waste (see Figure V.24) (shorter TIMEm a x), similar to the results observed in Labs 5 and 6. The TEMP m a x of reactors A, B, C, and D were 63 °C, 62°C, 68 °C and 65 °C respectively. The Duration55°c of reactors A, B, C and D were 2.2, 3.0, 3.4 and 3.6 days respectively. So all of them, except reactor A, passed the PFRP requirement (Duration55<>c >3 days). Duration45°c of reactors A, B, C and D were 5.2, 18.6, 6.4 and 14.1 days respectively and therefore all of them passed the Salmonella pasteurization requirement (Duration^c >3 days). All the reactors had similar temperature profiles in the thermophilic stage but the ones with non-shredded wastes had longer mesophilic periods. The active composting finished at day 14 for reactors A and C, at day 21 for reactor D and at day 28 for reactor B, i.e. the non-shredded wastes with fresh alder bark had the longest reaction times. All the reactors over-shot their set point temperatures (60 153 C) in the beginning of the thermophilic stage and were able to be cooled down to 60 C or below shortly after (within 24 hours) by applying full cycle aeration from the linear temperature feedback control. During those short periods when the temperature had over-shot the set point, the oxygen content went below 15% (between 10 to 15%). It went back up to above 15 % when the temperature was successfully kept below the set point as shown in Figure V.21. This showed that it was important to have a control strategy capable of supplying enough oxygen, as well as cooling in the thermophilic stage. In this case the linear temperature feedback control was again proven to be effective. The purpose of this experiment was to find out whether using composted alder basrk would enhance the process and whether shredding was necessary before composting the greenhouse prunings. Table V.23 shows the loss of carbon in each reactor after the composting process. The substrates with fresh alder bark had a slightly higher degree of degradation of C than the ones with composted alder bark (36% vs 33% for shredded, and 38% vs 35% for non-shredded waste). This could have been due to the lower initial biodegradable carbon content in the composted alder bark since it had already gone through a degradation process. The process times with the shredded prunings were the same for both when using fresh alder or composted alder (14 days) bark. The process times of the non-shredded prunings were slightly shorter for the composted alder than for the fresh alder (24 days vs 28 days). So, the process time did not appear to be much affected by whether fresh alder or composted alder bark was used. But it was affected by whether the prunings were shredded or not. Process time would be a less important issue than cost for greenhouse 154 operators. Therefore, fresh alder bark (less expensive) as bulking agent and non-shredded prunings (less material handling cost) would be preferable. Table V.23 Loss of carbon over composting period as calculated via mass balance Reactor Day % M C M M , M s % C A M , A M , / M C A (shredded with fresh alder) 1 70.89 +1.43 46.2 32.75 13.45 40.58 ±0 .82 5.46 14 71.75 ±1.13 28.0 20.09 7.91 44.13 ±0.85 3.49 -1.97 -36% B (Non-shredded with fresh alder) 1 71.32 ±0 .94 46.2 32.95 13.25 41.32 ±0 .96 5.47 28 70.51 ±1.41 27.0 19.04 7.96 42.59 ±1 .17 3.39 -2.08 -38% C (Shredded with composted alder) 1 72.35 ±1 .16 46.2 33.43 12.77 38.65 ±0 .57 4.94 14 72.90 ±1 .12 28.0 20.41 7.59 43.38 ±0 .74 3.29 -1.65 -33% D (Non-Shredded with composted alder) 1 72.98 ±1 .29 46.2 33.72 12.48 39.14 ±0 .68 4.89 24 70.27 ±1 .38 24.5 17.22 7.28 43.76 ±0.83 3.19 -1.70 -35% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. V.4.1.8Lab 8 Test In this test, since fresh alder bark and no shredding had been decided upon from previous tests, different amounts of fresh alder bark were used in order to find out the best mixing ratio. Table V.24 provides the feedstock makeup and Figure V.26 provides the temperature profiles in the 4 reactors. 155 Table V.24 Lab 8 Compost Feedstock Recipes Batch Materials VVt(lb) VVt% Reactor A Non-shredded prunings 22 100% Reactor B Non-shredded prunings 18.7 85% Fresh alder bark 3.3 15% Reactor C Non-shredded prunings 15.4 70% Fresh alder bark 6.6 30% Reactor D Non-shredded prunings 12.1 55% Fresh alder bark 9.9 45% o d 40 1 1 REACTORA REACTOR B REACTOR C REACTOR D few /> i !/ \ 0 2 4 6 8 10 12 14 16 Ambient Temp=7.8±4.2 °C (avg±SD) T | m e ( d j Figure V.26 Temperature Profile of Lab 8 All the reactors reached the TEMP m a x by about day 3 or 4. The T E M P m a x of reactors A, B, C, and D were 41.8 °C, 5 3 . 2 ° C , 5 3 . 6 °C and 5 0 . 9 °C respectively. None of reactors passed the PFRP requirement (Duration5 °c > 3 days), which was probably due to the extremely cold ambient temperature. Even though all the reactors were well-insulated, the fresh air supplied to the reactors cooled down the substrates significantly. 156 Duration45°c of reactors A, B, C and D were 0, 2.4, 3.5 and 4.1 days respectively and therefore only reactors C and D passed the Salmonella pasteurization requirement (Duratiori45°c >3 days). Reactor A, with no bulking agent, took longer to get started then the other 3 reactors, displayed lower temperatures throughout the active composting period and finished the active composting phase the soonest. This implies that addition of a bulking agent was beneficial as far as temperature profile was concerned. Reactor D, with 45% bulking agent, maintained a relatively high temperature over the longest time compared to the other reactors. Reactor C, with 30% bulking agent was next, followed by reactor B, with 15%> bulking agent. Table V.25 shows the amounts of leachate and condensate collected from each reactor. The amount of leachate seemed to be related to the initial moisture content of the substrate - the higher the initial moisture content, the higher the leachate amount. The initial moisture content was dependent on the amount of bulking agent used since the bulking agent moisture content was lower than that of the prunings (60.39% vs 87.88%). So, it was reasonable to conclude that the more bulking agent used (alder bark), the less leachate would be generated. The amount of condensate tended to correlate with the duration at high temperature of the process. It was highest with the highest sustained temperature reactor (reactor D) i.e. the higher the temperature and longer it was sustained, the more condensate that was generated. 157 Table V.25 Leachate and Condensate from Reactors Reactor A 100 % Reactor 15 85% / 15"/,. Reactor ( 71)-% / 301% Reactor 1) 55% / 45% Leachate (ml) 3000 1950 1550 1100 Initial M C 87.88±0 .94 87 .20±1.25 82 .38±1.05 75 .51±0 .92 Final M C 85.56±2 .24 84 .03±1.20 73 .86±0.93 72 .77±1 .12 Condensate (ml) 210 400 400 870 Duration45oC (D) 0 2.4 3.5 4.1 Table V.26 Loss of carbon over composting period as calculated via mass balance Reactor l)a> % M C M [*§„ m P \ M , \ M , / M , A (100 % non-shredded pruning) 1 87.88 ±0.97 22.0 19.33 2.67 30.33 ±0 .56 0.81 14 85.56 ±1.23 10.6 9.04 1.52 31.18 ±0 .85 0.48 -0.33 -41% B (85% non-shredded pruning/ 15%fresh alder) 1 87.20 ±0.85 22.0 19.18 2.82 36.65 ±1 .16 1.03 14 84.03 ±0 .59 14.5 12.20 2.32 31.51 ±1.11 0.73 -0.30 -29% C (70% non-shredded pruning/ 30%fresh alder) 1 82.38 ±0 .68 22.0 18.12 3.88 41.37 ±1 .74 1.60 14 78.89 ±0.83 15.0 11.80 3.16 36.73 ±0 .84 1.16 -0.44 -28% D (55% non-shredded pruning/ 45%fresh alder) 1 75.51 ±0 .48 22.0 16.61 5.39 42.22 ±0.93 2.27 14 72.77 ±1.13 16.7 12.17 4.55 36.51 ±0 .79 1.66 -0.61 -27% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. Table V.26 shows the carbon loss from each reactor with the different bulking agent mixing ratios. The one with zero bulking agent, reactor A, demonstrated the highest carbon loss (41%), while the rest of other reactors had more or less the same loss at 27 to 29%. Reactor A had largest loss of carbon, likely because the prunings have greater amount of available carbon as compared to alder having more lignocellulose content. 158 Even though reactor A had the highest carbon loss over the active composting period, it had the highest phyto toxicity, i.e. lowest germination rate and root index compared to the others (see Figure VI.5 from Chapter VI). Therefore, the percentage of carbon loss might not always be a good indicator of compost maturity and quality. Moreover, reactor A had a much higher nitrogen loss (52%) than did reactors B, C and D (18%, 18%, 12% respectively) (see Table V.27). Nitrogen loss occurring during composting is a key issue and nitrogen volatilization reduces the fertilizer value of the finished compost and thus constitutes an economic loss. Barrington et al. (2002) measured the effect of carbon source on nitrogen losses by volatilization during composting and found that the moisture content of the bulking agents and aeration regime had no consistent significant effect on N and C losses by volatilization; only the type and amount of bulking agent had a significant effect. Martin et al.(1993) investigated the effects of peat and sawdust, employed as bulking agents, in composting and found that mixing enough bulking agent to absorb the moisture of substrate was an important element in controlling nitrogen losses during composting. This agrees with the results found in this lab study. 159 Table V.27 Loss of Nitrogen over composting period as calculated via mass balance Reactor Day % M C rvi M„. M s %N M„ AM,, AM„/M„' : A (100 % non-shredded pruning) 1 87.88 ±0 .97 22.0 19.33 2.67 3.89 ±0.18 0.10 14 85.56 ±1.23 10.6 9.04 1.52 3.25 ±0.07 0.05 -0.05 -52% B (85% non-shredded pruning/ 15%fresh alder) 1 87.20 ±0.85 22.0 19.18 2.82 2.33 ±0 .16 0.07 14 84.03 ±0 .59 14.5 12.20 2.32 2.31' ±0.21 0.05 -0.01 -18% C (70% non-shredded pruning/ 30%fresh alder) 1 82.38 ±0.68 22.0 18.12 3.88 2.01 ±0.13 0.08 14 78.89 ±0.83 15.0 11.80 3.16 2.17 ±0 .09 0.07 -0.01 -12% D (55% non-shredded pruning/ 45%fresh alder) 1 75.51 ±0 .48 22.0 16.61 5.39 1.77 ±0.06 0.10 14 72.77 ±1.13 16.7 12.17 4.55 1.85 ±0.15 0.08 -0.01 -12% Notations : MC=Moisture Content; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %N=Mean Percentage of Nitrogen ± Standard Deviation; M„=Nitrogen Mass; AM„=Change in Nitrogen Mass. V.4.1.9Lab 9 Test At the end of a crop year (December), all the vines, leaves and the sawdust growing medium (yellow cedar) are removed from the greenhouses. Since the vines could be as long as 30 to 40 feet, they had to be shredded before putting them into the composter. Moreover, 75 tonnes per hectare of sawdust had to be treated as well. The sawdust could be a good carbon source to supplement the relatively low C:N ratio of the vines; however, it would not be as a good a bulking agent as fresh alder bark since it lacked the necessary structure and porosity. It would be beneficial for greenhouse operators to know how much the sawdust could be utilized without affecting the compost 160 process and compost quality. In this test, different ratios of alder bark and sawdust medium were used in order to find out the best mixing ratio. Table V.28 lists the components and their proportions in the compost mix for this test. Table V.28 Lab 9 Compost Feedstock Recipes Hatch Materials Wt(lb) Wt% Reactor A Shredded year-end vines 26.4 60% Alder bark 0 0% Used sawdust 17.6 40% Reactor B Shredded year-end vines 26.4 60% Alder bark 6.16 14% Used sawdust 11.44 26% Reactor C Shredded year-end vines 27.28 62% Alder bark 11.0 25% Used sawdust 5.72 13% Reactor D Shredded year-end vines 28.6 65% Alder bark 15.4 35% Used sawdust 0 0% 80 70 60 Ambient Temp=3.8±3.3 °C (avg+SD) Time (d) Figure V.27 Temperature Profiles of Lab 9 161 The reactors started to heat up at different times, see Figure V.27. The TIMEmax of reactors A, B, C, D were 5, 4.8, 4.3 and 3.6 days respectively. The T E M P m a x of reactors A, B, C, and D were 42.6 °C, 52.7°C, 55.2 °C and 55.6 °C respectively. None of reactors passed the PFRP requirement (Duration5 °c >3 days), which was probably due to the extremely cold ambient air temperature (0-5 °C) at the time. Duratiori45°c of reactors A, B, C and D were 0, 7.1, 21.6 and 29.2 days respectively and therefore reactors B, C and D passed the Salmonella pasteurization requirement (Duratiori45°c >3 days). Reactor D (35% alder bark, no sawdust) heated up most quickly and maintained the highest temperature longest implying that sawdust was detrimental to the composting process. Reactor C (25% alder bark, 13% sawdust) gave the next best thermal performance which was not greatly different from that of reactor D except near the end of the composting. Thus, it was shown that it was possible to tolerate some sawdust. Reactor B (14% alder bark, 26% sawdust) and reactor A (no alder bark, 40% sawdust), in that order, gave rise to worse thermal performances. As a result, the temperature profiles trends were as follows : 1. The substrate heated up faster with more fresh alder bark(TIMEmax D<C<B<A); 2. The substrate heated up to higher temperature with more fresh alder (TEMPmax D>OB>A); 3. The high temperature period sustained longer with a greater fresh alder bark ratio (Duratiori45°c D>C>B>A). 162 Table V.29 Leachate and Condensate from Reactors Reactor A 60%, 0%, 40%* Reactor 1$ 60%, 14%, 26% Reactor G 1 62%, 25%, 13% Reactor D 65%', 35%, 0% Leachate (ml) 2800 3000 1600 900 Initial M C 85 .02±1 .20 82.37±1.5 3 80 .75±0.84 77 .38±0 .97 Final M C 82.91 ±0 .94 79 .92±1.3 2 78 .42±1.15 71 .91±1.25 Condensate (ml) 660 780 2700 3720 Duration45oC (D) 0 7.1 21.6 29.20 *vines, bark, sawdust Table V.29 shows the amounts of leachate and condensate collected from each reactor. Again, similar to the results presented in Lab 8, the amount of leachate appears to be correlated with the initial moisture content of the substrate - the higher the initial moisture content, the higher the leachate amount. The initial moisture content was dependent on the amount of fresh alder bark used instead of sawdust (moisture contents of 60.40% and 82.77% respectively). So, it was reasonable to conclude that the higher the alder bark to sawdust ratio used, the less leachate would be generated. Also, the amount of condensate correlated directly with the duration of high temperature of the process. It was highest in highest sustained temperature reactor (reactor D) i.e. the higher the temperature and the longer it sustained, the more condensate generated. Table V.30 shows the carbon loss from each reactor with different sawdust to alder bark mixing ratios. Al l the reactors showed quite similar carbon losses (between 25 and 29%) and were relatively lower than for the substrates used in Lab 8. This was 163 because the year-end vines and sawdust did not contain as much readily biodegradable carbon and were not as easy to degrade as the daily prunings (leaves and cull fruit). Reactors A and B, with the greater proportions of sawdust demonstrated slightly lower carbon losses (24% and 25%), while the ones with the greater proportions of alder bark (reactors C, and D demonstrated slightly higher carbon losses (29% and 28%). Reactor D also showed the highest germination rate and root index compared to the others (see Figure VI.6 in Chapter VI). Table V.31 shows that reactors A and B had much higher nitrogen losses (44% and 34%) than reactors C and D (14% and 7%). Most of the nitrogen losses were due to NH3 volatilization. Leaching has been said to account for about one fifth of the N losses and only a little N is lost due to denitrification (Sommer, 2001). In this case, the relatively high volume of leachate, which contained a high concentration of ammonia and nitrate (see Table VI.9 from Chapter VI), from reactors A and B (2800 and 3000 ml ) than from reactors C and D (900 and 1600 ml) would probably account for part of the high N losses. Barrington et al. (2002) also recorded their highest N losses in compost substrate with shavings (sawdust) as bulking agent, compared to wheat straw, hay and oat straw. This showed that sawdust is not a good amendment to control nitrogen loss. A greenhouse operator would like to get rid of the waste sawdust by incorporating it into compost and, at the same time, produce high quality compost. With this in mind a mixing ratio of 62% vines, 13% sawdust and 25% alder bark is recommended as being a reasonable mixture to achieve good results. 164 Table V.30 Loss of carbon over composting period as calculated via mass balance Reactor l)a\ M M , %C m. \ M , \ M , / M , A 60%, 0%, 40% 1 85.02 ±1 .47 44.0 37.41 6.59 40.33 ±0 .87 2.66 30 82.91 ±1.83 31.2 25.90 5.34 38.06 ±0.65 2.03 -0.63 -24% B 60%, 14%, 26% 1 82.37 ±1 .76 44.0 36.24 7.76 43.66 +0.58 3.39 30 79.92 ±0 .96 30.8 24.62 6.18 40.89 ±0.81 2.53 -0.86 -25% C 62%, 25%, 13% 1 80.75 ±1.25 44.0 35.53 8.47 45.32 ±1.03 3.84 38 78.42 ±0.89 31.2 24.50 6.74 40.48 ±1 .08 2.73 -1.11 -29% D 65%, 35%, 0% 1 77.38 ±0 .78 44.0 34.05 9.95 48.56 ±0.91 4.83 38 71.91 ±1 .27 29.9 21.52 8.40 41.39 ±0.38 3.48 -1.35 -28% Notations : %MC = Mean Moisture Content ± standard deviation; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass; %C=Mean Percentage of Carbon ± standard deviation; Mc=Carbon Mass; AMc=Change in Carbon Mass. 1 6 5 Table V.31 Loss of Nitrogen over composting period as calculated via mass balance Reactor l)a\ % M C M M „ M , " „ \ M „ \ M „ \ M „ / M „ A 60%, 0%, 40% 1 85.02 ±1 .47 44.0 37.41 6.59 1.87 ±0 .12 0.12 30 82.91 ±1 .83 31.2 25.90 5.34 1.30 ±0.05 0.07 -0.05 -44% B 60%, 14%, 26% 1 82.37 ±1 .76 44.0 36.24 7.76 1.81 ± 0 . 1 2 0.14 30 79.92 ±0 .96 30.8 24.62 6.18 1.50 ±0 .09 0.09 -0.05 -34% C 62%, 25%, 13% 1 80.75 ±1.25 44.0 35.53 8.47 1.74 ±0.15 0.15 38 78.42 ±0 .89 31.2 24.50 6.74 1.89 ±0 .13 0.13 -0.02 -14% D 65%, 35%, 0% 1 77.38 ±0 .78 44.0 34.05 9.95 1.46 ±0 .09 0.15 38 71.91 ±1 .27 29.9 21.52 8.40 1.60 ±0 .07 0.13 -0.01 -7% Notations : MC=Moisture Content; M=Compost Mass; Mw=Water Mass; Ms=Solid Mass ; %N=Mean Percentage of Nitrogen ± Standard Deviation; M„=Nitrogen Mass; AM„=Change in Nitrogen Mass. Conclusions from Part 2 of Lab test results Following the plan of testing different substrates and bulking agents, even though there was no identical replicate of each lab run, each parameter was tested and re-confirmed in at least 2 different lab runs. In Lab 5 and Lab 6, alder bark was found to be a better choice of bulking agent than hemlock bark in terms of better substrate structure, more carbon degradation, less nitrogen loss, and higher process temperature. In Labs 5,6 and 7, shredding was shown not to be necessary before composting of prunings and non-shredded prunings also helped in minimizing the amount of leachate. A longer mesophilic temperature regime was maintained after peak temperatures had been 166 attained in the composting of non-shredded wastes. Since prunings are generated every day from the greenhouses, composting without shredding means a lot of cost savings. As shown in Lab 8 and Lab 9, the amount of leachate was correlated with the initial moisture content of the substrate and the amount of condensate was correlated with the duration of the high temperature part of the process. Fresh alder bark was found to be as good as composted bark in terms of process time and loss of carbon (extent of degradation) when used as the bulking agent in Lab 7. Morever, fresh bark is less expensive compared to composted bark. Lab 8 was designed to test the minimum amount of bulking agent needed for composting daily prunings and results indicated that using 30% alder bark (by weight) as bulking agent was best for composting prunings. Lab 9 was designed to investigate the best amount of used sawdust and alder bark to blend for composting year-end wastes. For year-end wastes, shredding is necessary, and a ratio of 62% vines, 13% used sawdust and 25% alder bark was recommended based on observations made over a range of bark to sawdust to waste plant material ratios. This was in consideration of good thermal performance (via temperature profile), less leachate, greater loss in carbon and smaller loss in nitrogen. Composting without sawdust was in fact the best recipe, however, for practical purposes, greenhouse operators need to have the used sawdust composted. 167 V.4.2 PILOT SCALE COMPOSTING DATA In total, there were 8 pilot scale composting tests done in this part of the study. Each of them used from 5 to 10 tonnes of greenhouse waste. They were designed to fine-tune the in-vessel composting process until it reached the objectives of processing the greenhouse wastes produced at different times of the year. Table V.32 shows the composition of each batch in the pilot scale compost tests. Table V.32 Pilot Scale Compost Feedstock Recipes Hatch Materials Proportion In Volume PI Unshredded tomato and/or pepper prunings (leaves and vines) 75% Shredded prunings and cull fruit 17% Used sawdust 8% P2 Shredded year-end pepper vines 67% Used sawdust 33% P3 Unshredded tomato and/or pepper leaf prunings (partially rotted) 27% Cull tomatoes 31% Used sawdust medium 38% Compost recycle 4% P4 Unshredded tomato and/or pepper leaf prunings (partially rotted) 43% Used sawdust medium 14% Alder bark 43% P5 Shredded year-end cucumber vines and fruit 66% Alder bark 34% P6 Unshredded tomato and/or pepper leaf prunings 60% Used sawdust 10% Alder bark 30% P7 Shredded year-end tomato vines 87% Used sawdust 13% P8 Unshredded tomato and/or pepper leaf prunings and fruit culls 50% Used sawdust 10% Alder bark 20% Compost coarse fragment recycle 20% 168 The pilot scale tests were not designed to test or compare the effects of any parameters as was done in the lab scale tests; but instead, their purpose was to serve as a real demonstration for greenhouse operators. In some cases, they also served as a second proof of some of the lab test results. There were limitations in pilot tests as compared to lab tests, for example, weighing input wastes and output compost were not possible in the pilot tests because of the volume of the wastes and lack of proper device. Table V.33 summarizes the results obtained from the pilot scale compost tests. Table V.33 Pilot Composting Experimental Details and Results Summary U of da\ s of acti\ c composting ( on t i n ii on s niiflow Netting ("/;, of in a \ i in ti in) Intel millcnl airflow dtil\ o d e (%) 1 em p. st 11 HI 111 t 1 ccd-slock M i l . I l l ' l l l l i i l Vol. in 5 (\ i l l u m e i e d i n l i n i i i ) IM KI' A d i i o ed l.eadiatc generated (litres) PI 43 80 20 60 9 4.5 (50%) Y 1 800 P2 51 60 20 60 9 6.9 (23%) Y 200 P3 29 90 20 65 10 8.8 (12%) N 1200 P4 35 60 20 65 7 5.7 (20%) Y 2 200 P5 34 60 20 65 10 6.1 (39%) Y 210 P6 30 100 20-40 65 11 8.8 (20%) Y 2 400 P7 28 50-100 20 60 10 6.9 (31%) Y 1200 P8 i „ 38 50-100 20 60 12 9.3 (23%) Y 2 750 Based on data from a single thermocouple. 2Achieved in some but not all the thermocouple locations. Note: PFRP is Process to Further Reduce Pathogens, and requires 3 days at 55 °C minimum. 169 Figures V.28, V.29, V.30, V.31, V.34, V.35, V.37 and V.38 are the compost temperature vs. time plots for the 8 pilot scale tests. Since the compost was replaced by new material after each run, the thermocouples were not always in the same place inside the composting pile in each experiment, nor were the same number of thermocouples always used. The location of thermocouples were dependent on the dimension of waste mass inside the reactor. For example, sometimes the volume of wastes could only occupy rear half of the container, and therefore the thermocouple would be placed at the back. No replicate tests were performed because of (i) the large scale at which the experiments were carried out, and in some cases year-end crop wastes were used, rendering replication impractical; (ii) the time limit due to the long duration of the experiments (almost one month per experiment); (iii) the need to test a wide range of different substrates and operational variables and (iv) budget constraints. VanderGheynst et al. (1997) faced similar difficulties in their pilot composting studies, and concluded that as long as the composter showed similar temperature profiles with similar initial conditions, the composter was then considered able to duplicate behaviour well. Pilot 1, 6 (Fig. V.28, Fig. V.35) and Pilot 4, 8 (Fig. V.31, Fig.V.38) in this study had similar initial conditions and showed similar temperature profiles. 170 Pilot 1 - prunings/ vines /fruit/sawdust Pilot 1 Compost Temperature \ 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 Time (day) Ambient Temp=6.8±5.2 °C (avg±SD) Figure V.28 Pilot 1 Temperature Profile In this run the first mixture of feedstocks was composted in the pilot composter, using pruning waste and used sawdust growing medium. However during active composting, the material collapsed and became anaerobic at the bottom, due to a lack of physical structure and porosity in the mixture. Because of excessive wetness and odour, the composter was left open with the fan on (100% cycle) for three days at the end of the active cycle. The material was mixed with more sawdust prior to placing in a curing pile. This failure demonstrated the need for a bulking agent with this mixture of prunings, vines, culled fruit and sawdust. 171 Pilot 2 - year-end vines/sawdust 80 70 60 50 | 40 30 20 10 Pilot 2 Compost Temperature V - T C 1 TC4 ( s \ y V / f s K / \ f •••If PR \ \ • 1 1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 Ambient Temp=5.5±4.6 °C (avg±SD) Time (d) TCI - Centre depth, near back TC4 - Centre depth, near front Figure V . 2 9 Pilot 2 Temperature Profiles In the second run, rather than using tomato prunings, year-end pepper vines were used, which were available because of the crop cleanup underway at that time. The vines were shredded and mixed with used sawdust growing medium. The year end vines, which would be expected to have greater porosity, provided a compost pile with better structure since the vines would be more resistant to collapse. During the process, some uneven heating was observed. The thermocouple close to the door (TC4) showed heating sooner than TCI, which was closer to the rear of the 172 composter. This may have been due to uneven aeration or non-uniform mixing of the compost before loading it into the composter. However, the temperatures did reach PFRP and the resulting compost had a pleasant smell and medium to dark brown colour, suggesting a good composting process. Leachate production was smaller than in Pilot 1 (200 L vs 800 L). Pilot 3 - prunings/fruits/ sawdust/ compost 60 50 40 o I 30 20 10 Pilot 3 Compost Temperature - — TC — T C 1 2 -< H MftUW Il -r •• "•yaw Its \ j 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Ambient Temp=8.3±3.9 °C (avg±SD) T i m e ( d) TCI - Centre depth, near back T C 2 - Near surface, near front Figure V . 3 0 Pilot 3 Temperature Profiles In this batch, the feedstock materials were similar to those of batch 1, but incorporating more tomato fruit, and a much larger proportion of sawdust in an attempt to utilize more of the greenhouse used sawdust waste as requested by the greenhouse operator. A small amount of finished compost was also included. 173 During the active composting process, the materials again proved to be too dense and too wet, and became anaerobic. As a result, thermophilic temperatures were not reached. Large amounts of leachate and odour were also generated (1200 L). After one month, the materials were removed, remixed with alder hog fuel, and placed in a windrow. The problems encountered appear to have been due to inadequate physical structure in the mixture (too much sawdust instead of alder as bulking agent), compounded by the partially rotted leaves and a larger amount of fruit. This was quite similar to the situation in Lab 8 Reactor A, which had 100% prunings and no bulking agent, and could not reach thermophilic temperatures. Again, the lack of sufficient bulking agent was obvious. Pilot 4 - prunings/ sawdust/ alder bark o Q. 80 70 60 50 40 30 20 10 0 Pilot 4 Compost Temperature i T C1 C3 -C4 • i T -I \ ••• • "XT' " V s , 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Ambient Temp=12.1±5.1 "C (avg+SD) T l m e (d) TCI - Centre depth, near back TC3 - Centre-lower depth, near centre TC4 - Centre depth, near front Figure V.31 Pilot 4 Temperature Profiles 174 Pilot 4 Compost Oxygen Concentration ! 1 A Oxygen at TC1 — • — O x y g e n at TC2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Time (d) Figure V.32 Pilot 4 Oxygen Profiles Based on the negative experiences with batches 1 and 3, it was determined that for composting fresh prunings and fruit in the pilot scale composter, an additional bulking agent was required. For batch 4, alder bark, as recommended from Labs 5 and 6, was added in equal parts by volume to the leaf prunings. Addition of the alder bark proved successful in preventing collapse and anaerobic conditions. Thermophilic temperatures were reached, though there was still some uneven heating, with PFRP reached in some areas and not in others. There was also extensive growth of visible fungus, including mushrooms on the surface during active composting. Several oxygen measurements confirmed conditions were aerobic but values as low as 8% were noted which are less than the recommended minimum of 14% (VanderGheynst, 1997). 175 80 70 60 o 50 d E 40 o> 1- 30 20 10 0 Pilot 4 Curing Pile Temperature 0 10 20 30 40 50 60 70 80 90 100 Curing Time (d) Ambient Temp=13.8±5.2 °C (avg±SD) Figure V.33 Pilot 4 Temperature Profile of Curing For this batch, temperature in the curing pile was also logged. The results showed limited reheating to about 53 °C initially, then the temperature dropped gradually to below 40 °C after 6 weeks, and to about 36 °C after 10 weeks (see Figure V.33). 176 Pilot 5 - year-end vines/ fruit/ hemlock bark mulch Pilot 5 Compost Temperature o r 1 1 T C 1 T C 2 r r / N 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Ambient Temp=20.7±3.8 °C (avg±SD) T i m e (d) TCI - Centre depth, near back T C 2 - Centre depth, near centre Figure V.34 Pilot 5 Temperature Profiles For this batch, composting of shredded year-end cucumber vines and fruit was attempted. Some alder bark was mixed with the shredded waste to ensure porosity was adequate, based on the apparent high density of the shredded material. Thermophilic temperatures were achieved, reaching nearly 80 °C and satisfying PRFP. Higher temperatures compared to shredded pepper and tomato vine composting could have been due to more energy available from the cucumber waste, or an increased oxidation rate due to the higher porosity resulting from the added bulking agent. Small amounts of leachate were generated (210 L). The appearance of the finished compost was slightly different, in that it had a stringy texture, probably due to the fibrous cucumber vines. Again, addition of a bulking agent resulted in improved performance. 177 Pilot 6 - prunings/ sawdust/ alder bark Pilot 6 Compost Temperature 80 70 60 50 | 40 30 20 10 0 TC1 T r Q [ •TC5 W r \ r M#f \ \ 1 r t 6 8 10 12 14 16 18 20 22 24 26 28 30 A Time (d) C O N T R O L ' F A I L U R E MIXING Ambient Temp=17.8±4.3 °C (avgtSD) TCI - Near surface, near back TC3 - Centre depth, near centre TC5 - Near surface, near front Figure V.35 Pilot 6 Temperature Profiles In Pilot 6 the proportion of bark mixed with prunings was decreased to 1/2 (volume basis) to determine the minimum amount required. During active composting, some anaerobic conditions and odour developed due to inadequate porosity and physical structure. Figure V.36 plots % O2 vs time and shows that indeed near anaerobic conditions occurred early in the cycle. After 10 days the material was removed and remixed with additional bark. Following remixing, thermophilic temperatures were reached, with PFRP achieved in some parts of the material. A moderate amount of leachate was generated. 178 Pilot 6 Compost Oxygen Concentration 20 a 10 x O 5 ii 0 2 4 6 8 10 12 14 16 18 20 22 24 Time(d) Figure V.36 Pilot 6 Oxygen Profile Pilot 7 - year-end vines/ sawdust Pilot 7 took advantage of the availability of year-end waste from the pepper greenhouse at that time. The shredded vines were mixed with a small amount of used sawdust. During active composting, thermophilic temperatures and PFRP were reached. Though large amounts of leachate were generated, it appeared that aerobic conditions were maintained and odour was not a problem. Thus it was concluded that year end tomato vines with some sawdust could be acceptably composted without adding a bulking agent. 179 Pilot 7 Compost Temperature 80 70 60 50 40 30 20 10 0 -TC1 -TC3 --TC5 * \ -' 1 0 2 4 6 8 Ambient Temp=15.6±5.3 °C (avgtSD) 10 12 14 16 18 20 22 24 26 Time (d) 28i TCI - Near surface, near back T C 3 - Centre depth, near centre T C 5 - Near surface, near front Figure V.37 Pilot 7 Temperature Profiles Pilot 8 - prunings /fruit /sawdust/ alder bark / Compost Pilot 8 was conducted in mid-summer, several months after pilot 7. In this batch, leaf prunings and fruit culls were mixed with alder bark, recycled coarse compost material, and used sawdust. This was based on experiences and results from the previous 7 runs, and the recommendations from the lab scale tests. This turned out to be a very successful run. This mixture composted very well, with PFRP met in most of the material, and a final product with a pleasant smell. 180 Pilot 8 Compost Temperature 80 i 70 60 -50 o d 40 -E 30 -20 10 -0 T C1 C3 -C5 T t i t , 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 Ambient Temp=21.4±5.4 °C (avgtSD) T ' m 6 T C 1 - Near surface, near back TC2 - Near surface, near centre TC5 - Near surface, near front Figure V.38 Pilot 8 Temperature Profiles Conclusions from Pilot scale test results The pilot scale tests were not meant to be as analytical as the lab scale tests since there were not as many reactors as in the lab tests and the pilot scale tests were even more affected by external factors, e.g. larger differences in ambient temperatures, different mixings by different Bobcat operators, etc. However, several observations were made that were quite similar to those noted in the lab scale compostings and serve as re-confirmations of the lab scale findings. 179 fln Pilot 7 Compost Temperature 70 i TC1 60 - I - T C 5 50 O / Tem| ™\ on N m • M 1 u n 0 Ambient Temp=lf 2 i >.6±5.3 « 1 ( : (a.vg±i 3 I D) J 1 0 1 2 1 Tim 4 1 e(d) 6 1 8 2 0 2 2 2 4 2 6 28 TCI - Near surface, near back TC3 - Centre depth, near centre TC5 - Near surface, near front Figure V.37 Pilot 7 Temperature Profiles Pilot 8 - prunings /fruit /sawdust/ alder bark / Compost Pilot 8 was conducted in mid-summer, several months after pilot 7. In this batch, leaf prunings and fruit culls were mixed with alder bark, recycled coarse compost material, and used sawdust. This was based on experiences and results from the previous 7 runs, and the recommendations from the lab scale tests. This turned out to be a very successful run. This mixture composted very well, with PFRP met in most of the material, and a final product with a pleasant smell. 180 Pilot 8 Compost Temperature 80 70 60 o 50 ci 40 E * 30 20 10 0 TC1 — T C 3 -TC5 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 Ambient Temp=21.415.4 °C (avg±SD) Time (d) T C 1 - Near surface, near back TC2 - Near surface, near centre TC5 - Near surface, near front Figure V.38 Pilot 8 Temperature Profiles Conclusions from Pilot scale test results The pilot scale tests were not meant to be as analytical as the lab scale tests since there were not as many reactors as in the lab tests and the pilot scale tests were even more affected by external factors, e.g. larger differences in ambient temperatures, different mixings by different Bobcat operators, etc. However, several observations were made that were quite similar to those noted in the lab scale compostings and serve as re-confirmations of the lab scale findings. 181 First of all, at the beginning of the tests, the greenhouse operator worried that large amounts of manure would have to be mixed with the greenhouse vegetable wastes in order to get high themophilic temperatures. However, the pilot tests proved that without any inoculum, the greenhouse plant wastes (both daily prunings and year-end wastes) could be composted with temperatures passing PFRP. The greenhouse operator always would like to utilize the used sawdust growing medium as a bulking agent and additional carbon source, since it was part of the wastestream (costless) while alder bark could be a cost burden. However, it was reconfirmed that alder bark was needed to mix with daily prunings to provide good structure and porosity for composting otherwise the compost pile would become anaerobic and cause lot of odor and leachate problems. It was also reconfirmed that daily prunings could be composted without shredding, and this should mean cost savings for a greenhouse operator. The year-end wastes contained a lot of woody vines, and they provided the structure and porosity for proper aeration. So, the year-end wastes, with some used sawdust, could be composted without adding a bulking agent. This should make for another significant cost saving. In most of the pilot tests, temperatures went over the setpoint even when the aeration was on 100% cycle for cooling. This again proved that fresh air cooling might not be able to cool the compost pile effectively below the setpoint temperature. For the purpose of cooling (without tremendously increasing the fresh air aeration, which would increase the emission of odour), air recirculation is still the best option. 182 V.4.3 COST ANALYSIS OF GREENHOUSE IN-SITU COMPOSTING SYSTEMS The feasibility of in-vessel composting technology for on-farm use To minimize capital equipment costs, a relatively simple container system was selected for study. Some commercial in-vessel systems include mixing, which is a desirable feature, however it adds significant complexity and cost. The limitation of a containerized system is that it is a batch operation and cannot be continuously fed, therefore several units are required to handle an ongoing waste stream. However it is simple to operate, requires little maintenance, and can be loaded with standard, relatively low-cost materials handling equipment such as a small tractor or a Bobcat. The in-vessel design proved to be well-suited for prunings and rejects waste, which was high in moisture, putrescible and was generated over a long period. Forced aeration maximized aerobic conditions, odours were minimized, and leachate was contained. Compost of consistent, high quality could be produced. Year-end waste on the other hand was less putrescible, and a large amount was generated in a very short period at the end of the season. This pattern of waste generation was difficult to handle with the in-vessel system, as it would require stockpiling of waste for potentially long periods. Moreover, stockpiling the wastes without proper treatment would introduce disease and pest back into the greenhouse. An alternate, lower cost batch system with large capacity, such as aerated static pile, would be more cost-effective and manageable for year-end waste. 183 A practical solution would therefore be to use the in-vessel system for prunings and rejects waste during most of the season, producing a high-quality product suitable for potential use as a greenhouse growing medium. A lower-cost alternate system would be used for year-end waste, producing a good quality product but under less controlled conditions and therefore suitable for less critical or intensive applications such as farm soil amendment. Coarse, woody material present in shredded year-end waste can be screened out of the final product, and re-used as a bulking agent for composting prunings and rejects, thereby reducing the requirement for bulking agents from outside. Economic Feasibility Economic feasibility of on-site composting for growers will depend primarily on the following factors: existing organic waste disposal costs - cost of composting - market demand and value for compost product Conventional Disposal Costs Due to the variety of disposal methods used by growers, the existing organic waste disposal costs for growers vary widely. In some cases, the disposal cost at some greenhouses (especially some small operations) was reported as zero if the greenhouse operators could stockpile the wastes on their own land for a long time. Based on the limited cost information from surveys, and information from Hazelmere greenhouses, existing disposal costs were estimated as shown in Table V.34. As mentioned earlier, 184 conventional disposal costs are likely to increase with more reliance on off-site landfill disposal. In the case of all waste being trucked off-site, projected costs are also shown in Table V.34. These costs were based on the assumption that all plant waste would be landfilled at $70/T, plus $55/T trucking; and sawdust waste would be trucked to a local site at $23/T for land application. Table V.34. Conventional Disposal Costs. Generation Hazelmere Disposal Costs Average Existing Disposal Costs Projected Conventional Disposal Costs T/ha $/T $/ha $/T $/ha $/T $/ha Prunings and rejects 45.1 80 3,600 27 . 1,200 125 733 Year end 55.3 39 2,160 46 2,552 125 6,916 Sawdust medium 74.6 10 750 12 898 23 1,715 Total 175 6,510 4,650 11,016 Note : 1. Average existing disposal costs were calculated based on the surveys from 8 greenhouses. 2. For projected conventional costs, plant waste would be landfilled at $70/T, plus $55/T trucking; and sawdust waste would be trucked to a local site at $23/T for land application. Composting Costs Composting costs are made up of capital equipment costs and operating costs. Capital costs includ primarily composting system equipment (capacity-dependent) and materials handling equipment. Operating costs included labour costs for shredding, mixing, loading and unloading, and screening, plus maintenance. Capital costs would obviously also depend on the selection of composting technology. Table V.3 5 summarizes estimated capital costs of an in-vessel container composting system, aerated static pile composting system, and materials handling equipment. The in-vessel system costs are based on those found for the study system, 185 with some additional costs to up-grade from pilot to commercial scale. The static pile system cost is based on a covered, concrete pad with forced aeration. Table V.35 Composting System Component Costs Component Cost (S) Annual Raw Waste Capacity In-vessel container system, 3 containers 69,000 180 T 2 Covered aerated static pile system 25,000 225 T Site paving, storage sheds 13,000 -Shredders 20,000 -Screen and conveyor 15,000 -Total 142,000 405 T Note that the above costs did not include those of a loader (eg small tractor or skid-steer). The overall cost of in-vessel composting was estimated at approximately $200 to $250 per tonne of compost produced (see payback example later) including operating costs. Alternative Systems Alternative commercial systems, including different types of in-vessel systems, are available. Similar systems using batch containers without mixing are available from suppliers in the USA, including NaturTech (Seattle, WA) and Green Mountain Technologies (Seattle, WA). A two-container Green Mountain system with a total single batch capacity of approximately 30 T is estimated to cost US$100,000 (C$155,000), not 2 8.5 T/unit x 10 batches/year = 85 T/unit/year; 85 T - 30% (bulking agent) = 60 T/unit/year 186 including the roll-off truck required for unloading, shredder, screen, or site-related costs. Agitated (mixed) in-vessel systems are also available, for example Wright Environmental (North Vancouver, BC), or Transform Compost Systems (Abbotsford, BC); generally the cost of agitated in-vessel systems will be higher than non-agitated systems. Product Value and Marketing A market survey for greenhouse compost products was done and is reported in Chapter IX. Market demand and product value are key factors in the economic equation. Potential benefits to users include improved soil quality, disease suppression (Cheuk, 2003a), water and nutrient conservation, and improved yield (Haug, 1993; Hoitink, 1997). Based on site visits to existing composting facilities and other greenhouses and nurseries, a significant potential market exists in the lower Fraser Valley. The potential market for bulk product includes: - flower and vegetable greenhouses nurseries organic farms - landscapers At the time of the study, a compost product made from hog manure was being marketed to flower and vegetable greenhouses in the Abbotsford area. Though the typical price for bulk compost was historically near $60/T, this product sold for over $100/T. According to Composting Factsheets from B.C. Ministry of Agriculture (1991), a market value of good quality manure compost could be approximately S 1 0 0 / T . 187 Demonstration of high-quality greenhouse compost as a growing medium that can enhance production would obviously increase the market value of the product. On-site production using an in-vessel system would be advantageous in this respect, since the feedstocks are consistent, do not contain unknown components, and are processed under controlled conditions. Compost produced on-site could be partly utilized at the greenhouse, and the surplus marketed to other users. When compost is utilized as a growing medium amendment, even a small improvement in yield can lead to significant revenue increase. For example, 5% yield improvement for a 4 ha tomato greenhouse could lead to an annual revenue increase of nearly $200,000 based on 1996 figures (B. C. Agdex 257-810). ANNUAL COST ANALYSIS Based on a 4 ha tomato or pepper greenhouse, and amortizing the capital equipment over five years, the net annual cost of composting represents a savings of $8,000 (Cheuk, 2003b) annually as shown in the example in Table V.36. This analysis is based on the following assumptions: 1. Most prunings, rejects and year-end waste are currently hauled for local landfill disposal. 2. Separate in-vessel/aerated static pile systems are used for prunings and rejects, and year-end waste respectively. Materials handling equipment and site costs are included with each, not including a tractor/loader. 188 3. Weight reduction after composting are 50% for prunings and sawdust, and 40% for year-end wastes based on the results from lab and pilot scale tests. 4. Small percentages of sawdust are included in the composting feedstocks as shown; the rest is land applied at the disposal cost indicated. 5. Variable costs include production labour, maintenance, and electricity/fuel. 6. The annual fixed cost is calculated as amortization of the capital cost over 5 years with interest rate of 8%. 7. The compost selling price is $60/ton for the compost made from the aerated static pile and $130 for the compost made from in-vessel composter since the later one would have better and more consistent quality. However, these prices are still very conservative since a lot of good quality of composts have been selling between S133/T to $312/T (wholesale) and between $200/T to $562/T (retail) (see Table LX.l in Chaper EX). Cost savings could be made by sharing materials handling equipment between greenhouses, since shredding and screening would typically be done only for short periods. Also, the analysis did not take into account potential financial benefits due to re-use of compost on site as a growing medium amendment, which may be more financially attractive than marketing the same material to other users. 189 Table V.36 Annual Cost Analysis Example Waste Parameters Waste Generation Tons Disposal Cost S/Tons % Composted Compost production Tons ASP In-vessel ASP In-vessel Total Prunings and rejects 180 80 0% 100% 0 90 90 Year-end plants 221 80 80% 20% 115 22 137 Sawdust medium 298 10 10% 20% 15 30 45 Total 699 130 142 272 Compost Selling Price ($/Tons) ASP 60 In-vessel 130 System Costs Amortization Period: 5 years Capital Costs Annual Fixed Costs (amrt. 5 yrs @ 8% int. rate) Variable Costs Total Annual Costs Annual Disposal Savings Annual Compost Revenue Annual Net Cost $ $ $/T comp. $ $/T comp. $ $ $ ASP 52,500 12,805 30 16,825 126 14,400 7,800 (5,375) In-vessel 89,500 21,829 85 33,899 239 18,000 18,460 (2,561) Total 142,000 34,634 50,724 32,400 26,260 (7,936) 190 V.5 CONCLUSIONS The results of this study can be used for a better understanding of the effects of different modes of forced aeration and substrate recipes on the composting process and may help in the design and operation of future on-site greenhouse composting plants. The greenhouses wastes contained enough bio-degradable carbon to start the self-heating process and no manure nor bacterial inoculation was needed for the composting process. This resolves a lot of potential hygienic and transportation problems for greenhouse operators because provincial regulation does not allow on-farm composting to process foreign material, e.g. manure. The application of recirculation cooling and aeration created a more homogeneous vertical distribution of temperature in the composting mass and was a more effective method for maintaining the process temperature and caused much less temperature variations between each aeration period, thus enhancing the stability of the process. The greenhouse wastes contain a lot of seeds, from rotten fruit, in the waste stream. If the compost process temperature was not high enough to kill the seeds, the seeds would germinate in the curing pile, and this would be unfavourable in terms of selling the final product. Less leachate and condensate were generated by the air recirculation control, and this would mean less problems for the greenhouse operators. In the time series study, the NH3 concentration in the compost somehow followed a similar path to the temperature profile. As the process activity increased in the beginning, NH3 release increased, as proteins in the organic matter were broken down. 191 This peaked early during the thermophilic phase, then decreased as the amount of degradable organic matter and proteins decreased. So, NH3 could be a good indicator of compost maturity. By investigating the substrate structure, carbon loss, nitrogen loss, and process temperature, fresh alder bark was found to be the best bulking agent of those tested. Shredding was not necessary for composting the prunings and this would mean a significant savings for greenhouse operators. The amount of leachate was related to the initial moisture content of the substrate and the amount of condensate was correlated with the duration of the high temperature phase of the process. A bulking agent (alder bark) in amounts of about 20-30% (in weight) was necessary for composting prunings. For year-end wastes, a ratio of 62% vines, 13% used sawdust and 25% alder bark was recommended. The results of the pilot scale investigations enabled one to prove the usefulness of the in-vessel containerized system for composting greenhouse wastes. A variety of greenhouse waste materials and blends were able to be composted successfully and efficiently in the in-vessel system. The composting process sanitized the material sufficiently (55 °C for 3 days), and the time of composting was significantly shorter than any windrow or aerated static pile composting. Addition of bulking agents to pruning waste (leaves and cull fruit) was critical to maintain an aerobic process. With maintenance of proper porosity and aerobic conditions, odour was not a problem during composting or curing. The on-site composting of greenhouse wastes provided an environmentally and economically sustainable solution for the greenhouse industry. It not 192 only cut down on waste hauling costs, but also generated good quality compost for re-use in the greenhouse and for sale in the commercial market. 193 V.6 REFERENCES American Society of Agronomy. (1982) Methods of Soil Analysis. 2 n d Ed. Soil Science Society of America, Madison. Wisconsin. APHA. (1985) Standard Methods for Examination of Water and Wastewater, 16 Ed. American Public Health Association. Washington, D.C. Annual Statistics^ (1994) Ministry of Agriculture, Fisheries and Food. Government of British Columbia. Bari, Q.H., Koenig, A. (2001) Effect of air recirculation and reuse on composting of organic solid waste. Resource, Conservation and Recycling. 33(2) : 93-111. Barrington, S., Choiniere, D., Trigui, M. , Knight, W. (2002) Effect of carbon soruce on compost nitrogen and carbon losses. Bioresource Technology. 83(3) : 189-194. Barrington, S., Choiniere, D., Trigui, M. , Knight, W. (2003) Compost convective airflow under passive aeration. Bioresource Technology. 86(3): 259-266. Beck-Friis, B., Smars, S., Jonsson, H., Eklind, Y., Kirchmann, H. (2003) Composting of Source-Separated Household Organics at Different Oxygen Levels : Gaining an Understanding of the Emission Dynamics. Compost Science & Utilization. 11(1) : 41-50. 2003, Bernal, M.P., Sanchez-Monedero, M.A., Paredes, C , Roig, A. (1998) Carbon mineralization from organic wastes at different composting stages during their incubation with soil. Agriculture, Ecosystems & Environment. 69(3): 175-189. 194 Bidlingmaier, W., De Bertoldi, M. , Diaz, L., Papaddimitriou, F.K. (1999) The control of the process and compost quality. Organic Recovery and Biological Treatment, pp. 47-53. Rhombos-Verl publisher. Berlin. Bollen, G.J. (1993) Factors involved in inactivation of plant pathogens during composting of crop residues. Pg : 301-318. Science and Engineering of Composting : Design, Environmental, Microbiological and Utilization Aspects. Renaissance Publications, Worthington, Oh. British Columbia Ministry of Agriculture, Fisheries and Food. (1991) Composting Factsheets. 382.505 series. Government of British Columbia. British Columbia Ministry of Water, Land and Air Protection. (2002) The Production and Use of Compost Regulation, B.C. Reg. 334/93. Brown, M.B. (1997) Composting Apparatus. Biotechnology Advances. 15(3-4) : 805-806. Cheuk, W., Lo, K.V., Branion, R., Fraser, B., Copeman, R., Jolliffe, P. (2003a) Applying compost to suppress tomato disease. Biocyle. 44 (1): 50-51. Cheuk,W., Lo, K.V., Branion, R., Fraser, B. (2003b) Benefits of sustainable waste management in the vegetable greenhouse industry. Journal of Environmental Science and Health. V.38(6) (accepted August 2003) Das, K., Keener, H.M. (1997) Numerical model for the dynamic simulation of a large scale composting system. American Society of Agricultural Engineers. 40(4): 1179-1189. 195 Delhomenie, M.C., Bibeau, L., Bredin, N. , Roy, S., Broussau, S., Brzezinski, R., Kugelmass, J.L. and Heitz, M . (2002) Biofiltration of air contaminated with toluene on a compost-based bed. Advances in Environmental Research. 6(3): 239-254. Diaz, M.J., Madejon, E., Lopez, F., Lopez, R., Cabrera, F. (2002) Composting of vinasse and cotton gin waste by using two different systems. Resources, Conservation and Recycling. 34(4): 235-248. Elsgaard, L., Jorgensen, L.W. (2002) A sandwich-designed temperature-gradient incubator for studies of microbial temperature reponses. Journal of Microbiological Methods. 49(1): 19-29. Environmental Guidelines for Greenhouse Growers in British Columbia. (1994) B.C. Ministry of Agriculture, Fisheries and Food and B.C. Federation of Agriculture and the Greenhouse Industry of BC. Government of British Columbia. Faure, D., Deschamps, A . M . (1991) The effect of bacterial inoculation on the initiation of composting of grape pulps. Bioresource Technology. 37(3): 235-238. Feachem, R.D., Bradley, D.J., Garelick, H., Mara, D.D. (1983) Sanitation and Disease Health Aspects of Excreta and Wastewater Management. World Bank Studies in Water Supply and Sanitation 3. The World Bank, Washington, pp. 501. Finstein, M.S., Miller, F.C., Strom, P.F., MacGregor, ST., Psarianos, K . M . (1983) Composting Ecosystem Management for Waste Treatment. Biotechnology. 1: 347-353. Fraser, B.S. (1997) An investigation of In-Vessel Composting Process Control Strategies. M.A.Sc. Thesis, Dept. of Bio-Resource Engineering, University of British Columbia. 196 Golueke, C.G. (1991) Principles of composting. The Staff of BioCycle Journal of Waste Recycling. The Art and Science of Composting. The JG Press Inc., Pennsylvania, USA. pp. 14-27. Hamoda, M.F., Abu-Qdais, H.A., Newham, J. (1998) Evaluation of municipal solid waste composting kinetics. Resources, Conservation and Recycling. 23(4) : 209-223. Hansen, C , Keener, H.M., Hoitink, H.A.J. (1989) Poultry manure composting : an exploratory study. Transactions oftheASAE 36: 2151-2157. Hassen, A., Belguith, K., Jedidi, N. , Cherif, A., Cherif, M . and Boudabous, A. (2001) Microbial characterization during composting of municipal soild waste. Bioresource Technology. 80(3): 217-225. Haug, Roger T. (1993) The Practical Guide to Compost Engineerings Boca Raton: Lewis. Hay, J.C., Kuchenrither, R.D. (1990) Fundamentals and application of windrow composting. Journal of Environmental Engineering. 116(4) : 746-763. Held, C , Wellacher, M. , Robra, K.H., Gubitz, G.M. (2002) Two-stage anaerobic fermentation of organic waste in CSTR and UFAF-Reactors. Bioresource Technology. 81(1): 19-24. Hoitink, H. A., Zhang, W., Han, D. Y., and Dick, W. A. (1997) Making Compost to Suppress Plant Disease. Biocycle 38(4):40-42. Hoitink, H.A., Boehm, M.J. (1999) Biocontrol within the context of soil microbial communities : A substrate-dependent phenomenon. Annu. Rev. Phytopathol. 37: 427-446. 197 Hoyos, S.E.G., Juarez, J.V., Ramonet, C.A., Lopez, J.G., Rios, A.A., Uribe, E.G. (2002) Aerobic thermophilic composting of waste sludge from gelatine-grenetine industry. Resources, Conservation and Recycling. 34(3): 161-173. Hutchinson, C M . (1999) Trichoderma virens-Inoculated Composted Chicken Manure for Biological Weed Control. Biological Control. 16(2) : 217-222. Ichida, J.M., Krizova, L., LeFevre, C.A., Keener, H.M., Elwell, D.L., Burtt, E.H. (2001) Bacterial inoculum enhances keratin degradation and biofilm formation in poultry compost. Microbiological Methods. 47(2): 199-208. Illmer, P., Schinner, F. (1997) Compost turning - A central factor for a rapid and high-quality degradation in household composting. Bioresource Technology. 59(2-3) : 157-162. Ingham, E. (1998) Anaerobic bacteria and compost tea. Biocycle. 39(6): 86-91. Jeris, J.S., Regan, R.W. (1973) Controlling environmental parameters for optimum composting. Part II. Compost Sci. 14: 8-15. Johnson, G., Crawford, S., Sark, S.A. (1993) Sampling Municipal Solid Waste Compost. Biocycle. December issue: 61-64. Kaiser, J. (1996) Modelling composting as a microbial ecosystem : a simulation approach. Ecological Modelling. 91(1-3): 25-37. Ketterer, N. , Fisher, B., and Weltzien, H. C. (1992) Biological Control of Botrytis cinerea on Grapevine by Compost Extracts and their Microorganisms in Pure Culture. Recent Advances in Botrytis Research: Proceedings of the 10th International Botrytis Symposium, Heraklion, Crete, Greece. 198 Kirchmann, H., Bemal, M.P. (1997) Organic waste treatment and C stabilization efficiency. Soil Biology and Biochemistry. 29(11-12): 1747-1753. Koenig, A., Bari, Q.H. (1998) Effect of air recirculation on single-reactor and two-reactor composting system. In : Proceedings of 91 s t Annual Meeting and Exhibition, Air and Waste Management Association, June 14-18, San Diego, California. Paper No.98. Lei, F., VahderGheyst, J.S. (2000) The effect of microbial inoculation and pH on microbial community structure changes during composting. Process Biochemistry. 35(9): 923-929. Liang, C , Das, K.C., McClendon, R.W. (2003) The influence of temperature and moisture contents regimes on the aerobic microbial activity of a biosolids composting blend. Bioresource Technology. 86(2): 131-137. Liang, Y., Quan, X., Chen, J., Chung, J.S., Sung, J.Y., Chen, S., Xue, D., Zhao, Y. (2000) Long-term results of ammonia removal and transformation by biofiltration. Journal of Hazardous Materials. 80(1-3) : 259-269. Maboeta, M.S., van Rensburg, L. (2003) Vermicomposting of industrially produced woodchips and sewage sludge utilizing Eisenia fetida. Ecotoxicology and Environmental Safety. 35(2): 295-302. Madejon, E., Diaz, M.J., Lopez, R., Cabrera, F. (2002) New approaches to establish optimum moisture content for compostable materials. Bioresource Technology. 85(1) : 73-78. 199 Mari, I., Ehaliotis, C , Kotsou, M. , Balis, C. and Georgakakis, D. (2003) Respiration profiles in monitoring the composting of by-products from the olive oil agro-industry. Bio-Resource Technology. 87(3): 331-336. Martin, A.M. , Evans, J., Portger, D., Patel, T.R. (1993) Comparative effects of peat and sawdust emploted as bulking agents in composting. Bioresource Technology. 44 : 65-69. Mathur, S.P., Cook, D.G., Meyboom, P.A. (1994) A blueprint for compost research and development activities in Ontario. Composting Council of Canada. ppl4-16. Mathur, S. P. (1996) The Use of Compost as a Greenhouse Growth Media. Waste Reduction Branch, Ontario Ministry of Environment and Energy, Toronto, ON. McKinley, V.L., Vestal, J.R., Erapl, A.E. (1985) Microbial activity in composting (I). Biocyle. 26(9) : 39-43. McQuilken, M . P., Whipps, J. M. , and Lynch, J. M . (1994) Effects of Water Extracts of a Composted Manure-Straw Mixture on the Plant Pathogen Botrytis cinerea. World Journal of Microbiology & Biotechnology. 10:20-26. Monteoliva-Sanchez, M. , Incerti, C , Ramos-Cormenzana, A., Paredes, C , Roig, A. and Cegarra, J. (1996) The study of the anaerobic bacterial microbiota and the biotoxicity in various samples of olive mill wastewaters (Alpechin) during their composting process. International Biodeterioration & Biodegrdation. 38(3-4): 211-214. Morisaki, N . , Phae, C.G., Nakasaki, K., Shoda, M. , Kubota, H. (1989) Nitrogen transformation during thermophilic composting. Journal of Fermentation and Bioengineering. 67(1): 57-61. 200 Nakasaki, K., Ohtaki, A., Takano, H. (2000) Biodegradable plastic reduces ammonia emission during composting. Polymer Degradation and Stability. 70(2) : 185-188. Nakasaki, K., Yaguchi, H., Sasaki, Y., Kubota, H. (1993) Effect of pH control on composting of garbage. Waste Management & Research. 11(2): 117-125. Ouedraogo, E., Mando, A., Zombre, N.P. (2001) Use of compost to improve soil properties and crop productivity under low input agricultural system in West Africa. Agriculture, Ecosystems & Environment. 84(3): 259-266. Paredes, C , Bernal, M.P., Cegarra, J., Roig, A. (2002) Biodegradation of olive mill wastewater sludge by its co-composting with agricultural wastes. Bioresource Technology. 85(1): 1-8. Raviv, M. , Medina, S., Krasnoysky, A., Ziadna, H. (2002) Conserving nitrogen during composting. BioCycle. 43(9) : 48-50. Real, J.L., Merillot, J. (1996) Starting materials. The Science of Composting. 2: 1039-1040. Sanchez-Monedero, M.A., Roig, A., Cegarra, J., Bernal, M.P. (1999) Relationship between water-soluble carbonhydrate and phenol fractions and the humification indices of different organic wastes during composting. Bioresource Technology. 70(2): 193-201. Sartaj, M. , Fernandes, L., Patni, N.K. (1997) Performance of forced, passive and natural aeration methods for composting manure slurries. Transactions ofASAE. 40(2) : 457-463. Schloss, P.D., Walker, L.P. (2000) Measurement of process performance and variability in inoculated composting reactors using ANOVA and power analysis. Process Biochemistry. 35(9): 931-942. 201 Shi, W., Norton, J.M., Miller, B.E., Pace, M.G. (1999) Effects of aeration and moisture during windrow composting on the nitrogen fertilizer values of dairy waste composts. Applied Soil Ecology. 1 l(l):17-28. Shulze, K.L. (1962) Continuous Thermophilic Composting. Appl. Microbiol. 10: 108-122. Sidhu, J., Gibbs, R.A., Ho, G.E., Unkovich, I. (2001) The role of indigenous microorganisms in suppression of salmonella regrowth in composted biosolids. Water Research. 35(4): 913-920. Smars, S., Beck-Friis, B., Jonsson, H., Kirchmann, H. (2001) Structure and Environment : An Advanced Experimental Composting Reactor for Systematic Simulation Studies. Journal of Agricultural Engineering Research. 78(4) : 415-422. Smars, S., Gustafsson,' L., Beck-Friis, B., Jonsson, H. (2002) Improvement of the composting time for household waste during an initial low pH phase by mesophilic temperature control. Bioresource Technology. 84(3) : 237-241. Smet, E., Van Langenhove, H., DeBo, I. (1999) The emission of volatile compounds during the aerobic and the combined anaerobic/aerobic composting of biowaste. Atmospheric Environment. 33(8): 1295-1303. Sommer, S.G. (2001) Effect of composting on nutrient loss and nitrogen availability of cattle deep litter. European Journal of Agronomy. 14(2): 123-133. Stelmachowski, M. , Jastrzebska, M. , Zarzycki, R. (2003) In-vessel composting for utilizing of municipal sewage sludge. Applied Energy. 75(3-4): 249-256. 202 Thamdrup, B., Hansen, J.W., Jorgensen, B.B. (1998) Temperature dependence of aerobic respiration in a coastal sediement. FEMS Microbiology Ecology. 25(2): 189-200. Tiquia, S.M., Tarn, N.F.Y. (2000) Co-composting of spent pig litter and sludge with forced-aeration. Bioresource Technology. 72(1): 1-7. Tiquia, S.M., Tarn, N.F.Y., Hodgkiss, I.J. (1998) Salmonella Elimination during Composting of Spent Pig Litter. Bioresource Technology. 63(2) : 193-196. Tuomela, M. , Vikman, M. , Hatakka, A., Itavaara, M . (2000) Biodegradation of Linin in a compost environment: a review. Bioresource Technology. 72(2): 169-183. United States Environmental Protection Agency. (1994) A Plain English Guide to the EPA Part 503 Biosolids Rule. EPA/832/R-93/003. Washington, D.C. VanderGheynst, J.S., Gossett, J.M., Walker, L.P. (1997) High-solids aerobic decomposition : pilot-scale reactor development and experimentation. Process Biochemistry. 32(5): 361-375. Vinneras, B., Bjorklund, A., Jonnson, H. (2003) Thermal composting of faecal matter as treatment and possible disinfection method - laboratory-scale and pilot scale studies. Bioresource Technology. 88(1): 47-54. Watanabe, H., Kitamura, T., Ochi, S., Ozaki, M . (1997) Inactivation of pathogenic bacteria under mesophilic and thermophilic conditions. Water Science and Technology. 36(6-7) :25-32. Weppen, P. (2001) Process calrimetry on composting of municipal organic wastes. Biomass and Bioenergy. 21(4): 289-299. 203 CHAPTER VI COMPOST QUALITY VI.l BACKGROUND AND LITERATURE RESEARCH One of the purposes of composting is to produce an organic soil amendment that is beneficial to plants. Al l composting process should be designed and operated to produce a stable product that is beneficial to plants and not phytotoxic. Organic components of compost undergo several transformations during the composting process, producing metabolities which exhibit inhibiting or stimulating effects on plant growth (Wong, 2001). Haug (1993) defined compost as an organic soil conditioner that has been stabilized to a humus-like product, that is free of viable human and plant pathogens and plant seeds, that does not attract insects or vectors, that can be handled and stored without nuisance, and that is beneficial to the growth of plants. A number of approaches have been used to measure the degree of stabilization and to judge the quality of the compost product, including temperature decline, C:N ratio, chemical characteristics, and plant bioassays. Temperature Decline The rate of heat production from the compost mass should be proportional to the rate of organic oxidation. The composting process first undergoes a thermophilic stage (high temperature, i.e. above 55°C), then a mesophilic stage (medium temperature, i.e. 40-45 °C), and finally drops back to ambient temperature. As materials are decomposed, 2 0 4 less bio-degradable carbon is available, and the rate of oxidation decreases. The maturation of compost is accompanied by a decline in pile temperature (Tiquia, 2002). Such a decline indicates that the process is nearing completion and that the compost has been stabilized enough to reduce its nuisance potential and its level of phytotoxic metabolities. C:N Ratio A carbon to nitrogen (C/N) ratio of less than 20 is often used as an indicator of compost maturity. However, the actual C/N ratio depends largely on the C/N ratio of the starting material as well as the proportion of degradable carbon, so a final C/N ratio of less than 20 is only a guideline, but not an absolute condition for compost maturity (Jimenez, 1989). If the organic matter in compost has a high C:N ratio and decomposes continuously, it can rob the soil, to which it is added, of the nitrogen needed to support plant growth. Moreover, if the organic matter has a low C:N ratio, it can release ammonia which can be phytotoxic. Chemical characteristics Chemical characterization of compost is generally based on two criteria, agronomic value and heavy metal content (Soumare, 2003). Agronomic values usually refer to the availability of major elements, such as N , P, K and trace elements (e.g. Cu, Zn, Mn, Fe, Co, Mo). For heavy metal content, different guidelines and rules have been established for compost specification. 205 Ammonia is usually present in the early stage of composting as organic nitrogen is decomposed. The ammonia concentration is eventually reduced through volatilization or oxidation to the nitrate form. During the composting process ammonia production decreases in the later stages, and production of nitrate increases. The presence of nitrate and the absence of ammonia can therefore be used as an indicator of compost maturity (Haug, 1993). Soluble phosphorus (phosphate) is an important nutrient for plant growth and is often lost through the leachate of composting (Eghball, 1999). Therefore, the concentration of phosphate is sometimes used for compost analysis as well. Phytotoxicity Studies have shown that toxicity during organic matter decomposition was strongly associated with the initial 3-4 weeks, subsequently decreasing rapidly and disappearing after 2 months (Zucconi, 1981). Plant bioassays are probably the most direct method to determine whether the compost product has been sufficiently stabilized. Potting growth trials could be very time consuming. Therefore, many researchers have used a cress seed phytotoxicity bioassay as a standard germination test for compost evaluation to determine compost maturity and to detect the presence of phytotoxic compounds (Baca, 1990; Fang, 1999; Ball, 2000). Cress seeds (Lepidium sativum L.) were chosen because of their sensitivity to toxic substances and their speed of germination (Murillo, 1995). Seeds are incubated in compost extract with distilled water 206 as control. Germination rates and root elongation are measured to determine the phytotoxicity of compost. Compost By-Prbducts - Leachate, Condensate and Compost Extract Apart from the compost mass produced from the bio-degradation process, a few-other "by-products" should also be investigated. These are leachate, condensate and compost extract. The leachate is commonly known as "compost tea" because of its brown colour, opacity and the analogy of water drawing nutrients and organics out of the compost as it passed through the compost piles. Condensate could be collected from the ceiling of an in-vessel composter and from the aeration outlet when the gases from the composter are condensed before emission. These compost by-products have been considered waste products but may prove to be beneficial. These liquids contain nutrients and organics that have the potential to pollute surface water, but also can be used to improve soil quality and fertilize plants, even in greenhouses, if applied properly. Considering the trend to return organics to agricultural soils in order to stem soil erosion, and the effort to reduce industrial fertilizer use, the reuse of this material should become a high priority (Clean Washington Center, 1997). Compost extracts are made by soaking compost with water and filtering out the solids afterwards. Compost extracts can suppress disease in plants via a number of mechanisms, including competition, antibiosis, hyperparasitism, and induction of systemic acquired resistance (SAR) in some host plants (Hoitink, 1997). Competition and hyperparasitism mechanisms would require the presence of active microorganisms, while 207 antibiosis and SAR could potentially be effective in a sterilized medium. Soilless media such as sawdust or peat support limited microbial populations, compared to properly prepared compost. For this reason, these suppressive mechanisms of compost can be effective for soilless media, due in part to the high microbial activity and biomass of the compost. The effectiveness of a compost amendment and compost extract is largely determined by the feedstock materials and the method of processing and storage, as well as the type of crop. As previously reported (Mathur, 1996), compost can be phytotoxic if it is not processed or applied correctly. Parameters including moisture, pH, carbon to nitrogen ratio, and maturity impact the effectiveness of disease suppression; low moisture content during curing favours fungal growth and development of Pythium diseases (Hoitink, 1997). Curing time was found to be important to the microbial populations and the disease suppression characteristics; inoculation of suppressive organisms early in the curing phase has also been found to improve disease suppression characteristics (Hoitink, 1997). In addition, selection of appropriate feedstock and processing methods can help optimize the microbiological makeup of the product for a particular crop in order to give the best results (Grobe, 1998). A high bacteria to fungi ratio has also been suggested as a desirable characteristic for increasing disease suppression (Kai, 1990). Beneficial liquid extracts of compost would provide another important avenue for application of beneficial compost products. Water extracts of compost were found to significantly reduce the severity of grey mould (Botrytis) for lettuce and grapevine (Ketterer, 1992; McQuilken et al., 1994). However, there are a number of variables 208 involved in the production of extract that may impact the product effectiveness, including aeration, steeping time, and provision of a carbon source (McQuilken, 1994; Ingham, 1999). The effect of storage and dilution when applying are also important for commercial application. VI.2 SPECIFIC OBJECTIVES The overall objective of this part of the project was to correlate the compost quality to the process control and substrate recipes, so that the compost and by-products generated from the process could be fully utilized. The specific objectives are: • Investigate any correlation between compost quality and process control. • Investigate any correlation between compost quality and substrate recipe. • Investigate the compost maturity and quality in terms of seed germination (compared to water). • Investigate the phytotoxicity level of the bio-conversion by-products, i.e. leachate and condensate, and thus find out the level of dilution for application onto plants. • Investigate whether the compost extract would enhance seed germination and shoot growth (for nursery use). 209 VI.3 METHODOLOGY Chemical Analysis Parameters, such as C:N ratio, pH, EC, ammonia, nitrate, phosphate were analyzed and compared in different lab and pilot tests. Total Organic Carbon was measured by combustion at 680 °C and CO2 measurement, using a Shimadzu TOC-5050 Total Organic Carbon Analyser with an SSM-5000 Soild Sampling Module. Total Nitrogen was measured using ignition at 950 °C in a Leco FP228 Nitrogen Determinator. pH and EC were measured by Hach pH and EC meters compensated for temperature. Water soluble Ammonium, Nitrate and Phosphate were measured using water extraction and the Technicon Autoanlyser II industrial method. Bio-Assay for Compost Extract Bioassays were carried out to determine the toxicity of compost using cress seeds (Lepdium sativum L.) and tomato seeds (Lycopersicum esculentum L.). Compost was collected from the greenhouse composting reactors and was sent to UBC Lab (Chemical and Bioloigcal Engineering) in transparent plastic bags immediately after collection. Approximately 15 grams of fresh compost were placed in a 250 ml Erlenmeyer flask and 135 ml of distilled water were added to the flask. The flask was placed on a shaker table and shaken for one hour. The liquid was then pipetted off and centrifuged for 15 minutes. The extract, with solids settled out, was collected. The extract was refrigerated at about 4°C for not more than 14 days if it was not used immediately. 210 To begin the bioassay, a Whatman no. 3 filter was placed in a 10 cm Petri dish and 2.5 ml of compost extract were added. Twenty-five seeds were added to the plate, arranged in a five by five square. Four dishes, i.e. a total of 100 seeds, were used for each test. For cress seeds, the plates were incubated in the dark in a temperature controlled incubator at 28.5°C for 24 hours. The number of cress seeds that germinated and the length of each seed's root were recorded for each plate. For tomato seeds, since they took a longer time to germinate, the plates were incubated in the dark in a temperature controlled incubator at 28.5°C for 144 hours (6 days). An additional 2.5 mL of the compost extract under testing were added after two days to the tomato seeds to prevent drying out of the seeds. The number of tomato seeds that germinated after 6 days and the length of each seed's root were recorded for each plate. Germination index (GI) was calculated according to the following formula (Zucconi, 1981; Tiquia, 1996): Equation VI. 1 Germination index GI = Seed germination % X Root length of treatment X 100 Seed Germination % X Root length of control Bio-Assays for Leachate and Condensate Bioassays were carried out to determine the toxicity of leachate and condensate from the composting process using cress seeds (Lepdium sativum L.) and tomato seeds (Lycopersicum esculentum L.) Leachate and condensate were collected from the greenhouse composting reactors and stored at about 4°C for not more than 14 days. Various dilutions were made with distilled water. 211 To begin the bioassay, a Whatman no. 3 filter was placed in a 10 cm Petri dish and 2.5 mL of a given leachate or condensate concentration were added. Twenty-five seeds were added to the plate, arranged in a five by five square. Four replicates (total 100 seeds) were done for each dilution. For cress seeds, the plates were incubated in the dark in a temperature controlled incubator at 28.5°C for 24 hours. The number of cress seeds that germinated and the length of each seed's root were recorded for each plate. For tomato seeds, since they took a longer time to germinate, the plates were incubated in the dark in a temperature controlled incubator at 28.5°C for 144 hours (6 days). An additional 2.5 mL of diluted leachate/condensate were added after two days to the tomato seeds to prevent drying out of the seeds. The number of tomato seeds that germinated after 6 days and the length of each seed's root were recorded for each plate. In some cases, both distilled water and nutrient water (from greenhouse operation) were used as control comparison. Statistical Analysis When data were subjected to statistical analysis, comparison of the mean was performed by the Student's t test. A significance level of PO.05 was used throughout the study. The following formula was used in the Mest for comparing two means : Equation VI.2 /-test ( x i - x 2 ) / = V((S,2/Ni) + (S22/N2)) 212 where X i and X2 are the means of two samples, Si and S2 are the standard deviations of the two samples, and Ni and N2 are the sizes of the two samples. The degrees of freedom are the smaller of Ni -1 or N2 -1 . In the bioassay study, this test was used to confirm whether the mean root length of one sample group was significantly larger or smaller than another group. For example, the mean root length of the seeds (NA=100) germinated from compost extract of Lab 8C is 4.87 mm with a standard deviation of 1.20 mm, and the mean root length of the seeds (N c = 100) from compost extract of Lab 8D (distilled water) is 5.24 mm with a standard deviation of 0.94 mm. The t-value (tgc,8D) is then calculated to be 2.43, which is larger than to.05,100, i.e. 1.660. This means the root lengths of 8D are significantly longer than that of 8C, with a confidence level of 95%. VI.4 RESULTS AND DISCUSSIONS VI.4.1 COMPOST ANALYSIS All the compost analyses for the different batches of composting described in Chapter V, are listed in Table VI. 1. Generally, most of the finished composts had a reduced C:N ratio from the initial values, except for a few occasions that started with low C:N. A lower final C:N meant the rate of carbon loss was higher than the rate of nitrogen loss in the process, which was a good indication of significant bio-degradation with some nitrogen conservation. Most of the C:N ratios were within the desirable range of 15 to 30 C:N(Haug, 1993). 213 The ammonia and nitrate concentrations were much higher in the compost inoculated by manure. That was to be expected since the manure contained high concentrations of ammonia and nitrate (Krapac, 2002). Thus, it is often used as nitrogen amendment source for fertilizers. The ammonia concentrations of compost from reactors with air recirculation were higher than from those without air-recirculation. This was due to less leachate generation and recirculation of the volatized ammonia during the beginning of the process (Bari, 2001). Ammonia concentration was highest with continuous, constant aeration in Lab 4 (reactor D), which indicated that the compost was not as stabilized as the others produced in the air-recirculated reactors (reactors B,C). In Lab 8 and Lab 9, the greater was the percentage of bulking agent (alder), the less was the ammonia and nitrate concentration in the finished compost, which also showed that the compost was more stabilized. Studies have shown that phosphate should decrease with degradation of substrate by microbial consumption for cell assimilation (Yun, 2000). Enrichment of partially composted crop wastes could be achieved by phosphate solubilizers to improve the nitrogen, available phosphorus and humus content of finished compost (Gaur, 1987). However, there was no particular trend nor correlation found in this present study for phosphate concentration. Yun et al. (2000) also showed that it was possible to predict a slurry-phase bioreactor's performance, under various conditions, from pH and DO measurements. The pH began to increase as degradation of solids began and the DO began to increase when the degradation process was completed. It was inferred that the reactor operated with a 214 stable decomposition of food wastes when the pH was about 8 and the DO greater than 2 mg/L. For most cases of the lab runs in this study, the pH values were at about 8, which provided a stable degradation. Table VI.2 shows a comparison of analyses made by a commercial laboratory, between compost from year-end wastes (Lab 9C), compost from daily pruning (Lab 8C) and the un-used yellow cedar sawdust medium. In each case, 3 to 5 samples were taken and mixed as a composite sample of 250 grams for analysis (Johnson, 1993). It is obvious that the pH values of both composts were higher (alkaline) than sawdust. This was unfavourable for a greenhouse vegetable crops, which prefer a pH value of 6 to 6.5 (slightly acidic) (BC Ministry of Agriculture, 1997). So, pH adjustment needed to be made before utilizing the compost in greenhouse. The EC was much higher in the daily prunings (15.02 mS/cm) than in the year-end wastes (1.18 mS/cm) and the year-end waste was slightly higher than sawdust (0.46 mS/cm). So, there were much more macro and micro-nutrients in the compost of daily prunings, which contained a lot of leaves and culled fruit, than in the compost of year-end vines and sawdust. The Cation Exchange Capacity (CEC) values, which are a measure of the nutrient holding (adsorption) power of the material and are related to potential fertility, were much higher in both compost (162.2 meq/lOOg and 143.8 meq/lOOg) than in the sawdust medium (8.2 meq/lOOg). So, the compost made from greenhouse wastes had a higher potential fertility and humification (Sanchez-Monedero, 1999). 215 The nutrient contents (N, P, K, Ca, Mg ,S) of both composts were considerably higher than that of the sawdust. Al l these nutrients are favourable to plants and could be used in combination with other media and fertilizers for plant production (Vogtmann, 1993). The organic matter content and C:N ratio were lower in the composts because a lot of the organic carbon had already been oxidized (degraded) in the composting process. The heavy metals concentrations (Fe, Cu, Mn, Zn) tended to be higher in the composts. However on the basis of seed germination, there was no obvious phytotoxicity, with seed germinations of 100% and 98% in the year-end and prunings composts respectively. Even though the composts had lower porosity than the sawdust (53.1% and 49.7%) vs 63.4%), they showed higher water holding capacity (22.8% and 26.9% vs 16.6%) which reduced the leaching out of nutrients. 216 . Table VI. 1 Analysis from different Lab tests Sample Ammonia mg/L Nitrate mg/.L ... Phosphate ; mg/L' C:N Ratio Input C:N Ratio Output Lab 1A with manure 41.2 80.2 119 - 24.0 23.7 Lab IB w/o manure 23.1 60.3 53.3 - 25.4 20.9 Lab 2A w/o air-recirculation 230 250 298 - 29.2 10.2 Lab 2B w/air-recirculation 270 150 322 - 29.2 16.3 Lab 3A w/o air-recirculation 92.4 75.4 311 8.29 9.13 13.7 Lab 3B w/ air-recirculation 187 45.1 277 8.64 9.13 • 17.1 Lab 4A fresh air aeration/fresh air cooling 38.1 75.2 72.3 8.39 12.8 18.9 Lab 4B fresh air aeration/ air-recirculation cooling 49.3 60.2 138 8.04 12.8 12.2 Lab 4C fresh air + recir. aeration/recir. cooling 56.1 75.1 89.1 8.43 12.8 13.2 Lab 4D fresh air continuous constant aeration 102 90.1 121 7.94 12.8 14.1 Lab 5A Non-shredded with hemlock 0.58 27.6 54.5 6.82 30.2 26.4 Lab 5B Shredded with hemlock 1.93 11.2 35.6 6.81 33.9 29.4 Lab 5C Non-shredded with alder 1.06 0.31 23.3 6.92 32.6 24.8 Lab 5D Shredded with alder 1.21 0.33 16.2 7.03 34.6 24.9 Lab 6A Non-shredded with alder 0.00 25.7 14.5 7.94 30.9 22.7 Lab 6B Non-shredded with hemlock 2.33 27.9 26.4 7.60 24.1 23.0 Lab 6C Shredded with alder 4.87 23.6 2.98 8.19 30.9 23.7 Lab 6D Shredded with hemlock 16.2 27.6 3.32 7.59 24.1 23.5 Lab 7A Shredded with fresh alder 24.3 56.4 245 8.36 38.2 20.2 Lab 7B Non-shredded with fresh alder 29.4 209 169 9.33 38.2 18.4 Lab 7C Shredded with composted alder 39.5 254 23.9. 8.73 26.9 18.4 Lab 7D Non-shredded with composted alder 38.1 157 . 109 8.66 26.9 16.2 Lab 8A Non-shredded 170 24.4 21.9 7.92 7.80 9.60 Lab 8B Non-shredded with 15% alder 108 11.7 15.2 7.91 15.7 13.6 Lab 8C Non-shredded with 30% alder 84.3 10.3 15.1 7.91 20.6 16.9 Lab 8D Non-shredded with 45% alder 29.9 1.79 8.67 7.72 23.9 19.7 Lab 9A 60% Prunings, 40% sawdust 8.68 37.3 27.9 6.92 21.6 29.3 Lab 9B 60% Prunings ,14% alder, 26% sawdust 5.34 10.1 15.9. 7.51 24.1 27.3 Lab 9C 62% Prunings, 25% alder, 13% sawdust 2.13 5.93 16.7 7.74 26.0 21.4 Lab 9D 65% Prunings, 35% alder 2.50 5.72 15.5 7.62 33.3 25.9 217 Table VI.2 Growing Medium and Amendment Analysis (Compost) I nil Compost from Year End Compost from Daily LJn-iised Yellow Wastes (from Lab l)C after Prunings (from Lab 8C Cedar Sawdust composting process) after composting Growing process) Medium pH 7.93 9.43 5.88 SMP Buffer pH 7.67 8.02 7.67 E C mS/cm 1.18 15.02 0.46 Sodium-Na ppm 1110 1390 80 C E C meq/lOOg 162.2 143.8 8.2 Ammonia-N ppm 2.62 4.28 0 Nitrate-N ppm 907 1470 8.4 Phosphate-P ppm 846 1874 20 Potassium-K ppm 24000 29300 190 Calcium-Ca ppm 14000 9200 1300 Magnesium-Mg ppm 3100 2000 100 Sulphate-S ppm 2010 12040 296 Organic Matter % 52.2 62.7 98.6 T O C %db 30.2 38.2 52.4 T N %db 2.2 1.6 0.71 C/N ratio 23.9 13.7 73.8 Moisture % 57.3 62.7 65.9 Iron-Fe ppm 740 220 200 Manganese-Mn ppm 122 83.7 78.2 Zinc-Zn ppm 31.9 26.1 21 Copper-Cu ppm 4.1 2.8 3.5 Chloride-Cl ppm 3300 3000 4200 Seed Germination % 100% 98 98 Bulk density kg/m3 570 587 409 Total Porosity % 53.1 49.7 63.4 Water-holding capacity % 22.8 26.9 16.6 Particle Size >4.8 mm 17 19 14 2.4-4.8 mm 35 36 42 1.0-2.4 mm 40 35 40 <1.0 mm 7 10 4 218 VI.4.2 BIOASSAYS ON COMPOST EXTRACT Chemical analysis before and after composting could not fully reflect compost quality nor be used correlate with the compost process. Therefore, bioassays were done as another means of comparing compost quality and maturity in the following sections. In this section, compost extracts from different compost runs were investigated by bioassays. The objectives were to investigate any beneficial nutrient values in compost extract and also, to compare quality variations in each test run. The bioassays were done in terms of the % germination, root length and germination index of cress and tomato seeds (in Lab 7 only). Table VI.3 and Figure VI.I present the bioassay results for Lab 5. Table VI.3 Bioassay results from Lab 5 % germ, 1- * day cress seed avg length, mm (mcaniSD) ( i . l . .% of control t-valuc to control t-value to others Control, distilled water 96 4.06±0.52 Lab 5A Non-shredded with hemlock 95 4.85±0.79 118 tC,5A>3 t5A,5B >3, t sA .SC -^ , t5A,5D>3 Lab 5B Shredded with hemlock 99 5.25±0.81 133 tc,5B>3 t5B,5C >3, t 5 B ,5D > 3 Lab 5C Non-shredded with alder 98 6.73±1.07 169 tc,5C>3 t5C,5D >3 Lab 5D Shredded with alder 99 6.28±0.60 160 tc,5D>3 * ? c 5 A is the /-value of comparing Control and 5A, etc. t-value at 95 % confidence level of sample number 100, i.e. /o.os.loo is 1.660. 219 In the Lab 5 bioassay test, all the compost extracts showed no phytotoxicity (95-99% vs 96% for the control). Al l compost extracts showed beneficial growth effects compared to the control (/-values >3) with G.I. (germination index) higher than 100. Both compost extracts from shredded and non-shredded wastes showed similar G.I. which further proved that there were not much difference in quality between composts made shredded and non-shredded prunings, as concluded in Chapter V. Also the results showed that composting with alder bark (G.I.=169, 160), whether or not with shredded or non-shredded wastes, was better than with hemlock bark (G.I.=118, 133). This further proved that alder bark was a better choice, as bulking agent, than hemlock bark as concluded in Chapter V. Control L5A L 5B L 5C L 5D Figure VI. 1 Bioassay on Lab 5 - 1-day cress seeds 220 Table VI.4 and Figure VI.2 present the bioassay results for Lab 6. In the Lab 6 bioassays, all the compost extracts showed no phytotoxicity (96-100% vs 100% for control). Al l the compost extracts showed beneficial growth effects significantly greater than the control (t-values >3), except that one, non-shredded with hemlock bark, did not pass the 95% confidence level test (t=0.51), with a G.I. of 100. Also, it is shown that the non-shredded prunings with alder bark as bulking agent stood out to be significantly the best among the other 3 combinations. This again further proves the argument that alder bark with non-shredded waste was found to be a better combination in Chapter V. Table VI.4 Bioassay results from Lab 6 (t-test compared to distilled water) % germ, 1-day cress avg length, mm (mean±SD) . c.i! '%ojf control t-value to control , t-value to others Ctrl, distilled water 100% 8.0±2.75 100 - -Lab 6A Non-shredded with alder 98% 11.2±2.42 t>3 137 tc,6A>3 t6A,6B>3, t6A,6C>3, t6A,6D>3 Lab 6B Non-shredded with hemlock 98% 8.2±2.84 t=1.22 100 tc,6B=0.51 t6B,6C>3, t6B,6D>3 Lab 6C Shredded with alder 96% 9.7±2.46 t>3 116 tc.6C>3 t6C,6D=l-48 Lab 6D Shredded with hemlock 100% 9.2±2.30 t=2.47 115 tc.6D>3 221 Control L6A L6B L 6C L6D Figure VI.2 Bioassay on Lab 6 - 1-day cress seeds Tables VI.5 and VI.6 and Figures VI.3 and VI.4 present the bioassay results for Lab 7. In the Lab 7 bioassays, two kinds of seeds were used: cress and tomato. They were allowed to germinate in the Petri dishes for 1 and 6 days respectively. The results are shown in Table VI.5, and Table VI.6. The compost extracts again showed no phytotoxicity (98-100% germination vs 98%) for the control; GI between 116 and 132) in the cress seeds and even higher difference in germination rate in tomato seeds (90-97%) vs 85% for control; GI between 270 and 439). This should be of interest to propagation companies since using compost extract could then cut down their seed loss through failure to germinate. The results from cress seeds and tomato seeds are very similar in 222 terms of germination index. This shows that both cress ands tomato seeds can be used for compost bioassay, except that cress seeds can give much faster results. There were improvements in growth and germination rate when using compost extract, especially with tomato seeds. Within the different extracts, the one composted with non-shredded prunings and fresh alder bark showed significantly better results. This correlates with the results from Chapter V, where it was shown that reactor B had the highest degradation rate and longest reaction period. Table VI.5 Bioassay results from Lab 7 (Cress seeds) % germ. 1-day cress seed avg length. mm ( m c a n i S D ) ( i . l . % of control t-value to control t-value to others Control, distilled water 98% 9.2±0.99 100 - -Lab 7 A shredded with fresh alder 99% 10.9±0.93 120 tc,7A>3 T7A,7B > 3, t 7 A , 7 C > 3 , t7 A,7D=l-34, t7A.PAB=2.52 Lab 7B non-shredded with fresh alder 100% 11.9±0.76 132 tc.7B>3 t7B,7c=2.45, t 7 B,7D>3,, t 7B,PAB > 3 Lab 7C shredded with composted alder 98% 11.5±1.44 125 tc.7C>3 T7C.7D > 3, t7CPAB >3 Lab 7D non-shredded with composted alder 100% 10.7±1.17 119 tc,7D>3 t7D,PAB=0-72 Pure alder bark 99% 10.6±0.74 116 tc,PAB >3 -223 Figure VI.3 Bioassay on Lab 7 - 1-day cress seeds Table VI.6 Bioassay results from Lab 7 (Tomato seeds) "A> g e r m , 6-day tomato seed avg l e n g t h , m m(mean±SI)) G . I . % of control t-value to control t-value to others Control, distilled water 86% 11.3±4.88 100 - -Lab 7A shredded with fresh alder 97% 38.3±16.38 382 tc.7A>3 t7A,7B=2-78, t7A,7C =l-31, t7A,7D=2-11, t7A.PAB >3 Lab 7B non-shredded with fresh alder 94% 45.4± 19.57 439 tc,7B>3 t7B,7c=2.45, t7 B,7 D>1.54, t7B.PAB >3 Lab 7C shredded with composted alder 96% 41.4±17.05 408 tc.7C>3 t7C,7D=0.93, t7C,PAB >3 Lab 7D non-shredded with composted alder 90% 43.9±20 .75 406 tc,7D>3 t7D,PAB >3 Pure alder bark 95% 27 .7±10 .80 270 tc,PAB >3 224 ro 2 c V i V 0. 500 450 400 350 300 250 200 150 100 50 0 I Germination Rate, % I Germination Index, % Root Length, mm • • Control L 7 A I 50 45 ;- 40 - 35 r 30 | 25 £ ro 20 q 15 I- 10 5 0 L 7 B L 7 C L 7 D Pure Alder Figure VI.4 Bioassay on Lab 7 - 6-day tomato seeds Table VI.7 and Figure VI.5 present the bioassay results for Lab 8. The Lab 8 bioassays verified the results discussed in Chapter V. The reactor A compost, which could not even attain a temperature above 45 °C, showed the highest phytotoxicity (70% vs 96% for the control; GI=64). Then came reactor B with 75% germination (GI=76), and reactor C with 78% germination (GI=98). Reactor D, which had the highest and longest sustained temperature, showed much lower phytotoxicity (92%) and much higher GI (124) than the others. Reactor A compost extract inhibited the growth of seeds (3.53 mm vs 4.04 mm for control), while Reactor D compost extract enhanced the growth of seeds (5.24 mm vs 4.04 mm for control). The Lab 8 bioassay showed an obvious trend: the lower the ratio of bulking agents, the lower the quality of the finished compost. The germination rate, root length, germination index and the number of days of thermophilic 225 and mesophilic reactions were correlated, i.e. they all showed an upward trend from A, B, C, and D, with A being the lowest. Table VI.7 Bioassay results from Lab 8 avg length, G.I. t-value to t-value to % germ, 1-day mm % 0 f control others crest seed ( n i e a n ± S D ) control Control, distilled water 96 4 .04±0.48 100 Lab 8A non-shredded 70 3.53±0.78 64 tc,8A>3 t8A,8B >3, t8A,8C >3, t8A.8D >3 Lab 8B non-shredded with 15% alder 75 3.95±0.89 76 tc.8B=0.89 t8B,8C >3, t8B,8D >3 Lab 8C non-shredded with 30% alder 78 4 .87±1 .20 98 tc,8C>3 t8C,8D=2.43 Lab 8D shredded with 45% alder 92 5.24±0.94 124 tc.8D>3 Control L8A L 8B L 8C L 8D Figure VI.5 Bioassay on Lab 8 - 1-day cress seeds 226 Table VI.8 and Figure VI.6 present the bioassay results from Lab 9. In Lab 9, used sawdust from the greenhouse waste stream was tested as an alder bark replacement. However, the best results were still the ones with high proportions of alder bark. Reactors A and B, with low proportions of alder bark, showed higher phytotoxicity (88% and 76% respectively than 94% for the control). Reactors C and D, with more alder bark, showed low phytotoxicity (90% and 92% vs 94% for control), significantly higher growth rates and GI than the control. This also correlated with the results and recommendations in Chapter V (the recommend best ratio was about 25% alder, 13 % sawdust, and 62% raw wastes) and there is nothing in the bioassay results that would suggest modifying this recommendation. Table VI.8 Bioassay results from Lab 9 % germ, 1-day crest seed :|\g length, m in (n iean±SD) . C . L | % olj control t-value to control t-value to others. Control, distilled water 94 4 .25±0.96 100 Lab 9A prunings, 40%sawdust 88 4 .35±1.55 96 tc,9A=0.54 t9A,9B=0.48 4, t9A,9C>3, t9A.9D >3 Lab 9B prunings ,14%alder, 26%sawdust 76 4 .25±1.36 80 tc,9B=0 t9B,9C >3, t9B,9D>3 Lab 9C prunings ,25% alder, 13%sawdust 90 5.3±1.49 119 tc,9C>3 t9C,9D =l-75 Lab9D prunings,35%alder 92 5.65±1.33 130 tc,9D>3 Figure VI.6 Bioassay on Lab 9 - 1-day cress seeds 228 VI.4.3 BIO ASSAYS ON WASTES, LEACHATE AND CONDENSATE Studies have shown that leachate and condensate from composting facilities were phytotoxic and they were treated as water pollutants (Peot, 1997; Agassi, 1998; Tyler, 2000; Kaschl, 2002). High levels of nitrogen, phosphorus and potassium in these by-products of composting can have detrimental effects on surrounding surface waters. However, for the same reasons, they can be used to improve soil quality and fertilize crops if applied at proper rates. The leachate from greenhouse wastes should be considered as much "cleaner" than leachate from any municipal wastes composting sites since it contains no human pathogens, food debris, or dead animals. This study was designed to examine more closely the by-products - leachate and condensate from a greenhouse waste composting facility and investigate their potential as a marketable source of plant nutrients. Table VI.9 Analysis of Wastewater from Rotten Wastes, Leachate, and Condensate Ammonia, mg/L Nitrate, m»/L Phosphate, mg/L ,. ,'. pH ECi, ins . Wastewater from rotten wastes 314 .7±53 .7 29.2±7.1 123.8±24.55 7 .15±0.57 2 .63±0 .67 Leachate (mix of Pilot 1,2,3) 376 .0±226 .4 43 .7± 18.1 55.55±21.41 8 .29±0.63 15.85±4.71 Condensate (mix of Pilot 1,2,3) 110.5±72.1 2 .42±2.36 253 .3±115 .4 5 .73±0.52 1.07±0.23 *mean±standard deviation 229 Rotten Waste In most cases, if greenhouses do not treat their wastes properly, for example, stockpiling them at the available land beside the greenhouse, the wastes will be rotten and create undesirable smells and water contamination. Water samples collected from a rotten pile at a greenhouse site were used for a bioassay test, and the results are shown in Figure VI.7. Table VI.9 provides some analysis made on this compared to leachate and condensate. 120 100 80 c (0 | 60 c Ui >s 40 20 0 • %germ| 3 0 O length 25 20 £ 15 2 O) 10 ? rotten waste rotten waste nutrient water distilled water 100% 10%, treatment Figure VI.7 Bioassay on Rotten Greenhouse Waste - 1-day cress seeds These data (Figure VI.7) showed that the "juice" created by the rotten greenhouse wastes, without any dilution, was highly contaminated and created infertility of the land on which they were stored (close to zero germination and growth). Even with 10 times 230 dilution, they only showed a 64% germination rate, compared to 86 % with distilled water and 98% with a nutrient water used in conventional greenhouse operation. Leachate / Condensate The following tests were done to investigate the toxicity of by-product leachate and condensate, the main by-products from an in-vessel composting process, and the dilution at which they could become non-toxic and possibly beneficial to seed growth. Since the leachates and condensates from Pilot 1 to 3 had been collected into the same leachate tank (1200 L tank) and condensation barrel (200 L plastic barrel) and mixed together, they probably represented an average concentration and therefore were used in these tests. Figures VI.8, VI.9 and VI. 10 show the bioassay results with leachate at different dilutions. Figure VI.8 Bioassay on Leachate, 1-day Cress seeds 231 Figure VI.9 Bioassay on Leachate (10%-100%), 6-day tomato seeds Figure VI.10 Bioassay on Leachate (10%-20%), 6-day tomato seeds Undiluted leachate completely suppressed germination of both cress and tomato seeds. Dilution with distilled water improved germination and root length performance for cress seeds. For tomato seeds dilutions to 20% leachate still totally suppressed 232 germination. This could be due to the presence of humic acids in the leachate, which are known to stimulate shoot and root growth, accelerate water uptake, and enhance cell elongation and mobilization of microelements (Inbar, 1992). At 10% dilution of leachate germination of cress seeds and tomato seeds were approximately equal to the control seeds exposed to distilled water or a nutrient solution. For the cress seeds exposed to leachate diluted to 10%, the root lengths were less than those of controls (distilled water and greenhouse nutrient water). For the tomato seeds the root lengths were less than those of the controls but not by much (24mm vs 26mm). Figure V.10 shows a repeat of the test on tomato seeds between 10 and 20 % dilution. It still shows that only 10% dilution is the threshold dilution, with germination rate above 90%. Another set of tests was then done on leachate, diluted to 10%, from the pilot scale composting (pilot tests 1,2,3), to verify the above results. The results are shown in Figure VI. 11 and Figure VI. 12. Figure VI. 11 Bioassay on 10% Leachate from Pilot 1-3 (1-day cress seeds) 233 Figure VI.12 Bioassay on 10% Leachate from Pilot 1-3 (6-day tomato seeds) It can be seen that at 10% dilution, the leachate did not inhibit germination and growth of cress seeds. On the other hand, the tomato seeds improved their growth rates when treated with leachate. At concentrations up to 10%, the tomatoes showed an increase in total length grown while maintaining consistent germination rates of compare with the control values. As the leachate proved beneficial to the tomatoes, there could be a potential to market the diluted leachate. When grown in condensate (mix of Pilot 1, 2, and 3) at various dilutions, a percentage of cress seeds germinated, which was comparable to the control values but somewhat lower. There was a decrease in the length of seed roots as the concentration of condensate increased (see Figure VI. 13). No pattern was observed between the percentage of seeds germinating and the concentration of condensate in which the seeds were grown. 234 Figure VI.13 Bioassay on Condensate, 1-day cress seeds The tomato seeds grown in the presence of condensate showed a more favourable response. Figure VI. 14 shows that the percentage of seeds germinating was fairly high in all the tests with the tomatoes seeds. The best germination (100%) was noted with a 75% condensate concentration. This was even better than the control germination. The length of seed roots was higher than the distilled water control and was similar to the nutrient water control value. From the relative comparison of the nutrient water control with the condensate, it appears that the condensate must contain some nutrients that the tomatoes use. 235 Figure VI.14 Bioassay on Condensate, 6-day tomato seeds Based on the above results, the by-products from the greenhouse wastes composting process could be re-used as fertilizer supplements depending of the application rate and dilutions. An important step in any such reuse plan is a complete analysis of the liquids. The parameters listed in Table VI. 10 profile the nutrients contained in the leachate and condensate (collected from Pilot 1, 2 and 3 as a mix). The leachate has very high nutrient values in macro-nutrients (Potassium, Nitrigen, Phosphorus), and micro-nutrients (calcium, magnesium, zinc, etc). When utilized in a balanced loading, these nutrients could be use as nutrients to enhance plant growth. 236 Table VI. 10 Compost Leachate and Condensate Analysis (from Pilot 1,2, and 3) Analysis: den Haan Hort. Consultancy, NL, via Westgro Sales Inc. Parameter, ppm Leachate (no dilution) C o i K l e i i M i t e (no dilution) Approv Target'-' PH 8.5 5.9 5.5 E C , ms 18.9 0.9 3.7 N 0 3 60 229 1400 C l 920 7 142 so4 2210 10 384 H C 0 3 6770 31 30 P 45 210 39 NH„ 310 97 1.8 K 4730 4 313 Na 340 9 184 C a 20 8 401 M g 290 4 97 Si 28 2.8 0 Fe 4.6 0.01 1.4 M n 0.8 <0.01 0.5 Zn 0.3 0.08 0.5 B 2.4 0.08 0.8 C u 0.1 <0.01 0.04 Mo 0.4 <0.01 0 * Target based on recommended concentrations for greenhouse tomato crop, Westgro Sales. From the analysis of the nutrient contents of leachate and condensate from the greenhouse wastes composting process, quite a few parameters were either too high or too low for the agronomic needs of greenhouse crop. Seed germination bioassay is only a short test to show acute phytotoxicity. Growth trials are still needed to fully understand the effects on plants after applying the compost by-products. However, studies have shown positive results from application of leachate on field crops (Clean Washington Center, 1997). The greenhouse operator can possibly sell these by-products to field crop farmers, especially those organic farmers who are constantly looking for organic nutrient sources. 237 VI.5 CONCLUSIONS Bioassays are a relevant way of proving compost quality. Compost extracts from different batches of compost showed different phytotoxicities and growth rate of seeds, which correlated to the degree of degradation and stability of the finished products. Chemical analysis of compost extract, leachate and condensate showed that they contained a number of valuable nutrients. It was generally accepted that there is an inhibitory effect on germination and early plant growth, as a consequence of using non-stabilized compost, which might contain ammonium and short-chain fatty acids (Murillo, 1995). However, the data presented in this study indicates that there is potential for the reuse of the leachate and condensate from the composting process. Leachate, when diluted to a desired level, could be reused as nutrient supplement for plants. In the case of greenhouse waste, a 10 time dilution was non-toxic or even beneficial in the case of tomato seed. Condensate from composting could be reused for plants without dilutions. If the reuse is possible, costs for disposal of these by-products may be reduced, and they can be instead benefit an end user, e.g. organic farmers. Based on the nutrient content of leachate and condensate and the agronomic needs of the crop, the farmer can calculate a loading rate suitable for the crop. 2 3 8 VI.6 REFERENCES Agassi, M. , Kirsten, F.A., Loock, A.H., Fine, P. (1998) Percolation and leachate composition in a disturbed soil layer mulched with sewage biosolids. Soil and Tillage Research. 45(3-4): 359-372. Baca, M.T. Delgado, I.C., Sanchez-Raya, A.J., Gallardo-Lara, F. (1990) Comparative use of cress seed germination and physiological parameters of Helianthus annuus L. to assess compost maturation. Biological Wastes. 33(4): 251-261. Ball, A.S., Shah, D., Wheatley, C.F. (2000) Assessment of the potential of a novel newspaper/horse manure-based compost. Bioresource Technology. 73(2): 163-167. Bari, Q.H., Koenig, A. (2001) Effect of air recirculation and reuse on composting of organic solid waste. Resources, Conservation and Recylcing. 33(2): 93-111. British Columbia Ministry of Agriculture, Fisheries and Food.(1997) Greenhouse Vegetable Production Guide. Government of British Columbia. Clean Washington Center. (1997) Evaluation of Compost Facility Runoff for Beneficial Reuse. Eghball, B., Gilley, J.E. (1999) Phosphorus and nitrogen in runoff following beef cattle manure or compost application. Journal of Environmental Quality. 28(4): 1201-1210. Fang, M. , Wong, J.W.C. (1999) Effects of lime amendment on availability of heavy metal and maturation in sewage sludge composting. Environmental Pollution. 106(1) : 83-89. Gaur, A.C. (1987) Recycling of organic wastes by improved techniques of composting and other methods. Resources and Conservation. 13(2-4): 157-174. 239 Grobe, K. (1998) Fine-Tuning the Soil Food Web. BioCycle. 39(l):42-46. Haug, Roger T. (1993) The Practical Guide to Compost Engineering. Boca Raton: Lewis. Hoitink, H. A., Zhang, W., Han, D. Y., and Dick, W. A. (1997) Making Compost to Suppress Plant Disease. Biocycle 38(4):40-42. Inbar, Y., Hadar, Y., Chen, Y. (1992) Characterization of humic substances formed during the composting of solid wastes from wineries. The Science of The Total Environment. 113(1-2) :35-48. Ingham, E. (1999) Making a High Quality Compost Tea. Biocycle. 40(4): 94. Jimenez, E.I., and Garcia, V.P. (1989) Evaluation of City Refuse Compost Maturity : A Review. Biological Wastes 27:115-142. Johnson, G., Crawford, S. Sark, S.A. (1993) Sampling Municipal Soild Waste Compost. Biocycle. December issue : 61-64. Kai, H., Ueda, T., and Sakaguchi, M . (1990) Antimicrobial Activity of Bark-Compost Extracts. Soil Biol. Biochem 22(7):983-986. Kaschl, A., Romheld, V., Chen, Y. (2002) The influence of soluble organic matter from municipal soild waste compost on trace metal leaching in calcareous soils. The Science of Total Environment. 291 (1 -3): 45-47. Ketterer, N. , Fisher, B., and Weltzien, H. C. (1992) Biological Control of Botrytis cinerea on Grapevine by Compost Extracts and their Microorganisms in Pure Culture. Recent 240 Advances in Botrytis Research: Proceedings of the 10th International Botrytis Symposium, Heraklion, Crete, Greece. Krapac, I.G., Dey, W.S., Roy, W.R. Smyth, C.A., Storment, E., Sargent, S.L. Stelle, J.D. (2002) Impacts of swine manure pits on groundwater quality. Environmental Pollution. 120(2) : 475-492. Mathur, S. P. (1996) The Use of Compost as a Greenhouse Growth Media. Waste Reduction Branch, Ontario Ministry of Environment and Energy, Toronto, ON. McQuilken, M . P., Whipps, J. M. , and Lynch, J. M . (1994) Effects of Water Extracts of a Composted Manure-Straw Mixture on the Plant Pathogen Botrytis cinerea. World Journal of Microbiology & Biotechnology. 10:20-26. Murillo, J.M., Cabrera, F., Lopez, R., Martin-Olmedo, P. (1995) Testing low quality urban composts for agriculture : germination and seedling performance of plants. Agriculture, Ecosystems & Environment. 54(1-2): 127-135. Peot, C. (1997) A marketable product from runoff. Biocycle. 38(9): 60. Sanchez-Monedero, M.A., Roig, A., Cegarra, J., Bernal, M.P. (1999) Relationship between water-soluble carbonhydrate and phenol fractions and the humification indices of different organic wastes during composting. Bioresource Technology. 70(2): 193-201. Soumare, M . Tack, F.M.G., Verloo, M.G. (2003) Characterisation of Malian and Belgium solid waste composts with respect to fertility and suitability for land application. Waste Management. 23(6): 517-522. Tiquia, S.M., Tarn, N.F.Y., Hodgkiss, I.J. (1996) Effects of composting on phytotoxicity of spent pig-manure sawdust litter. Environmental Pollution. 93(3): 249-256. r 241 Tiquia, S.M., Wan, J., Tarn, N.F.Y. (2002) Dynamics of yard trimming composting as determined by dehydrogenase activity, ATP content, arginine ammonification, and nitrification potential. Process Biochemistry. 37(10): 1057-1065. Tyler, R. (2000) Managing corrosion inside composting buildings. Biocycle. 41(4) : 62-64. Vogtmann, H., Matthies, K., Kehres, B., and Meier-Ploeger, A. (1993) Enhanced Food Quality: Effects of Composts on the Quality of Plant Foods. Compost Science and Utilization. 1(1):82-100. Wong, J.W.C., Fang, L.K., Su, M . (2001) Toxicity evaluation of sewage sludges in Hong Kong. Environ. Intern. 27(5): 373-380. Yun, Y.S., Park, J.I., Suh, M.S., Park, J.M. (2000) Treatment of food wastes using slurry-phase decomposition. Bioresource Technology. 73(1): 21-27. Zucconi, F., Forte, M. , Monaco, A. De Bertoldi, M . (1981) Biological evaluation of compost maturity. Biocycle. July/August :27-29. 242 CHAPTER VII UTILIZATION OF COMPOST VII.l BACKGROUND AND LITERATURE RESEARCH On-site composting and re-use of greenhouse waste as a growing medium amendment provides an end use for a waste that might otherwise require disposal, thereby reducing trucking costs and associated environmental impacts, and reducing the reliance of growers on externally supplied growing medium materials (Cheuk, 2002). Currently, most BC vegetable greenhouses use yellow cedar or hemlock sawdust, which are in limited supply, as growing medium. In addition, utilisation of compost in growing media potentially offers significant benefits to growers in terms of disease suppression (Szczech, 1993; Zhang, 1996; Hoitink, 1997). Concurrent with this study, a disease suppression study carried out at the University of BC using the same amendments demonstrated significant yield improvement through reduction of soil-borne crown rot disease caused by Fusarium oxysporum on greenhouse tomatoes (Cheuk, 2003). Suppressing disease by utilizing organic amendments would reduce dependence on chemical disease control. 243 VII.2 SPECIFIC OBJECTIVES The overall objective of this part of the study was to evaluate the feasibility of utilising high-quality compost manufactured from greenhouse wastes as an amendment or replacement to sawdust growing medium for greenhouse tomatoes. VII.3 METHODOLOGY The study comprised two main elements: first, comparing the characteristics of the compost amendment and conventional sawdust growing medium; and second, demonstration of the feasibility of incorporating the compost amendment in a commercial greenhouse operation. VII.3.1 GROWING MEDIA ANALYSIS Compost of Pilot 4 test (Chapter V.4.2), produced from a mixture of waste tomato leaves and culled fruit, red alder tree bark mulch, and used yellow cedar sawdust growing medium, was used throughout this study. The materials were composted using a pilot-scale, in-vessel composting system. The materials were composted for approximately 30 days in the container, cured for several months, and screened to V-i\ Maximum temperature in the composting vessel was approximately 65 °C, and the composting process had passed the PFRP requirements (Process to Further Reduce Pathogens) of 55 °C for 3 days. 244 Prior to starting the growing trials, the physical, the chemical and microbiological characteristics of the sawdust, compost amendment (from Pilot 4), and sawdust plus amendment mixture (2:1 by volume) were measured using the methods described in Table VII. 1. Measurements were conducted both before and after the season. Table VII. 1 Growing Media Analysis Methods Test Laboratory Method Moisture U B C Oven-drying gravimetric (Amer. Soc. Agron., 1982) Bulk density U B C Gravimetric/volume estimation (Kasica, 1997) Particle size - pre-season U B C Manual dry sieving Porosity - pre-season (aeration porosity and water holding porosity) U B C Gravimetric water saturation and drainage using bulk materials (Kasica, 1997) Porosity - post-season (water holding porosity) Soilcon Laboratories (Richmond, BC) In-situ samples from 2.5cm depth analysed using desorption from saturation under 10 kPa Total nitrogen U B C Ignition at 950 °C in Leco FP228 nitrogen determinator Total organic carbon U B C Combustion at 680 °C in Shimadzu total organic carbon analyser with solid sampling module Nutrients Norwest Labs (Langley, BC) CMPT-Turf C E C Norwest Labs CL11 pH, electrical conductivity (EC) U B C lOx dilution distilled water extraction (shaken and centrifuged) Total bacteria and fungi Cantest Laboratories (Burnaby, BC) Bacteria: standard plate count for solid materials Yeast and mold: analysis in solid samples -peptone water rinse, PDA medium 245 VII.3.2 GROWING TRIAL The growing trial was conducted in the test greenhouse operated at Hazelmere Greenhouse, Surrey, BC. Seedlings were planted on 25 February 1999 (week 8) and removed at the end of the season on 10 December 1999 (week 49). The test greenhouse environment, covering 153 m , was similar to the commercial greenhouse, with computer control of climate and irrigation. The growing medium was enclosed in cylindrical polyethylene bags of approximately 18 litres each; seedlings were rooted in rock wool blocks, placed on the growing medium through an opening in the bag. Four main treatments were used, testing the two media and nutrient solutions as shown in Table VII.2. Each treatment was divided into 3 north-south rows that were interspersed as evenly as possible in the east-west direction. Two more treatments using pure amendment were included, each using only a single end row. "Mississippi" beefsteak tomato cultivar was used, with three plants per bag, and each plant double-headed on a " V " crop wire system. Figures VII. 1 and VII.2 show the experimental layout. Table VII.2 Growing trial treatments - medium, nutrient solution & number of rows Treatment Growing medium Nutrient No. 'o f rows; , No. of bags No. (if solution ...plants 1 Sawdust NI 3 24 72 2 Sawdust N2 3 24 72 3 Sawdust + amendment 2:1 v/v NI 3 24 72 4 Sawdust + amendment 2:1 v/v N2 3 24 72 5 100% Amendment N2 1 8 24 6 100% Amendment N3 1 8 24 Total 14 110 336 246 Nutrient solution 1 was intended to be as close to a typical commercial feed as possible and was based on the one used by the grower at Hazelmere Greenhouse. A sample recipe is shown in Appendix F. As a recommendation from the professional grower, another recipe (nutrient solution 2) with higher ammonium concentration (by adding ammonium nitrate) was used. According to the grower, the addition of ammonia would possibly increase the yield on tomatoes; however, it would create problems with high EC and low pH. Nutrient solution 3 was the same recipe as nutrient solution 2, with a lower EC to compensate for the higher drain EC expected from the pure amendment. The ECs of all the nutrient solutions were controlled by an irrigation control unit (PRIVA) through different ratios of dilution. Table VII.3 shows that nutrient solution 3 always had a lower average EC than both the conventional solution and nutrient solution 2. Nutrient solutions 2 and 3 had a higher percentage of ammonia. The irrigation feed and drain streams were monitored and analyzed regularly. The nutrient recipes, volume, and EC of the feed were adjusted where necessary to maintain drain parameters within acceptable ranges. Table VII.3 Growing Trial Nutrient Solutions - EC, Recipe And Ammonia Concentration Nutrient Solution Average E C Average ammonia as % of nitrogen ' N l - Conventional 3.2 4.3 N2 - Modified 3.1 6.4 N3-Modi f i ed 2.8 6.4 247 Irrigation Tank ZONE 1 ZONE 2 Entrance ZONE 3 ZONE 4 9 I I 3 N3 4 NI 5 NI 6 N2 7N2 8 NI 9 NI 10 N2 11 N2 12 NI 13 NI 14 N2 15 N2 16 N2 1A 3A 4A 2A IB 3B 4B 2B 1C 3C 4C 2C Rows 3-16: 7.3m (8 bags x 3 plants/bag) Figure VII. 1 Test Greenhouse Layout 248 Figure VII.2 Test Greenhouse for Growth trial Tomato yield was recorded per row, after harvesting one to three times per week depending on the season. Changes in the number of plants due to human intervention (for example breakage due to lowering) were recorded, and the actual yields corrected. Plants lost to disease were recorded separately and not included in yield corrections, with the exception of three plants that were removed shortly after planting due to a viral infection that likely occurred prior to planting. Plant growth parameters of shoot height, leaf length, and stem diameter were monitored on one plant (two shoots) in each row, weekly through the season and they are shown in Appendix G. 249 VII.3.3 FRUIT QUALITY A tomato shelf life test was also performed at B C Hot House Quality Assurance also in June, 1999. Three, approximately 5 kg, composite samples of similarly sized and ripened tomatoes were sampled from each row, and grouped according to growing medium. The tests performed included observation of colour development, tray weight, calyx condition, firmness, wrinkles, soft spots, and mold or rot, and were conducted in both laboratory (18-19 °C) and warehouse (12—13 °C) conditions. Each test was conducted after 1, 5, 8, and 14 days of storage. STATISTICAL ANALYSIS Statistical data analysis was performed using SPSS 7.5 for Windows software, including ANOVA in Appendix G (analysis of variance) to compare data means. 250 VII.4 RESULTS AND DISCUSSIONS VII.4.1 GROWING MEDIA TESTS VII. 4.1.1 Pre-Season Media Analysis Density and porosity characteristics of sawdust and compost amendment are compared in Table VII.4. The compost amendment showed significantly higher bulk density, lower total and aeration porosity, and higher water holding porosity than sawdust; the sawdust-amendment mixture displayed characteristics generally between the two. Table VTI.4 Pre-Season Physical Characteristics Of The Growing Media Media status Parameter • Unit Sawdust Sawdust + , •: amendment Amendment. . Moist Bulk density kgm"3 409 476 587 Total porosity % 63.4 58.6 49.7 Aeration porosity % 46.8 35.8 22.8 Water holding porosity % 16.6 22.8 26.9 Air-dried Bulk density % 171 200 332 Total porosity % 76.0 72.4 74.1 Aeration porosity % 39.4 36.1 28.5 Water holding porosity % 36.6 36.3 45.6 Table VII.5 shows the particle sizes of different media and Figure VII.3 shows the distribution. The sawdust was the ordinary growing media for greenhouses. The compost was from Pilot scale test 4 and a commercial compost from HOME DEPOT (Envirowaste 251 Mushroom compost) was used for comparison purpose. The amendment (compost) has higher percentage of large particles (>4.8), which provided better structure for the media. Table VII.5 Particle size of different media (Pre-season) Mesh (mm) Pilot 4 compost Commercial Compost Sawdust Sawdust + Pilot 4 compost Mass (g) 4.8 80 50 45 40 2.4 110 95 135 110 1.0 90 150 120 145 0 30 10 10 20 Total 310 305 310 315 Percentage (%) Fraction Pilot 4 compost Commercial Compost Sawdust Sawdust + Pilot 4 compost >4.8 26 16 15 13 4.8-2.4 35 31 44 35 2.4-1.0 1029 49 39 46 1.0-0 10 3 3 6 Total 100 100 100 100 >4.8 4 .8-2 .4 2 .4 -1 . 0 1 .0 -0 mm Figure VII.3 Particle size distribution of different growing media 252 Tables VII.6 and VII.7 show the media nutrient and chemical characteristics. The compost amendment was relatively rich in macro- and micro-nutrients, most of which were 10 to 20 times higher in concentration compared to sawdust. EC and pH were also higher than sawdust, as expected; however they were both within an acceptable range for use as a growing medium. Cation Exchange Capacity (CEC) was significantly higher for the amendment compared to sawdust. The compost amendment had a much lower C:N ratio than sawdust. Table VII.6 Pre-Season Nutrient Characteristics Of The Growing Media Materials Measurement ' Unit Sawdust Amendment C E C meq lOOg"1 8.2 127.1 Sodium-Na ppm 80 2460 Ammonia-N ppm 0 25.8 Nitrate-N ppm 8.4 94 Phosphate-P ppm 20 716 Potassium-K ppm 190 14200 Calcium-Ca ppm 1300 12000 Magnesium-Mg ppm 100 2400 Sulphate-S ppm 296 1626 Iron-Fe ppm 20 560 Manganese-Mn ppm 6.3 149 Zinc - Zn ppm 0 35.4 Copper - Cu ppm 0 2.6 Chloride-Cl Hgg"1 44 4800 253 Table VII.7 Pre-season chemical characteristics of the growing media materials Medium Moisture (% wet basis) Total organic carbon (% dry basis) Total nitrogen (%dry basis) E C ] • (mS/cm' ! I'll ';" C/N ratio Mean (n=3) Std. dev. Mean (n=3) Std. dev. Mean (n=3) Std. dev. Mean (n=3) Std. dev. Mean (n=3) Std. dev. Sawdust 65.9 0.2 52.4 0.6 0.7 0.1 <0.1 0.02 6.2 0.2 74.9 Sawdust + compost 65.1 1.0 49.3 0.6 0.8 0.1 0.2 0.03 6.8 0.2 61.6 Compost 63.8 2.8 43.7 1.4 2.0 0.3 0.9 0.0 7.3 0.2 21.9 Another batch of sample from Pilot 4 compost was sent to a commercial laboratory for bacteria and fungi counts as shown in Table VII.8. The microbial counts indicate an approximately 50 times higher bacterial population for the amendment compared to the sawdust, while the amendment fungal populations (as reflected by yeast and mold counts) were slightly lower. The resulting bacteria to fungi ratio was approximately 100 times higher for the amendment. This higher bacterial population and bacteria to fungi ratio of the amendment compared to sawdust may contribute to disease suppression capability - for vegetable crops, a high ratio of bacteria to fungi in the medium is beneficial (Kai, 1990). However, in this study, there was no evidence of significant soil-borne disease; therefore no yield improvements due to disease suppression would be expected. Table VII.8 Microbial counts of compost amendment Med iu iii Bacteria C F U g ' \ east C F U g 1 Mold C F U g ' Bacteria/fungi ratio (calculated), ; Sawdust 8x 104 1.6 x 104 7x 102 5 Amendment (Pilot 4) 4.5 x 106 2.1 x 103 7.1 x 103 489 254 VII.4.1.2 Post-Season Media Analysis Post-season analysis results of the media used in the growing trial are shown in Tables VII.9 and VII. 10. For the post-season density and porosity, the commercial laboratory (Soilcon) had the media air dried before analysis. The bulk density of the post season sawdust, mix and amendment were all much less than the pre-season values. Since there was not much change in volume (the plastics bag were still full of media), there must be some weight loss due to degradation of the media (loss of carbon) during the growing season. Nevertheless, the post season amendment's bulk density was greater than that of the mix which, in turn, was greater than that of the sawdust. See Tables VII.4 and VII.9. At season-end, the sawdust had the lowest water holding porosity and lowest total porosity, those of the sawdust/compost mix and compost were higher. This shows the amendment media would have better water retention ability than sawdust even at the end of the growing season. Table VII.9 Post-season media density and porosity 1 ..Parameter Sawdust . Sawdust|+ compost Compost Mean Range Mean Range Mean Range Bulk density (kg m-3) 144.0 11.6 183.1 9.6 209.9 17.5 Total Porosity % 63.9 2.9 73.8 1.3 79.0 1.8 Aeration Porosity % 28.1 2.5 20.7 4.7 25.5 0.9 Water Holding Porosity 35.8 0.6 53.0 3.3 53.4 0.8 (% by volume) Mean and range based on n=2 samples. 255 Table VII. 10 Post-season media chemical characteristics Growing medium Nutrient solution Moisture (%) Total organic carbon (%) Total nitrogen (%) E C • /,'< (mS cm-l) p H , ''i'!'•.•;'"',':•.;'. C/N ratio; Mean (n=3) Std. dev. Mean (n=2) Range Mean (n=3) Std. dev. (n=l) (n=l) Sawdust NI 75.3 1.4 50.8 0.1 1.41 0.06 0.70 6.3 36.1 Sawdust N2 79.9 0.6 49.7 0.1 1.59 0.09 0.50 6.4 31.2 Sawdust + compost NI 78.2 0.5 44.8 1.6 2.13 0.17 0.80 6.7 21.1 Sawdust + compost N2 78.3 0.3 42.4 0.8 2.56 0.11 0.90 6.7 16.6 Compost N2 78.3 0.6 42.3 3.0 2.38 0.26 0.50 6.3 17.7 Compost N3 79.4 0.3 38.8 1.6 2.78 0.05 0.40 6.8 13.9 The post-season moisture contents were all higher than the pre-season since the media had absorbed a lot of nutrient solution. The total organic carbon had decreased which confirmed there must be some loss of carbon during the season as discussed above. Also, since there were losses of carbon and absorption of nitrogen from the nutrient solution, the C:N ratio would decrease. Over the season, the carbon to nitrogen (C/N) ratio of the sawdust medium decreased by approximately 57 %, the mixture approximately 69%, and the amendment approximately 28%. A white fungus was also observed immediately under the rock wool block and on the surface of the medium in many of the sawdust-amendment mix bags. A few of the sawdust bags had smaller amounts, and virtually none was observed in the pure amendment. The fungus appeared to be more common in Treatment 4, bags which had higher feed ammonia than Treatment 3. Analyses at government and commercial laboratories both identified the fungus as non-pathogenic and saprophytic. 256 VII.4.2 GROWING TRIAL VIL4.2.1 Fruit Yield The total marketable yield for all treatments harvested was 7642 kg, or, on an area basis 49.8 kgm"2. Table VII. 11 compares the marketable yield, culls, and average fruit size for all treatments. Table VII. 11 Tomato yield, size and culls for the trial growing season Nutrient solution Medium Corrected marketable yield - '(kg).' . Average fruit size ' (Si) Culls (%) Mean (n=3) Std. dev. Mean (n=3) Std. dev. Mean (n=3) Std. dev. N l Sawdust 51.4 (A) 2.0 173.6 (A) 3.1 8.6 (A) 3.2 N l Sawdust + compost 52.6 (A) 1.0 177.7(A) 1.8 6.5 (A) 1.5 . N2 Sawdust 52.5 (A) 0.8 176.3 (A) 3.8 7.9 (A) 0.5 N2 Sawdust + Compost 47.7 (B) 1.5 170.2(A) 1.0 10.2(A) 1.4 N2 Compost 43.7 n/a 171.1 n/a 10.4 n/a N3 Compost 53.3 n/a 186.2 n/a 9.7 n/a (A), (B) indicate which treatment means are statistically the same(A) or different (B) for p < 0.05. For conventional feed, differences in yield between sawdust and sawdust-amendment mix were not significant, while for modified feed (higher ammonia N2), the sawdust-amendment mix and compost had a 9% and 17% lower yield respectively (significant for PO.05). This is could be due to the high EC in the medium caused by both the compost and additional ammonium nitrate. The yield in the EC-reduced compost media achieved the highest yield. It has to be noted that each row's yield data 257 were collected from 24 plants. So, even though treatment 6 (compost, N3) had only one row it's yield data represented the total yield of 24 plants. The comparison of the 100% compost amendment treatments (treatment 5 and 6) is as representative as the other 4 treatment since they were single rows on the periphery of the greenhouse; however yield results are encouraging in that Treatment 6 showed a larger fruit size in addition to higher yield. VI1.4.2.2 Fruit Quality Shelf life results are shown in Table VII. 12. The conclusion drawn by the shelf life tester (the quality supervisor at BC Hothouse Food Inc.) was that there was very little difference between the groups, and all groups showed acceptable quality. The data also suggest that increased use of amendment might have contributed to slight increases in softness and weight loss during storage; however further study would be required to confirm this. Table VII. 12 Tomato shelf life results C o l o u r stage t iraj weigfct loss C a l w condition Firmness (hand) Suga rs Number \\ rinkled N II in her with soft spots N u m b e r with mold/rot Range/Unit 1 - 12 % 1-5 0 - 5 1 -6 - - -Key 1 -Green - 1-Fresh 0-Hard 1 -Lowest - - -Sawdust 8.38 3.9% 3.25 0.75 5 0 0.75 0 Sawdust+ Amendment 8.25 4.7% 3.25 0.88 5 0 0.88 0 Amendment 8.38 5.3% 3.25 1.00 5 0 1.13 0 258 VJI.4.2.3 Plant Growth Results for cumulative plant growth, leaf length, and stem diameter were analyzed. Statistical analysis of Student's t test of the growth results did not show any significant differences between the treatments (PO.05), and therefore they are not further discussed here. The results can be found in Appendix G. VII.4.2.4 Disease Botrytis stem rot, with only a few exceptions, was the cause of losses due to disease. The relatively high losses largely occurred near the end of the season (approximately half in the last 5 weeks); the losses were consistent with the outbreaks in the main greenhouse to which the test greenhouse is attached. Although there were differences in the number of diseased plants between treatments, they were not statistically significant. This suggests that the growing media and nutrient combinations tested did not impact Botrytis development. VII. 4.2.5 Medium A cidity The modified feed with increased ammonium concentration depressed yield slightly. This may have been due to excessive amounts of ammonia nitrogen (up to 12% of TN) that were used during a period in June. In that period, the grower recommended to increase the ammonia concentration in order to see whether that could cause any yield increase (Cheuk, 2003b). Unfortunately, because of that, the pH in the media (drain) 259 dropped to an undesirable level. The minimum recorded drain pH was 4.2 for Treatment 2 (sawdust), and 5.0 for Treatment 4 (sawdust-amendment mix), both using the same feed. Subsequent to the minimum pH occurring in the sawdust medium, bicarbonate buffering was added to all nutrient solutions to maintain a minimum feed pH. In that period the sawdust medium fluctuated between pH 4.2 to 6.1 (Range=1.9) while the amendment fluctuated between pH 5.0 to 6.0. This showed that the amendment mix demonstrated a significant buffering capability (see Figure VII.4). This agrees with other research works done on buffering capacity of compost (Stamatiadis, 1999; Shojaosadati, 1999; Jakobsen, 1996). Stabilization of pH enhances the nutrient uptake by plants and root health. Figure VII.4 Drain pH comparison of Treatments 2 (sawdust) and 4 (sawdust plus amendment), both using nutrient solution N2 (higher ammonia). 260 VII.5 CONCLUSIONS There are likely several factors related to the growing medium that may have contributed to the yield positively or negatively. Positive effects provided by amendment may include nutrient retention (as reflected by CEC), additional nutrients or other compounds, and effects of beneficial microorganism populations for disease suppression. Negative factors may include lower porosity and oxygen diffusion, and excessive moisture retention. Further optimization of fertilizer and irrigation schedules to take into account the different physical and chemical medium characteristics could potentially increase yield, or alternatively allow a larger proportion of amendment in the medium mix. The combination of increased ammonium concentration with compost amendment had a negative effect on yield, suggesting an interaction between the feed composition and the medium; this may be related to the white fungus observed, but could not be confirmed. However, addition of the compost amendment to the conventional sawdust growing medium can provide mitigation of excessively low pH. Also, the positive results from the 100% compost with lower EC should encourage further research on the optimum use of compost. This study found that compost manufactured from greenhouse crop waste in a controlled process can be suitable for use as a growing medium amendment in soil-less tomato greenhouses! Using conventional management techniques, a similar yield could be 261 achieved compared to conventional sawdust medium using a 2:1 sawdust to amendment mix by volume. By demonstrating the feasibility of compost as a growing medium amendment in a commercial setting, the results of this study could form the first step in moving soil-less vegetable greenhouses in BC toward more sustainable waste management and growing practices. 262 VII.6 REFERENCES American Society of Agronomy. (1982) Methods of Soil Analysis, 2nd edn. Soil Science Society of America, Madison, Wis., U.S.A. Cheuk, W., Fraser, B., Lau, A. (2002) On-site composting of greenhouse crop residuals. Biocyle. 43(10): 32-34. Cheuk, W., Lo, K.V., Branion, R., Fraser, B., Copeman, B., Jolliffe, P. (2003) Applying Compost to Suppress Tomato Disease. Biocycle. 44(1) 50-51. Cheuk, W., Lo, K.V., Fraser, B. (2003b) Use of composted greenhouse waste as a growing medium component will contribute to a sustainable waste management solution for vegetable greenhouses. Journal of Horticulture, (accept Aug 2003) Jakobsen, S.T. (1996) Leaching of nutrients from pots with and without applied compost. Resources, Conservation and Recycling. 17(1) : 1-11. Kai, H., Ueda, T., and Sakaguchi, M . Antimicrobial Activity of Bark-Compost Extracts. Soil Biol. Biochem 22(7):983-986. 1990. Kasica, A. F. & Good, G. L. (1997) Media: Rooted in Success. From Cornell University Dept. of Floriculture and Ornamental Horticulture (Ithaca, NY, U.S.A.) web site: http://www.hort.comell.edu/department/faculty/good/growon/media/ Hoitink, H. A., Zhang, W., Han, D. Y. & Dick, W. A. (1997) Making Compost to Suppress Plant Disease. Biocycle, 38(4): 40-42. Portree, J. (1996) Greenhouse Vegetable Production Guide for Commercial Growers. BC Ministry of Agriculture, Fisheries and Food, BC, Canada. 263 Shojaosadati, S.A., Elyasi, S. (1999) Removal of hydrogen sulfide by the compost biofilter with sludge of leather industry. Resources, Conservation and Recycling. 27(1-2) : 139-144. Stamatiadis, S., Werner, M. , Buchanan, M . (1999) Field assessment of soil quality as affected by compost and fertilizer application in broccoli filed (San Benito County, California). Applied Soil Ecology. 12(3) 217-225. Szczech, M. , Rondomanski, W., Brzeski, M . W., Smolinska, U. & Kotowski, J. F. (1993) Suppressive Effect of a Commercial Earthworm Compost on Some Root Infecting Pathogens of Cabbage and Tomato. Biological Agriculture and Horticulture, 10 : 47-52. Zhang, W., Dick, W. A. & Hoitink, H. A. (1996) Compost-Induced Systemic Acquired Resistance in Cucumber to Pythium Root Rot and Anthracnose. Phytopathology, 86 : 1066-1070. Special Acknowledgements This project was implemented pursuant to the British Columbia Investment Agriculture Foundation Program, funding for which is provided through Agriculture and Agri-Food Canada's (AAFC) Safety Net Program and Canadian Adaptation and Rural Development Fund. The authors also thank Houweling Nurseries Ltd., Delta, B C for tomato seedlings, materials and time; Mr. Bud Fraser and Mr. Raymond Wong for research assistance and field crop management; Mr. Kevin Haddrell and Ms. Jennifer Miller at U B C for their dedication; and Mr. Jim Portree of the B.C. Ministry of Agriculture and Food for his support. 264 CHAPTER VIII DISEASE SUPPRESSION TRIALS V I I I . l B A C K G R O U N D A N D L I T E R A T U R E R E S E A R C H Disease is an ongoing problem that continues to impact greenhouse productivity and product quality. Historically, diseases such as Fusarium crown rot in greenhouse tomatoes have accounted for significant crop losses, in some cases as high as 30 to 50%. Pathogenic organisms continue to change and adapt to new cultivars and controls, developing resistance to chemical agents. These pathogen adaptations mean that plant resistance genes often give protection only for a few years (Campbell, 1989). Positive disease suppression results with compost products in terms of disease suppression have been observed in recent studies. Pythium root rot in cucumbers grown in peat was decreased when the peat was amended with a composted pine bark mix (Zhang, 1996). A floriculture study (Mathur, 1996) concluded that good quality compost can replace peat, increase nutrient availability, and increase plant growth and survival, though many of the commercial composts were found to be unsatisfactory due to poor quality. Compost can suppress disease in plants via a number of mechanisms, including competition, antibiosis, hyperparasitism, and induction of systemic acquired resistance (SAR) in some host plants (Hoitink, 1997). The effectiveness of the compost amendment is largely determined by the feedstock materials and the method of processing and 265 storage, as well as the type of crop. As previously reported (Mathur, 1996), compost can be phytotoxic if it is not processed or applied correctly. Parameters including moisture, pH, carbon to nitrogen ratio, and maturity impact the effectiveness of disease suppression (Hoitink et al., 1997). In addition, selection of appropriate feedstocks and processing methods can help optimize the microbiological makeup of the product for the particular crop in order to give the best results (Grobe, 1998). A high bacteria to fungi ratio has also been suggested as a desirable characteristic for increasing disease suppression (Kai, 1990). VIII.2 SPECIFIC OBJECTIVES As part of a larger project evaluating the feasibility of utilising high-quality compost amendment manufactured from greenhouse wastes as an amendment to soil-less growing medium for greenhouse tomatoes, this part of the study focuses on the disease suppressive abilities of the compost amendment. In the Lower Mainland of British Columbia, most vegetable greenhouses currently use yellow cedar sawdust as the growing medium. The specific objectives are : • Find out whether compost has any disease suppressive abilities on seedlings, and if so, find out what the best way of application; • Find out whether compost has any disease suppressive abilities on adult plants, and if so, investigate the improvement in plant health and yield. 266 VIII.3 METHODOLOGY The study comprised two main components - seedling growth tests, and a yield test. In both cases, FORL spores were introduced into the growing media, and the plants monitored for growth, health, and yield parameters, depending on the test. Media treatments compared the conventional growing medium (yellow cedar sawdust) with a medium amended or replaced with compost. Compost amendment was produced from a mixture of greenhouse waste materials and other amendments, as shown in Table VIII.l. The materials were composted using a pilot-scale, in-vessel composting system (indicated by "P" in the batch number) which was designed by the author for greenhouse use, and a laboratory scale system (indicated by " L " in the batch number), depending on the batch. Materials were composted for 3-4 weeks in the vessel, then cured for several months, and screened to Yz". The composting system provided controlled environmental conditions designed to ensure sufficient reduction or elimination of potential pathogens in the raw materials (BC MWLAP, 2002). Maximum temperature in the composting vessels was approximately 65 °C. The controlled process used in the laboratory scale vessels was similar to that of the pilot-scale vessel (see Chapter V). 267 Table VIII.l Recipes and dates of compost amendment batches .Amendment batch .number ' ' • • .' .;' , Recipe ' ,' , :: : P4 Tomato leaves and cull fruit Alder bark hog fuel Used sawdust medium P6 Same as P4 P7 Shredded whole tomato plants Used sawdust medium P8 Tomato leaves and cull fruit Hemlock bark mulch Used sawdust medium 5A Tomato leaves and cull fruit Hemlock bark mulch Recycled compost coarse fragment Sawdust 5C Tomato leaves and cull fruit Alder bark hog fuel Recycled compost coarse fragment Sawdust VIII.3.1 TOMATO SEEDLING TESTS Seedling propagation generally followed established greenhouse methods (Portree, 1996) as follows. Rock wool plugs were pre-soaked and flushed twice over 48 hours with EC=0.5 mS/cm nutrient solution. Rock wool plugs and amendment plugs were then seeded and the seeds were covered with either vermiculite or amendment, and the trays were covered until 90% of the seeds had germinated. Aliquots (1 ml) of spore suspensions (2x10 spores/ml of Fusarium oxysporum f. sp. radicis- lycopersici (FORL) Race V) were then inoculated into selected plugs. Nutrient solution EC was increased to 3 Isolated from diseased Dombito plants in 1985 268 1.5 mS/cm after inoculation, then 2.5 to 3.0 mS/cm after one more week. Seedlings were removed and evaluated according to a scoring system for root disease after several weeks. Al l tests used Dombito (a FORL susceptible greenhouse beefsteak tomato) cultivar. To construct plugs from amendment, rock wool plugs were cut into three cylindrical wafers. One wafer was placed in the bottom of each hole in the plug tray; amendment was then added on top of the rock wool, filling the hole 4/5 full. Seeds placed on the amendment were then covered with either vermiculite or amendment. In later tests, a "sandwich" method was tested, where %" of the rock wool plug was cut from the top and replaced with amendment; this results in the seed being sandwiched between two layers of amendment. Plug trays were separated into small groups, and placed in 25 cm plastic trays, to avoid cross-contamination between groups. VIII.3.2 PREPARATION OF FORL SPORES Two week old FORL cultures grown on potato dextrose agar (PDA) at 20 °C were used. Sterile distilled water (10 ml) was added to each culture. A sterile glass rod was rubbed lightly over the fungal colonies to release the spores. The spore suspension was collected, filtered through cheesecloth, diluted and the spore concentration was determined by a haemocytometer. This suspension was then further diluted to the levels used in the individual experiments. VIII.3.3 SEEDLING SCORING The following parameters were recorded for scoring of tomato seedlings. 269 1. Plant height: root crown to plant apex. 2. Stem diameter: measured approximately 2 cm above crown. 3. Leaf colour: three measurements were taken on upper leaves of each plant using a Minolta SPAD-502 chlorophyll meter. 4. Root health : a visual observation to reflect the overall root health, recorded as 1 to 3, corresponding to good, fair and poor. However, this parameter was not used for discussion (only shown in Appendix K) since it is not as systematic and objective as the root disease score. 5. Root disease score: a visual scoring system reflecting the severity of crown and root rot symptoms on each plant, as shown in Table VIII.2. Table VIII.2 Seedling test root scoring Score . Interpretation; 1 White, healthy roots; no sign of disease 2 Slight browning of vascular tissue in the crown 3 Extensive browning of vascular tissue in the crown 4 Extensive necrosis of the crown 5 Dead plant with typical symptoms of crown and root rot Six individual tomato seedling disease tests were conducted. The first two were "dry runs" to help establish the method. To compare treatments in terms of disease, a significant amount of disease must be created, at least in some of the plants. In Tests 3, 4 and 6, disease was evident in the plant roots; in Test 5 no disease was observed in any plants, thus the results of this test have not been included in the analysis. The time 270 required to develop disease, and the growth rate of the plants, also varied somewhat from test to test. Note that the number of seedlings scored was usually less than the starting number of replicates planted, due to seeds that did not germinate or develop. For Tests 4 and 6,10 extra replicates were provided in the inoculated control group; these were used to monitor disease development prior to full scoring. Treatments and experimental conditions for tests 3, 4 and 6 are shown in Tables VIII.3, VIII.4 and VIH.5. Table VIII.3 Treatments and conditions for seedling test 3 - amendment as plug and covering Location U B C Pathology greenhouse Temperature 18-25 °C Lighting 18 hour photoperiod, sodium vapour lamps Inoculation time (days from seeding) 4 Scoring times (days from seeding) and number of replicates per scoring per treatment 21,28,35, 42, 50 days 5-6 replicates each Tray plug type Keem plug (2.5 cm dia. cylinder) Trays (blocks) per treatment 2 Treatments No. Medium Covering Nutrient solution Inoculation No. of replications scored* 1 Rock wool Vermiculite Nutrient None 15 2 Rock wool Vermiculite Nutrient FORL 15 3 Rock wool Amendment P6 Nutrient None 15 4 Rock wool Amendment P6 Nutrient FORL 15 5 Amendment P6 Vermiculite Nutrient None 15 6 Amendment P6 Vermiculite Nutrient F O R L 15 *There were at least 30 plants per treatment total, of which 15 were included in the scoring results. 271 Table VIII.4 Treatments and conditions for seedling test 4 - amendment as plug and covering, second trial. Location Growth chamber (UBC Plant Science) Temperature 24 °C (1 * two weeks); 23 °C (subsequent) Lighting 18 hour photoperiod, fluorescent grow-lights (after germination) Inoculation time (days from seeding) 8 Scoring times (days from seeding) and number of replicates per scoring per treatment 46, 53 days 14 - 17 replicates each Tray plug type Small cube, 4 cm Trays (blocks) per treatment 3 Treatments No. Medium Covering Inoculation No. of replications scored 1 Rock wool Vermiculite None 40 2 Rock wool Vermiculite F O R L 50 3 Rock wool Amendment P6 F O R L 40 4 Amendment P6 Vermiculite F O R L 40 5 Amendment P7 Vermiculite F O R L 40 272 Table VIII.5 Treatments and conditions for seedling test 6 - amendment comparison, sandwich covering method Location Growth chamber (UBC Plant Science) Temperature 24 °C (1 s t two weeks); 23 °C (subsequent) Lighting 18 hour photoperiod, fluorescent grow-lights (after germination) Inoculation time (days from seeding) 7 Scoring times (days from seeding) and number of replicates per scoring per treatment 48 days 28 replicates each Tray plug type Small cube, 4 cm Trays (blocks) per treatment 2 Treatments No. Medium Covering Inoculation No. of replications scored 1 Rock wool Vermiculite None 40 2 Rock wool Vermiculite F O R L 50 3 Rock wool Amendment P6 (sandwich) F O R L 40 4 Rock wool Amendment 5A (sandwich) F O R L 40 5 Rock wool Amendment 5C (sandwich) F O R L 40 273 VIII.3.4 YIELD TEST The disease suppression yield test, conducted in the UBC Horticulture Greenhouse, investigated the effect of the sawdust-amendment mix growing medium on yield, over a full growing season, and under severe disease pressure from FORL inoculation. The horticulture greenhouse is a modern glass research greenhouse, with a controlled climate, automatic sunshade, and misting system. The greenhouse provides conditions as close to a commercial greenhouse as are available at UBC. However the climate control and irrigation system available were not as sophisticated as those of a commercial greenhouse. Climate could not be controlled separately from the main greenhouse; temperatures were typically near 20 °C except on sunny summer days, when they could reach the high 20's. Nutrient solution was delivered from a manually mixed tank, typically at EC 2.8 mS/cm, through drippers controlled by a timer with a manually set program. Later in the season, feeding was controlled by a photo-sensitive controller. No artificial lighting was provided. Dombito tomato seedlings, seeded in rock wool plugs, were transplanted to rock wool blocks on March 8. In the horticulture greenhouse, the bags of yellow cedar sawdust growing medium were inoculated on March 19 by distributing 30 ml of FORL suspension (2x10 spores/ml) into the medium directly under the future locations of each rock wool block. The seedlings were then planted on the bags on March 21, and grown under the same irrigation and climate conditions until December 6. Each treatment, shown in Table VIII.6, was separated into three bags of three plants each, and distributed as evenly as possible over an area of approximately 12 m 2. 274 Table VIII.7 lists that number of treatment. Al l treatments were located randomly in the greenhouse test site. Figure VIII. 1 shows the experimental layout and Figure VIII.2 is a picture of the greenhouse site. Table VIII.6 Procedures - Disease Suppression Yield Test Location : UBC Horticulture Greenhouse Cultivar: Dombito tomato Layout: 2 bags per row 3 plants per bag 9 plants per treatment (i.e. 9 replicates) 36 plants total Bags distributed, but inoculated plants all kept in the same rows. Monitoring: Fruit Yield - Total mass per plant and number of fruit per plant Plant measurements - height, stem diameter Plant health scoring - external indicators 1 plant 1 I B K B 1C E C N A Note : Italics indicate sawdust/compost medium. Fertilizing and watering regime were the same (conventional) for all treatments. Figure VIII.l Experimental Layout of Yield Test 275 Figure VIII.2 Greenhouse site of disease suppression yield test Table VIII.7 Yield test treatments Treatment Growing medium Inoculation 1 Sawdust None 2 Sawdust FORL 3 Sawdust, amendment P4, 2:1 v/v None 4 Sawdust, amendment P4, 2:1 v/v FORL After planting, weekly growth measurements were taken on each plant. Fruit yield was measured for each plant, with the yield divided into cull and marketable fruit, based on the BC Hot House grading guidelines. 276 Microbial counts on medium materials were conducted by a commercial laboratory (Cantest Labs, Burnaby, BC) using a standard plate count for bacteria and peptone water rinse and PDA medium for yeast and mold. STATISTICAL ANALYSIS Statistical data analysis was performed using SPSS 7.5 for Windows software, including ANOVA (analysis of variance) to compare data means. 277 VIII.4 RESULTS AND DISCUSSIONS VIII.4.1 TOMATO SEEDLING TESTS VIII.4.1.1 Test 3 (see Table VIII.3 for treatments and conditions) The scoring was done weekly for 5 weeks starting at week 3, using 5-6 replicates, so that progression of the disease could be observed. The first symptoms of crown and root rot began to appear 4 weeks after seeding, then developed so that after 6 and 7 weeks, the disease was very apparent in some of the seedlings. Table VIII.8 shows the average root disease score and shoot height for each treatment, based on the last 3 scorings. Reduced" disease symptoms were evident in compost amendment treatments, but no clear relationships with shoot height were apparent (Cheuk, 2003). This test also incorporated FORL negative (no FORL inoculation) controls in every treatment; the results for the negative controls were all similar. In future tests, the number of scorings would be reduced and the number of replicates increased; also a FORL negative control would only be included in the rock wool control group. Test 3 statistical analysis of root health, shoot length and leaves color is shown in Appendix H. Figure VIII.3 shows sample pictures of the seedling tests. 278 Table VIII.8 Results summary for test 3 - amendment as plug and covering (week 3 to week 8) Treatment No. Medium-covering-inoculation Root discas Mean shoot height cm 1 Rockwool/vermiculite/FORL- 1.00 0.00 55.1 2 Rock wool/vermiculite/FORL+ 1.85 0.88 48.9 3 Rock wool/amendment P6/FORL- 1.00 0.00 51.6 4 Rock wool/amendment P6 /FORL+ 1.40 0.20 46.1 5 Amendment P6/vermiculite / F O R L -1.00 0.00 49.7 6 Amendment P6/vermiculite /FORL+ 1.13 0.23 49.6 Figure VIII.3 Sample seedlings with FORL (left: treatment 6, right: treatment 2) 279 VIII.4.1.2 Test 4 (see Table VIII.4 for treatments and conditions) Results for Test 4 scorings are shown in Table VIII.9. The scoring of disease clearly shows reduced disease in seedlings treated with amendment. The increased root disease in the rock wool inoculated treatment was significantly greater than in the non-inoculated control. However the root disease in the amendment inoculated treatments was significantly lower than in the rock wool inoculated treatment (P<0.0T in all cases). Test 4 statistical analysis of root health, shoot length and leaves color is shown in Appendix I. Table VIII.9 Results summary for test 4 - amendment as plug and covering, second trial •• Treat meri Medium-eovering-inoculation Root disease score Mean shoot height • t No, .''Vv-;:'' SD • /:*•;• ;/,'cm. '• . 1 Rock wool/vermiculite/FORL- 1.33 0.96 53.8 2 Rock wool/vermiculite/FORL+ 2.14 1.11 53.7 3 Rock wool/amendment P6/FORL+ 1.35 0.63 59.0 4 Amendment P6/vermiculite /FORL+ 1.00 0.00 46.5 5 Amendment P7/vermiculite /FORL+ 1.00 0.00 46.8 VIII.4.1.3 Test 6 (see Table VIII.5 for treatments and conditions) Results from test 6 are shown in Table VIII.9. Similar to Test 4, the visual scoring of disease clearly showed reduced disease in seedlings treated with amendment, using the sandwich method. The increased root disease in the rock wool inoculated treatment was significantly greater (P<0.01) than in the non-inoculated control, based on the root disease score. The root disease score in the amendment inoculated treatments was 280 significantly lower than in the rock wool inoculated treatment (PO.01 in all cases). There was also significantly less (P<0.05) root disease for amendment L5A compared to amendment P6, but it did not differ significantly from L5C. Germination rates for all treatments were similar. Table VIII. 10 Results summary for test 6 - amendment comparison, sandwich covering method Tint. • No.. Medium-covering-inoculation Root disc 1 ascscore 5 Mean shoot height cm - Mean stem dial m m. Germination rate % 1 Rock wool/ Vermiculite/ FORL-1.00 0.00 49.7 6.3 88.0 2 Rock wool/ Vermiculite/ FORL+ 2.18 1.02 55.1 5.8 92.5 3 Rock wool/ amendment P6/ FORL+ 1.25 0.59 52.5 6.0 92.5 4 Amendment L5A/ Vermiculite/ FORL+ 1.00 0.00 46.9 6.3 95 5 Amdendment L5C/ Vermiculite/ FORL+ 1.11 0.31 52.4 5.9 87.5 It is interesting to note that shoot height (both Tests 4 and 6) and stem diameter (Test 6 only) measurements consistently showed that plants with more visible disease (higher root score) were taller and thinner (larger shoot height, and smaller stem diameter). In many but not all cases, these differences were statistically significant for 281 P<0.05. The reasons for this trend are not clear, however reduced height and diameter do not appear to be an indicator of disease in the FORL seedling tests. Leaf colour measurements (SPAD meter) did not provide significant differences between treatments. Test 6 statistical analysis of root health, shoot length and leaf color are shown in Appendix J. Disease scoring was limited mainly by the time required to develop disease. Based on the observations of all the disease tests including the yield test, it is very likely that in spite of the size trends mentioned above, seedlings with a high disease score would soon die, given more time. Characteristics of the FORL crown and root rot disease on tomato were such that the plants showed little external symptoms, then begin a very rapid decline and death follows quickly. This end result could not be achieved in the seedling tests due to the size limitation of the seedlings growing in the small plug trays in the growth chamber. 282 VIII.4.2 YIELD TEST (see table VIII.6 for procedure) During the approximately 9 month course of the growing season, a substantial number of plants died. In each case, death was attributed to Fusarium crown and root rot, as shown in Table VIII. 11. Table VIII. 11 Diseased plants during yield test by treatment Treatment •;:[• ' .'No': • / • >,•'' . '.Mccliti'm-ino'eulation Dead or dying plants by number and date 1 Sawdust-FORL- None observed 2 Sawdust-FORL+ • 2A-2 Died 26-May 2A-3 Died 26-May 2A-1 Died 31-May 2B-1 Died 31-May 2B-2 Wilting-Dec (FORL Infection) 2B-3 Wilting-Dec( F O R L Infection) 3 Sawdust-amendment P4-FORL- None observed 4 Sawdust-amendment P4- FORL+ 4A-1 Died 21-Apr With the exception of a single plant in treatment 4, all of these deaths occurred in treatment 2, conventional sawdust medium with FORL inoculation. It is reasonable to assume that the treatment 4 plant, 4A-1, was accidentally infected with FORL at the propagation stage, and would have died regardless of the subsequent treatment, based on the following observations: 1. Death occurred 1 month after inoculation. Of all disease suppression tests conducted, not one case was observed of plant death within one month of inoculation; at this stage symptoms are usually only visible internally. 283 2. The seedling was over 5 weeks old at inoculation; it is generally accepted that an older, healthy plant will take longer to develop the disease after inoculation than a small seedling. 3. Examination showed necrotic crown tissue and visible Fusarium spores, indicating the disease was well developed. 4. Al l other plants exhibiting Fusarium infection symptoms or death required at least 2 months after inoculation. The infection of 4A-1 may have occurred in the pathology greenhouse where the seedlings were propagated, and which was also used previously for FORL disease tests. With this in mind, 4A-1 was not included in the statistical comparison of yield, in that treatment 4 had one less replicate in the calculated average yield. Table VIII. 12 shows the marketable yield and fruit size for the four treatments; Figure VIII.4 illustrates the time-trend for total marketable yield for the inoculated control and amendment mix treatments. Statistical analyses of the yield data showed that the 50% decrease in yield of the inoculated control compared to the non-inoculated control was significant (P<0.05), and the 74% increase in yield of the inoculated sawdust plus amendment treatment over the sawdust control was significant (PO.05). The 10% decrease in yield for the non-inoculated sawdust plus amendment treatment compared to the non-inoculated sawdust treatment was not significant. The data also suggest a larger fruit size for sawdust plus amendment medium, but this difference was not statistically significant. The yield test statistical analysis is shown in Appendix K. 284 Table VIII. 12 Tomato yield fruit size results by treatment Treatment No:'' ' Medium-inoculation ' ' ' ' ' • , ' Average yield per plant - kg \ Average yield per plant - kg (adjusted)* . Average fruit size g l Sawdust-FORL- 10.69 10.69 152 2 Sawdust-FORL+ 5.39 5.39 146 3 Sawdust-amendment P4-FORL- 9.79 9.79 158 4 Sawdust-amendment P4-FORL+ 8.34 9.38 155 *Does not include plant 4A-1. 285 20-May-99 9-Jul-99 28-Aug-99 17-Oct-99 6-Dec-99 25-Jan-00 Date -©—Treatment 2: sawdust-FORL+ -A—Treatment 4: sawdust-amendment P4-F0RL+ Figure VIII.4 Cumulative marketable tomato yield for sawdust medium vs. sawdust and amendment mixture, both inoculated with FORL VIII.4.3 MICROBIAL ANALYSIS Samples from compost amendment batch P4, and fresh sawdust, were taken to commercial laboratory for microbial analysis and the results are shown in Table VII. 8 (Chapter VII). The microbial counts indicated an approximately 100 times higher bacteria to fungi count for the amendment compared to the sawdust. 286 VIII.5 CONCLUSIONS Significant reduction of crown and root rot disease caused by Fusarium oxysporum f. sp. radicis- lycopersici in susceptible tomatoes was achieved by addition of the greenhouse compost amendment to seedling plugs or blocks, and by mixing with the sawdust medium. For propagation, the compost amendment can be used as a rock wool„ plug covering, or a rock wool plug replacement. For growing, the compost amendment can be mixed with sawdust growing medium; a mixture of 2:1 sawdust to amendment by volume was shown to be effective. The reduction in disease resulted in 74% improved yield over a full growing season under high disease pressure (Cheuk, 2003). In the absence of high disease pressure, addition of the compost to the growing medium did not have a significant effect on yield. Based on the microbial counts, the compost amendment likely increased the microbial population and bacteria to fungi ratio in the growing medium; this may have contributed to the suppressive effect, however determining the suppression mechanisms would require further study. Positive effects were observed with several different batches of compost amendment produced using the in-vessel composting systems; this encouraging result suggests that the disease suppression effects observed are reproducible using the appropriate materials and composting process. The results of this study will help provide greenhouse vegetable growers with alternative solutions to manage soil-borne disease. 287 VIII.6 REFERENCES American Society of Agronomy. (1982) Methods of Soil Analysis, 2 n d Ed. Soil Science Society of America, Madison, Wis. BC MWLAP. (2002) Organic Matter Recycling Regulation, Reg. 18/2002. BC Ministry of Water, Land and Air Protection, Victoria, BC, Canada. Campbell, R. (1989) Biological Control of Microbial Plant Pathogens. Cambridge University Press, Cambridge, UK. Cheuk, W., Lo, K.V., Branion, R., Fraser, B., Copeman, R., Jolliffe, P. (2003) Applying compost to suppress tomato disease. Biocyle. 44 (1): 50-51. Grobe, K. (1998) Fine-Tuning the Soil Food Web. BioCycle 39(l):42-46. Hoitink, H. A., Zhang, W., Han, D. Y., and Dick, W. A. (1997) Making Compost to Suppress Plant Disease. Biocycle 38(4):40-42. Kai, H., Ueda, T., and Sakaguchi, M . (1990) Antimicrobial Activity of Bark-Compost Extracts. Soil Biol. Biochem 22(7):983-986. Mathur, S. P. (1996) The Use of Compost as a Greenhouse Growth Media. Waste Reduction Branch, Ontario Ministry of Environment and Energy, Toronto, ON. Portree, J. (1996) Greenhouse Vegetable Production Guide for Commercial Growers. BC Ministry of Agriculture, Fisheries and Food, Province of British Columbia. 288 Zhang, W., Dick, W. A. & Hoitink, H. A. (1996) Compost-Induced Systemic Acquired Resistance in Cucumber to Pythium Root Rot and Anthracnose. Phytopathology, 86 : 1066-1070. 289 CHAPTER IX M A R K E T I N G STUDY OF COMPOST AND COMPOST E X T R A C T IX.l BACKGROUND Increasing numbers of communities and businesses are turning to composting to divert materials from landfills, reduce pollution, and lower waste management costs Composters are converting a wide variety of otherwise wasted materials into safe, valuable, and marketable soil amendment products. Researches have shown that the use of compost as supplements has produced higher yields in crops than use of fertilizers alone (Maynard, 1989; Ribeiro, 2000; Korboulewsky, 2002; Tejada, 2003). Some governments also considered using compost to reduce raindrop impact and erosion and to retain moisture (Garner, 1985). Compost was used or distributed in one of four ways : used internally by the farm (greenhouse), sold in bulk, sold in bags, or given away free. Compost must be marketed as a resource which can benefit plants in order to get high returns (Faucette, 2003). The horticultural market has a great potential for compost utilization. Since the plants produced by nurseries, greenhouse industries, landscapers and home gardeners are high-value crops, it is important that the compost used be of the highest quality. These markets usually utilize mostly peat moss, pine bark, composted hardwood bark and spent mushroom soil (Gouin, 1985). The feedstock of greenhouse compost is very consistent and homogenous, unlike that of the municipal wastes. Since compost produced from greenhouse wastes showed positive results in plant growth and 290 disease suppression (Cheuk, 2003), it was proposed to develop a series of products, manufactured by crop waste composted from vegetable greenhouses. These products include: 1. An organic amendment product (OAP), suitable as a growing medium for many types of plants including vegetables and flowers. 2. A liquid growth promoter, which would be an OAP extract (a type of compost tea) derived from the same process. The author planned and has commissioned a study to assess the market for OAP and liquid growth promoter in the retail, floriculture and nursery markets, with market information to include: the size of the market, key market segments, potential interest in the product for each segment, competitor products (description, purpose, pricing, etc.) and distribution channels. The marketing research was assisted by Ference Weicker & Co., a company specialized in agricultural product marketing research and was funded by MART (a marketing research assistance program from National Research Council of Canada). IX.2 METHODOLOGY First a review of readily available information was conducted. Then questionnaires and sampling strategies, that were used for the interview programs were prepared. The major steps involved in the field research included: • Review of studies, reports and other materials relevant to the study issues. 291 • Conducted interviews with government representatives, industry association representatives, and other selected experts. A partial list of the organizations contacted includes: • BC Ministry of Agriculture and Food • BC Landscape & Nursery Association • United Flower Growers Co-operative Association • B.C. Horticultural Coalition • B.C. Vegetable Marketing Commission • Agriculture and Agrifood Canada • Statistics Canada Structured interviews were conducted with a sample of potential buyers and distributors. The sample was stratified to include a cross section of distributors, retailers (primarily garden centres) and large flower greenhouses and nursery producers. The sample included: • 12 representatives serving the retail market including 6 big box stores (i.e. Home Depot, Revy), 5 garden centres, and one custom soil supplier • 14 representatives associated with the nursery industry including 4 container only growers, 2 field only growers, 4 field and container growers, 1 container grower of seedlings for reforestation and 3 bulk suppliers of soils and growing medium 292 • 12 representatives associated with floriculture industry, consisting of 3 cut flower producers, 5 potted flower producers, 2 produces of both cut and potted flowers and two bulk soil suppliers A profile of key competitive products was developed. To do so, the followings were to be reviewed : the results of the market research, the product materials produced by other suppliers, market pricing, and a sample of the competing suppliers. 293 IX.3 RESULTS AND DISCUSSIONS IX.3.1 MARKET FOR THE ORGANIC AMENDMENT PRODUCTS 1X3.1.1 RETAIL MARKET According to Statistics Canada, the average household in Canada spent $26 annually on fertilizers, soils and soil conditioners in 1997. It was estimated that the average household in the Greater Vancouver area spends about $34 annually on fertilizers, soils and soil conditioners (Agri. Statistics, 1997). Given that there are more than 700,000 households in the Greater Vancouver area, it was estimated that the value of the retail market for fertilizers, soils and soil conditioners for home use totals approximately $24 million annually in that area alone. The survey of the various retail outlets indicated that premium soil conditioner products such as 100% organic soils, specialty soil mixes, worm castings and liquid growth promoters accounted for slightly less than 10% of their total soil conditioner sales. Based on these figures, it was roughly estimated that the value of the retail market for premium soil conditioners was approximately $2.4 million. The retail market for fertilizer and soil conditioner products consists of retail nursery outlets, hardware stores and home improvement centres, department stores and grocery store chains. Common wholesale and retail prices for premium soil conditioner products were summarized in the table below: 294 Table IX. 1 Wholesale And Retail Prices For Premium Soil Conditioner Products Product Wholesale Price Retail Price Pride - 100% organic potting mix (45 litre bag) $6.50 $9.50 Pride - All Purpose Potting Mix (45 litre bag) $6.50 $9.50 Pride - Soil Builder (35 litre bag) $6.95 $9.95 Eddi's - Potting Mix (25 litre bag) $3.60 $5.95 Eddi's - Potting Mix (40 litre bag) $5.40 $7.95 Black Gold - Earthworm Castings (8 litre bag) $2.50 $4.50 Black Gold - Earthworm Castings (32 litre bag) $4.95 $7.95 Steer Manure Compost (10 kg bag) $1.50 $2.50 Mushroom Manure Compost (15 kg bag) $1.99 $3.00 Sure Gro - Plant Food (250ml) $8.00 approx $12.00 Sea Spray - Plant Food (250ml) $8.00 approx $12.00 Alaska - Fish Fertilizer (1/2 gallon) $7.00 $10.00 Alaska - Fish Fertilizer (1 gallon) $13.00 $20.00 Later's - Fish Fertilizer (1 litre) $4.00 $6.00 Black Crystal - Humic Acid (1 litre) n/a $9.00 Black Crystal - Humic Acid (4 litres) n/a $30.00 Guardian Angel - Growth Promoter (500ml) n/a $24.95 Guardian Angel - Growth Promoter (4 litres) n/a $169.95 Source: Industry interviews, January 2000 The following paragraphs described some of the main types of products, their use and the most common brands available. 295 Specialty Potting Soils This is one of the largest categories. They contain a blend of soil conditioning products such as peat moss, bark mulches, manure composts, perlite, vermiculite and slow release fertilizers. Major brands carried by local retail outlets are Pride, Scott's, Bell's and Eddi's. Based on the representatives surveyed, the smaller independent retail outlets sell around 100 bags per year while the bigger stores with central buying sell over 2,000 bags per store (40 litre size equivalent). Manure Composts Steer and mushroom manure are the two compost manure products carried by most retail nurseries and home improvement stores. Like potting soils, manure is normally sold in 20 and 40 litre bags. The volume of manure compost sold per store ranges from about 100 bags in small retail nurseries to more than 2,500 bags in the larger retail nurseries. Pricing is closer to the standard peat moss and topsoil mixes. Earth Worm Castings Earthworm castings are sold in bag sizes of 8, 11, 20 and 32 litres. The two most common brands carried by local retail outlets are Walter Worms and Black Gold. Only about 25% of the retail outlets surveyed carried earthworm castings but all were aware of the products. Most stores said that sales were too low at 10 to 100 bags per year to continue stocking the earthworm casting products. 296 Other Composts Several bagged composts made from organic waste materials such as yard wastes are available through horticultural product wholesalers. Most retail buyers knew of one or more compost products but none carried them due to the small volume of sales. Liquid Organic Growth Promoters/Humic Acids Common brands carried are Sure-Gro, Sea Spray, Alaska Fish Fertilizer and Later's Plant Food. Liquid organic growth products usually come in 500ml, one, two and four litre sizes. Most retail outlets said that less than 1% of their sales of premium soil conditioner products are liquids. Volume sold ranges from about 10 to 500 litres per year depending on the size of the retail outlet. Distribution Channels The products handled by retailers in BC come primarily from garden wholesalers such as Eddi's Wholesale Garden Supplies and Green Leaf Garden Supplies or direct from manufacturers. For example, 85% of the products handled by retail nurseries are sourced through garden wholesalers. Most manufacturers provide the selling function for their products in the form of product representation and marketing programs. Larger retailers (usually with central buying power) preferred to purchase directly from the manufacturer because the product is less expensive. Smaller stores, however, sourced almost all of their product through the wholesalers. 2 9 7 Most of the leading manufacturers for soil conditioner products sold in the Lower Mainland were based in British Columbia or Alberta. Some of the leading manufacturers included Greenleaf Products Inc. (Burnaby), Eddi's Wholesale Garden Supplies (Surrey), Evergro (Delta), Sun Gro Horticulture Inc. (Surrey), Green Valley (Abbottsford), Consolidated Envirowaste (Aldergrove), Fison's Horticulture, Keefer Greenhouses (Richmond), Later's Chemicals (Richmond), Garden Friendly (Surrey), Premier Horticulture (Sask) and Bell's (Olds, Alta). INTEREST IN THE PROPOSED PRODUCT Ten of the 12 retailers that were surveyed expressed at least some interest in the proposed products. The most common reasons given for this interest were: 1. The ability to position the product as an organic product. Most of the retailers noted a strong consumer trend towards organic products. 2. The increasing demand for premium soil conditioner products. This increased demand was attributed to greater interest in both organic products and in gardening in general. The home gardening market segment is the fastest growing market segment for most of the retailers surveyed. 3. The product would be manufactured locally. This could benefit the appeal of the product to consumers, improve the ability of the retailers and/or distributors to manage the supply chain, and reduce transportation costs. 298 The decision of whether to purchase the product would be based on a number of key factors including price, product characteristics, promotional support, and service. If the decision was made to purchase, the potential volumes that could be purchased annually ranged from around 50 bags (1 skid) for smaller stores to several truckloads (1,000+ bags) for the larger stores. RETAILER RECOMMENDATIONS AND COMMENTS The retailers also provided a number of recommendations and comments regarding the proposed products. These are summarized below: 1. The retail buyers which were under survey suggested that it would be better to focus on a bagged product or at least start first with a bagged product (finished compost) and perhaps try a liquid product (compost extract) later once brand recognition has been established. • 2. The characteristics that buyers want in a bagged product are 'quality', sterilized product, truly organic without a lot of fillers, disease free, weed free, and conforms to pesticide and other regulatory requirements. 3. Packaging in clear plastic, high quality bags with colourful text and graphics was strongly recommended by buyers. 4. The soil conditioner market tends to be very sensitive to price. 5. Maintaining adequate supply for reorders is considered critical, given the seasonal nature of soil conditioner products. 6. Retailer awareness of the product is usually created through sales reps, mailings and faxes, and advertisements in trade publications. 299 7. Targeting specific market segments was recommended to create demand for the product. 8. The main barriers to penetrating the premium soil conditioner market were identified as the large variety of products already on the market and the highly price competitive nature of the industry. 1X3.1.2 FLORICULTURE MARKET The floriculture industry is one of the fastest growing sectors of the horticultural industry in British Columbia. The current rate of growth in the industry has leveled off to around 5% per annum. According to Statistics Canada, floriculture production is a $200 million industry in British Columbia. The businesses encompass a production base of over 200 hectares of greenhouse space and 100 hectares of field production. Ninety percent of the industry is located in southwestern British Columbia in the Fraser Valley. The leading product categories include cut flower, potted flowers and bedded plants. The size of individual greenhouse operations varies from a few hundred square metres to over 80,000 square meters. The average size is around 4,000 square metres (just over one acre). Cut flower production differs from potted plant production in that most production is ground based, either in field or greenhouse operations. The potential size of the market for compost type soil conditioner products needed would be around 375 yards. Unlike cut flower production, container plant production in the floriculture industry 300 consumes large quantities of soil medium products. A rough estimate of the amount of growing medium used in the industry is about 60,000 yards. PRODUCTS CURRENTLY USED The growing media used by the companies surveyed varies depending upon whether they produce cut flowers or potted flowers as outlined below: 1. Of the 5 cut flower producers, 4 add liquid fertilizer or manure to the soil primarily for nitrogen boosting. One of the firms, Houweling's, produced its flowers in rock wool blocks. Two of the cut flowers producers also added coconut fibre and sawdust to the soil to improve aeration. 2. Four out of seven of the potted flower producers surveyed use brand name potting mixes (Sun Gro, Pro Mix) while the other three producers use custom mixes. Only Burnaby Lake Greenhouses currently used compost, which they make themselves. 3. Brookside Greenhouses in Langley indicated that they have tested a sample of the compost from this project (compost from one of the pilot scale test) for seedling growth and obtained good results. They indicated an interest in testing the compost for propagating and as potting mix. None of the floriculture producers uses liquid organic growth promoters as it is too difficult to quantify the nutrients and other elements getting to the plant. 301 The volume of growing medium used in potted flower production varied according to the size of pots used and the length of time to plant maturity for wholesaling. Smaller operations generally used between 20 yards and 100 yards per acre, while some of the larger high production operations such as Burnaby Lake Greenhouses (30 acre) indicated that they used hundreds of yards of potting soil per acre. SOURCES OF SUPPLY The leading brands of potting soil appear to be Pro-Mix, which is manufactured by Premier in Manitoba and distributed by Westgro Sales in Vancouver; and Sun Gro Mix, which is distributed by Sun Gro Horticulture Inc. of Surrey. Prices to the grower are $12 to $14 per 3.8 cu.ft compressed bale. Custom mixes are supplied by West Creek Farms of Fort Langley at a price of $37 per yd for standard mixes ranging up to about $70 per yd depending on the additives requested. POTENTIAL INTEREST IN THE PROPOSED PRODUCT Nine of the 10 floriculture producers who expressed an opinion indicated at least possible interest in purchasing the proposed compost product. However, it should be stressed that the floriculture producers are generally satisfied with their current growing regimes; as a result, the new product would need to demonstrate significantly better performance or significant savings before the producers could be induced to try the product. 302 In terms of performance, the growers are primarily interested in yields and controlling the length of time to maturity. Only two growers expressed an interest in the disease suppression capability of the compost, reflecting the perception that diseases are not generally a major issue for the industry. Because the compost would represent only one of several ingredients used in a mix, the growers were very uncertain as to what volumes they may purchase if the decision was made to use it. The estimates ranged widely from several yards up to over 1 0 0 0 yards. RESPONDENT RECOMMENDATIONS AND COMMENTS The floriculture producers also provided a number of recommendations and comments regarding the proposed product. These are summarized below: 1. The selection criteria identified the most often included price, aeration, sterilization, weed free, consistency, good particle size, and durability for plants such as roses that have a longer growing cycle. 2. The effectiveness of the product must be clearly demonstrated in growth trials. 3. Most of the growers surveyed have no direct experience in using a compost product. 4. The growers are viewing the compost product as a replacement product, not as an additional ingredient. 5. Interest in a liquid product was very limited. 6. The best way to make buyers in the floriculture industry aware of the product would be visits by sales reps bringing samples for trial, advertising in grower 303 magazines, attending key industry trade shows (such as the Can West Horticulture Show), faxes and mailings, and promoting word-of-mouth (encouraging growers to talk to other growers). 7. A quality product was identified as key to market success. 8. The growers indicated that the greatest barriers to entry for a new supplier is strong competition (especially price competition) and the reluctance of growers to try a new, unproven product. IX.3.1.3 NURSERY MARKET The nursery industry has been one of the fastest growing segments of Canadian agriculture. Current growth rate is forecast at 4% to 5% per annum, as the market has matured following the explosive growth of over 13% that existed through the 1980s and early 1990s. According to Statistics Canada, nursery stock production was a $115 million industry in British Columbia in 1998. There are approximately 2,500 hectares of production area for nursery stock in BC. Nursery plant production is either done in the field or in containers and may be open or in greenhouses. Most growers are active in both field and container production. A principal advantage of container production is that three to eight times the number of plants can be produced in containers on the same area of land as in the field. There are considerable differences in the volume of media required for field production versus container production. In field production, plants are grown in rows 304 directly in the ground. Using an estimate of 50% nursery production area for field crops (3,000 acres) the potential volume of compost and manure type soil conditioners needed for adding organic content to the soil is roughly estimated at around 45,000 yards. For container production, growers use a soil-less mix for planting consisting of about 75% composted bark mulch and 25% peat with other additives such as perlite and vermiculite. Using an estimate for area of container production at around 50% of total nursery production area (3,000 acres) and 50 yards per acre as an average usage rate, the volume of growing medium used in container production is roughly estimated at about 150,000 yards. PRODUCTS CURRENTLY USED The products used by nursery producers vary depending upon whether the focus is container production or field production as described below: 1. The most common growing media used by the container growers surveyed is a blend of peat, composted bark mulch and perlite. Coconut fibre is also frequently used in place of peat when a more durable medium is required for plants such as roses that have to be in containers for 4 to 5 years. None of the growers add compost to their potting mixes. The only liquid growth enhancers mentioned were liquid fertilizers. 2. None of the growers involved in field production use compost products. Normally steer or chicken manure is tilled into the soil between plantings to increase the nitrogen and other nutrients. 305 The volume of various soil conditioners and mixes used annually ranged from a minimum of approximately 30 yards/acre to over 100 yards/acre. It depends on the size of the pots used and the number of crop rotations per year. Prices for bagged products were provided earlier in this chapter. Delivered prices for various bulk growing medium products, which was obtained from three suppliers based on a minimum of 7.5 yards of product, are as follows: Pricing for coconut fibre is around $18 for a 6 kg block, that expands to approximately 3 cubic feet of growing medium. 306 Table IX.2 Bulk Prices For Selected Growing Media Media . / • • ' • Bulk Pricing (CdS) . Composted bark mulch $18.00 yd Screened peat moss $22.50 yd Potting soil mix (with trace elements) $37.00 yd Organic soil amendment $22.50 yd Decorative bark mulch $22.50 yd Customized mixes $40 to $70 yd SOURCES OF SUPPLY Five of the container growers have their potting soils custom mixed by West Creek Farms of Fort Langley. Al l the growers said that they were very satisfied with the service and quality provided by West Creek Farms. The other three container producers buy the individual components and mix their own soils. The major components are sourced from Augustine Trucks in Pitt Meadows. Interestingly enough, Augustine Trucks also supplies bark mulch and peat to West Creek Farms for their custom mixes. POTENTIAL INTEREST IN THE PROPOSED PRODUCT Some interest in the proposed product was expressed by both container and field growers as outlined below: 307 1. The key criteria used by container producers for deciding what to use in the soil mixes is aeration and drainage/water retention. Other key considerations are past growing success (plant health and yields) and cost. Several growers mentioned the need for sterilized and weed free mixes. 2. Six of the 9 container producers said they would possibly be interested in a compost but the product would have to have proven benefits in growth trials or be as good as and cheaper than their current mix. 3. Two field producers said that they might be interested in adding compost to increase the organic content of the soil. To be competitive, the cost would have to be comparable to other methods such as planting selvage crops to till into the soil. Three of the container growers and the two field growers said that they were not interested in the compost product. Some of the concerns that were expressed about a compost product included: - The possibility of introducing disease; - The fine particle size, which would result in poor drainage; - Uncertainty regrading how to incorporate it into their current custom mixes; and Cost effectiveness. Any of the growers who had previous experience with compost found it to be too costly. 308 RECOMMENDATIONS AND COMMENTS FROM THE NURSERY INDUSTRY The nursery producers also provided a number of recommendations and comments regarding the proposed product. These are summarized below: 1. Although the growers surveyed do not currently use compost in their operations, most growers were aware of other compost products available for adding to potting mixes. 2. The potential disease suppression aspects of the proposed product generated little response from the growers. 3. Certification would be very important for accessing this market. 4. The consistency of the product would also be an important consideration is deciding whether to use it. 5. Direct marketing, in the form of having sales representatives bring samples for testing and information on the product, was identified as the best means to create grower awareness of the product and induce trial. 6. Growers are satisfied with their current growing media. 7. The factors that most growers felt would make it difficult for a new supplier to gain entry into the market were grower reluctance to try a product and the level of satisfaction with current soil mixes and suppliers. 309 IX.4. CONCLUSIONS The major findings and conclusions arising from this review of the market for organic amendment products are as follows: 1. While there are a variety of compost products sold in the retail market, most nursery and floriculture producers are not currently using compost as a growing media. Only one of the floriculture or nursery products currently used compost in their mixes. Burnaby Lake Greenhouses manufactures the compost themselves from their own wastes and blends it in with a purchased mix. They have found that this approach not only reduces their waste stream but also serves to reduce the overall cost of their mixes. However, they claimed that the composting system need to be re-designed to give consistent high quality compost, which was one of the main objectives of this research project. 2. Potential interest in the proposed compost product was expressed by all three of the markets reviewed. Ten of the 12 retailers expressed at least some interest in the proposed product, noting that: • The product could be positioned as an organic product; • The retail demand for premium soil conditioner products is increasing; • The product would be manufactured locally, which could increase the appeal of the product to consumers, improve the retailers ability to manage the supply chain, and reduce transportation costs. 310 Nine of 10 floriculture producers and 6 of 9 container nursery producers expressed potential interest in purchasing the proposed compost product. The producers are generally satisfied with their current growing regimes but would be open to a new product that could demonstrate better performance, e.g. improvement in yield, shelf life, etc, or cost savings. 3. The potential impact of the product in suppressing disease was not of direct interest to most of the buyers. Retailers felt that most consumers would not appreciate the disease suppression qualities of the product. The floriculture and nursery producers generally indicated that diseases are not a major issue for their industry. However, all the buyers recognized that, if the compost were to outperform other media, regardless of the reason, they would be interested in the product. 4. Positive results from product trials and product certification will be key to penetrating the nursery andfloriculture markets. Formal testing will be required to demonstrate that the product is effective and safe. The testing should identify for what applications, and in what mixes, it will be most effective. The nursery and floriculture producers and customer mixers noted that they would need to know what is in the compost in order to adjust the slow release fertilizers and other components of their mixes. It should also be noted that a compost product may not be appropriate for some applications. For example, some concern was expressed that compost may compact too easily, plug up drain holes, or not be durable enough for plants will long growth cycles. Other possible concerns that were expressed included 311 consistency (the possible variable nature of its nutrient content) and sterility (the possibility of introducing disease). However, compost produced from the greenhouse plant wastes in an in-vessel composter would be acceptable since it is produced from consistent wastestream and has passed through the requirements for pathogen reduction. Several growers said that, even with certification and testing, they would still conduct their own tests. Certification is also important for retailers due to liability and other issues. 5. The biggest barriers to market entry are that other products are well established in the marketplace and the perceived risk of altering growing regimes can be significant. The level of price competition in most segments is very high; a product has to either provide superior benefits or provide direct costs savings in order to be successful. It was noted that the garden centre retail market tends to be less price competitive and more focused on quality than are the big box home centres. 6. The broad and fairly rapid acceptance of cocunut fibre as a new growing media demonstrates that producers are willing to vary their regimes when trial results indicate that significant benefits exist. Coconut fibre is now used extensively in the nursery industry for plants with a long growing cycle. 7. Direct marketing is a key to introducing the product in all three markets. Visits by sales representatives (bringing samples and information materials), direct mail programs, advertisements in trade publications, and attendance at industry trade shows were identified as the best means to introduce new products. 312 8. Packaging and, to a lesser extent, consumer advertising are also key for introducing a new retail product. Packaging in clear plastic, high quality bags with colourful text and graphics was strongly recommended by buyers as a means to create in-store demand for the product. A co-op fee would have to be paid to many of the retailers to have the product advertised in flyers. 9. The level of service provided by the supplier is also important. Retailers noted that, if there were problems obtaining supply, they would discontinue the product. An option for serving the nursery and floriculture industry would be to work through custom mixers such as West Creek Farms. 10. Interest in a liquid product was limited. Even in the retail segment, the buyers recommended that it would be better to focus on a bagged product or at least start first with a bagged product and perhaps try a liquid product later once brand recognition has been established. 313 IX.5. REFERENCES Agriculture Statistics. (1997) Ministry of Agricultural, Food and Fisheries. Cheuk, W., Lo, K.V., Branion, R., Fraser, B., Copeman, R., Jolliffe, P. (2003) Applying compost to suppress tomato disease. Biocyle. 44 (1): 50-51. Faucette, B., Governo, J. Graffagnini, B. (2003) Compost pricing and market survey in Georgia. BioCycle. 44(4): 32-33. Garner, M . (1985) Summary of Principal Provisions of State Laws Providing for Erosion and Sediment Control as of July 1,1985. National Association of Conservation District. Gouin, F.R. (1985) Greenhouse and Nursery Crops In Chapter 4. Benefits of Land Spreading of Sludges in the Northeast. The Perm. State Univ. Penn Agr. Exp. Sta., Univ. Park, PA. Korboulewsky, N. , Bonin, G., Massiani, C. (2002) Biological and ecophysiological reactions of white wall rocket (Diplotaxis erucoides L.) grown on sewage sludge compost. Environmental Pollution. 117(2): 365-370. Maynard, A. A. (1989) Agricultural composts as amendments reduce nitrate leaching from soil. Frontiers of Plant Science. 42(1): 2-4. Ribeiro, H.M., Vasconcelos, E., dos Santos, J.Q. (2000) Fertilization of potted geranium with a municipal solid waste compost. Bioresource Technology. 73(3): 247-249. Tejada, M . Gonzalez, J.L. (2003) Effects of the application of a compost originating from crushed cotton gin residues on wheat yield under dryland conditions. European Journal of Agronomy, 19(2): 357-368. 314 CHAPTER X OVERALL CONCLUSIONS & RECOMMENDATIONS FOR FUTURE WORK This is a comprehensive project that investigated and transferred the science of composting into the modern application of agriculture. The project developed and demonstrated an appropriate technology for the bio-conversion of wastes generated by greenhouses in B.C. Wastes related to crop production were categorized and the box-type (in-vessel) bio-conversion technology was investigated both in lab scale and pilot scale. The quality of the finished compost and the benefits of utilizing the bio-converted products as high-quality organic medium amendment for growing vegetable crops in greenhouses were evaluated. Greenhouse wastes, when not handled properly, can become a threat to the environment. Current practices are polluting the agricultural lands and ground waters. The greenhouse industry has a continuous organic waste stream, and a consistent waste characteristic. This project showed a very efficient way to treat the wastes, i.e. turning the wastes into valuable compost. By doing so, it not only reduces the disposal costs for farmers and mitigates the landfill space problem for the government, but also creates a new kind of organic growing media for the farmers. Greenhouse year-end waste, mostly composed of stems, leaves, fruit rejects, pruning and twines, needs to be shredded to proper sizes prior to the bio-conversion process. In this study it was found that a shredder 315 with a rotating bladed drum would be most suitable for greenhouse applications. Not only because it can handle twines and clips, but also it is economical and mobile inside a greenhouse, whereas most of other commercial composting shredders were too big and too expensive. Shredding the conventional plastic twine along with the organic waste would reduce the quality and resale value of the compost. Therefore, utilization of 5 different bio-degradable twine products were evaluated. They were tested for their physical properties during the greenhouse usage and their compostabilites in an in-vessel composting system. From the results obtained from both tests, it was concluded that it was feasible to utilize alternative twines in greenhouse vegetable production. Furthermore, among the five types of twines, EcoPLA®2000D, a synthetic bio-degradable plastics, was found to be the most suitable substitute for the traditional plastic twines currently used in greenhouses. In total, there were 9 lab scale and 8 pilot scale composting experiments being investigated. The lab scale experiments were designed to investigate different parameters of composting and their effects on the process and compost quality, while the pilot scale only served as a demonstration for the greenhouse operators, hi the first part of lab scale experiment, different control logarithm of composting were tested. Bacterial inoculation was found not necessary to start up the composting process which helped resolved a lot of hygienic and transportation problems for greenhouse operators. With a good control of algarithm and heat loss, composting of greenhouse wastes could reach satisfactory temperatures for pathogen reduction and carbon losses for compost stabilization. Ammonia emission might be a problem but it could be reduced by using air-recirculation 316 or removed by biofilter with compost as media. Recirculation cooling control was a more effective method to maintain the process temperature not to go beyond the set point than any kind of temperature feedback control. Less leachate and condensate were found from the reactors with air recirculation control, and this would be a big advantage for greenhouse on-site composting. System with air recirculation for cooling and part of the aeration showed higher degradation rate, and also more consistent moisture content of final compost. In Part 2 of the lab scale test, different substrate recipes (C:N ratio, moisture content, porosity, etc) were investigated. Alder bark was found to be a better choice of bulking agent than hemlock bark in terms of better substrate structure, more carbon loss, less nitrogen loss, and higher process temperature. Shredding was proven not necessary before composting of pruning wastes and it also helped minimizing the amount of leachate. The amount of leachate was directly correlated with the initial moisture content of the substrate and the amount of condensate was directly correlated with the duration of high temperature of the process. Bulking agent (alder bark) of about 20-30% (in weight) was necessary for composting pruning wastes. For year-end wastes, a ratio of 62% vines, 13% used sawdust and 25% alder bark was recommended. Pilot scale tests demonstrated the successful use of in-vessel system, the proper control algorithm and substrate recipe developed in the lab scale. A cost analysis was done based on the current disposal cost, investment on composting facility and sale of compost. Based on a 4 ha tomato or pepper greenhouse, and amortizing the capital equipment over five years, the net annual cost of composting represented a savings of $8,000 annually. 317 Bioassay successfully showed that the growth rate of seeds correlated to the degree of degradation and stability of the finished products. Chemical analysis of compost extract, leachate and condensate showed that they contained a number of valuable nutrients. Leachate, when diluted to 10 times was proven non-toxic or even beneficial in the case of tomato seed. Condensate from composting could be reused for plant without dilutions. In order to enhance the economic and environmental sustainability of on-farm greenhouse composting, a growth trial and disease suppression trial were done by using compost as growing media in a greenhouse. These studies found that compost manufactured from greenhouse crop waste in a controlled process could be suitable for use as a growing medium amendment in soil-less tomato greenhouses. Using conventional management techniques, a similar yield can be achieved compared to conventional sawdust medium using a 2:1 sawdust to amendment mix by volume. Significant reduction of crown and root rot disease caused by Fusarium oxysporum f. sp. radicis- lycopersici in susceptible tomatoes was achieved by addition of the greenhouse compost amendment to seedling plugs or blocks, and by mixing with the sawdust medium. The reduction in disease resulted in 74% improved yield over a full growing season under high disease pressure. The results of this study helped provide greenhouse vegetable growers with alternative solutions to manage soil-borne disease. Through the marketing study, it was found that selling the compost in another market like nursery, retail store required great amount of marketing and selling skill. So, greenhouse re-utilization might be the best ultimate goal and it also formed the first step in moving soil-318 less vegetable greenhouses in BC toward a more sustainable waste management and growing practice. It is recognized that continued efforts in the subject areas are needed to expand the current knowledge and understanding of the composting process and utilization of compost in the greenhouse application. The followings are recommended: 1. To investigate the mass balance and track study of carbon and nitrogen loss with air-recirculated compost reactor; therefore to find out more about the degradation process related to process and aeration control. 2. To conduct a bioassay track study to understand the change of level of phytotoxicity during composting. 3. To carry out a full scale growth trial with compost as growth media for different greenhouse crop, especially for organic farming greenhouses. 4. To carry out different disease suppression trial using other common greenhouse diseases, e.g. pythium, botrytis, etc. 319 APPENDIX A Waste Tracking Data Sheet PRUNING WASTES EVAULATION DATA SHEET Month : Number of Totes Day Leaves Vines Culls Total 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 320 APPENDIX B Greenhouse Solid Waste Management Survey Survey Information This survey was created as part of a project studying composting of greenhouse wastes, conducted for the Western Greenhouse Grower Society, in conjunction with the University of B.C. The aim of the study is to determine the feasibility of composting vegetable greenhouse crop waste as an economically viable solution to waste disposal. This is a voluntary survey. You may choose to fill out all or none of the fields, including identification, at your option. Identification Information Greenhouse Name Address Contact name(s) Telephone Date Waste Characterization Please note, this survey pertains to crop (organic) waste, and does not include plastic film or other non-crop wastes, except where the crop waste contains limited amounts of other materials such as plastic. Please fill in the crop type(s) at the top. If you are not sure of the numbers for individual crops, you may fill only the total. Amounts may be specified by mass or volume using either Imperial or SI units. Please indicate which units you are using (tons (t), Tonnes (T), cubic yards (cy), cubic metres (m3)) 321 Crop 1: Area (Acres or Ha) Crop 2: Crop 3: Total Growing season (Start-F i n i s h ) ^ . Crop 1 Crop 2 Crop 3 Total Amount of Prunings (during growing seas.) Amount of Rejects Amount of Year-end Waste Type of Growing Medium Amount of Growing Medium Waste Most significant pathogen organisms Pesticides used (please identify Does any crop waste contain contamination, such as twine or plastic clips? If so, which type of waste and which types of contaminants? Waste Collection Please describe the method of waste collection that you use in the greenhouse, including what type of equipment is used (forklift, tractor, type of bins, etc.). Prunings Year-end Waste Disposal Please indicate how waste is disposed and the approximate cost of disposal (including trucking), for the different waste types. Disposal method - landfill, land application, composting, or other (please describe) Cost Prunings Rejects Year-end Growing Medium Total If you are using land application of wastes, is it on-site or trucked to another location? If you are composting waste, is it composted on-site or trucked to another location? 323 If it was an economically competitive waste disposal method, would you have an interest in composting on your property? ' If no, why? If yes, what concerns would you have about composting? APPENDIX C Sample Aeration Requirement Calculations 1. Compost Mass: 1.4t/d*21 days = 200 tonnes Compost Mixture is 60% greenhouse waste, 40% bulking agents, therefore Total Mass = 200 tonnes/0.60 2. Compost component weights: Based on 333 tonnes of total compost: 200t waste + 133t of bulking agents 4. Stoichiometric Demand C 1 29Hi84O 8 0N + 134.25 0 2 -» 129C02 + 90.5 H 2 0 +NH3 3029 g:2148g Specific Oxygen demand: x = 2148/3029 or 0.7Ig 0 2/g BVS oxidized When all of the BVS is oxidized, total 0 2 requirement is : 0.71*069.6*1000 = 49416kg 0 2 Aeration Rate Required: Air required = specific oxygen demand*degradability*VS content*(g air/g 0 2) = (0.71gO2/g BVS)* (0.565 g BVS/g VS)* (0.86g VS/g ds)*(lg air/0.23 g 0 2) = 1.50 g air/ g ds =333 tonnes Moisture Solids Volatile Solids BVS 199.8 tonnes 143.2 tonnes 123.1 tonnes 69.6 tonnes 3. Chemical Composition: C129H184O80N 5. Moisture Removal: Initial Mass, Mj = 333 Initial Solids Content, Sj = 0.43 Initial VS content, V; = 0.86 Assume: Final solids content, Sf = 0.65 Final VS Content, Sf = 0.50 Net water loss per unit mass of dry substrate: M n = (l-S i/S i)-[(l-V i/a-Vf)*(l-Si/Sd] = (l-0.43/0.43)-[(l-0.86/0.50)*(l-0.65/0.65) =1.17tH2G7tds = 1.17gH 20/gds From psychrometric chart, Assumptions: - ambient air supplied to process is 20°C, 75% relative humidity - exit gas is 55°C, 100% relative humdiity Net Moisture Removal = w2-wi = (0.1147-0.011) g H 2 0 / g dry air = 0.1037 gH 2 0/dry air Mass of air required to remove 1.17 g H 2 0: (1.17 g H 20/g ds)/(0.1037 g H 2 0 / g air) = 11.28 g air / g ds 6. Heat Removal: 0.71 g0 2 /gBVS oxidized =xbs =(0.71 g)* degradability * VS content = (0.71 g)* (0.86)*(565) = 0.3450 g 0 2/g ds Heat Production, based on a rule of thumb: (3620 cal/g)*(0.3450g 0 2/g ds) = 1124.66 cal/g ds Enthalpy values: Assumptions: -ambient air supplied to process is 20°C, 75% relative humidity -exit gas is 55°C, 100% relative humidity From psychrometric chart: hi = 48kJ/kg and h2 = 353 kJ/kg Change in enthalpy = h 2 - hi = 305 kJ/kg = 305 J/g = 73 cal Ig air 326 Air requirement for heat removal is: (1124.66 cal/g ds)/ (73 cal/g air ) = 15.406 g air Ig ds 7. Aeration Rate: Since the greatest amount of aeration is required for heat removal, the aeration rate calculations is based on this value. Aeration rate = (air/BVS)*(dMBvs/dt) air/BVS: (15.406 g air/ g ds)*(69.6t BVS/143.21 solids)-l = 31.697 g air / g BVS dMevs/dt: dMBvs/dt = - k B i M B i - k B 2 M B 2 kei = k B U 2 0 (1.066 T ' 2 0-l.21 T"*°)FiF2 Fast fraction k B 2 = kB2,20(1.066T-20-1.21T-60)FiF2 Slow fraction M B i = (0.35)*(69.6t) = 24.361 k B , , 2 0 = 0.019 M B 2 = (0.65*(69.6t) = 45.241 k B 2 , 2 0 = 0.0024 From Figure 11.9 of Haug, 1993 Fi = 0.95 From Figure 11.11 of Haug, 1993 F 2 = 0.882 (Assume 0 2 in FAS = 15%) Assuming initial compost temperature, T = 20°C: k B i = 0,01592 d - 1 = -78621 g BVS/d"1 Aeration Rate = (31.697 g air/g BVS)*(78621 g BVS/d) = 2492049 g/(28*0.79*32*0.21)g/mol = 86409.5 mol = 86409.5 mol * 25.4 L/mol * 1/1000 mol/L = 2194.8 m3/d = 1.52 m3/min or 53 cfm Reference: Haug, Roger T. (1993) The Practical Guide to Compost Engineerings Boca Raton: Lewis. 327 APPENDIX D Rutgers Temperature Feedback: C-Program Code of control Algoritm and Flowchart /* T M P _ C T U */ /* This program accepts the following inputs: - temperature (from X input) - temp, setpoint and default duty cycle (user def. inputs) and outputs a flag (0 or 1) indicating whether the aeration should be off (0) or on (1), according to temperature feedback and a fixed schedule when the temperature is below the setpoint. It is based on the example code provided (rbv.c) which reads the current value of the block specified as the X input block in the menu. */ #define tempsp UsDefVarl /* temperature setpoint (int) */ #define duty_percent UsDefVar2 /* duty cycle in percent (int) */ #include "windows.h" #include "math.h" #include "ltuserw.h" /* global variables */ int counter; int maxcount; float temp; int airflag; int dutycount; int airarray [ 10] [ 10]= {1,0,0,0,0,0,0,0,0,0, /* array that determines air on/off cycle */ 1,0,0,0,0,1,0,0,0,0, /* a given duty cycle uses a single row */ 1,0,0,1,0,0,1,0,0,0, /* eg. duty cycle 20% uses row 2 */ 1,0,1,0,0,1,0,1,0,0, 1,0,1,0,1,0,1,0,1,0, 1,0,1,0,1,1,0,1,0,1, 1,1,0,1,1,0,1,1,0,1, 1,1,1,1,0,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,1}; struct ReadBlocklnput Datablock; struct ReadBlockOutput Value; void F A R P A S C A L LT_CIcon_init(fh_name, ptr, callback) char *fh_name; C I C O N D A T A *ptr; void *call_back(int fh, void far *inp_data, void far *outp_data); { int i; Ciconptr = ptr; /* initialize algorithm variables */ counter = 1; max_count= 10; Data_block.usNum=Xin; /* set X input block as the block to read */ > . void F A R P A S C A L LTCIconclosefhandle) int handle; { } void F A R P A S C A L LTCIconOpen(ptr) C I C O N D A T A *ptr; { int i; Ciconptr = ptr; } void F A R P A S C A L tempfblfptr, callback) C I C O N D A T A *ptr; void *call_back(int fh, void far *inp_data, void far *outp_data); { /* run time */ Ciconptr = ptr; call_back(READ_BLOCK_VAL,(void *)&Data_block,(void *)&Value); /* get current value */ if(Value.error) /* check for error condition */ { error_status=Value.error; BlockValReadErr; /* use L A B T E C H error macro to return correct error code */ LTReturn; /* pass back error information and return to L A B T E C H */ } /* start algorithm */ dutycount = floor((duty_percent/10)+0.5); /* duty cycle represented by integer eg 20% is 2 */ 329 if (counter = maxcount + 1 ) counter = 1; /* reset at end of cycle */ temp = Value.dData; /* input temperature */ if (temp >= tempsp ) airflag =1; /* air on when temp exceeds setpoint */ else { if ( dutycount > 0 ) airflag = air_array[duty_count-l][counter-l]; /* air on when temp below setpoint based on duty cycle */ else airflag = 0; } dCalcResult = airflag; /* output air flag */ counter++; /* end algorithm */ LTReturn; } 330 AIR ON OR OFF FOR 10 MINUTES ACCORDING TO 20% DUTY CYCLE CONTROL SEQUENCE SAMPLE TEMPERATURE FROM THERMOCOUPLES EVERY 10 MINUTES AIR ON FOR 10 MINUTES DUTY CYCLE CONTROL TABLE 331 APPENDIX E Linear Temperature Feedback: C-Program Code of control Algoritm and Flowchart /*t2_CTL3 */ / T h i s program accepts an input of temperature (Xin), and outputs a flag (0 or 1) indicating whether the aeration should be off (0) or on (1) during each cycle based on the temperature and the temperature trend (increasing or decreasing). Air duty cycle is checked and potentially adjusted once every "long cycle" (maxcount2 counts). Aeration is off when temperature exceeds the setpoint. Based on the sample code rbv.c and example.c */ #define tempsp UsDefVarl /""temperature setpoint (int) */ #define duty_percent UsDefVar2 /* duty cycle in percent (int) */ #include "windows.h" #include "math.h" #include "ltuserw.h" /•global variables*/ int counter; int counter; int maxcount; int max_count2; int airinterval; /*interval(#counts) between air on cycles */ int tempinterval; /*#counts since temperature exceeded setpoint*/ int dcadjdelay; /*#counts temp must be <setpoint before adjusting aeration duty cycle*/ float tempcurr; float tempavgcurr; float temp_avg_prev; float tempsum; float tempchange; /•current temp*/ /•average temp, from previous period*/ /•average temperature current period*/ /*sum used to calculated first temp.average*/ /•change from previous to current period*/; int airflag; int dutycount; /•sum used to calculated temp.average*/ /•output (0 or 1)*/ int dutycountincr; int dutycountmin; intair_array[10][10]= {1,0,0,0,0,0,0,0,0,0, /*array that termines air on/off cycle*/ 1,0,0,0,0,1,0,0,0,0, /*a given duty cycle uses a single row */ 1,0,0,1,0,0,1,0,0,0, /*eg. Duty cycle 20% uses row 2 */ 1,0,1,0,0,1,0,1,0,0, 1,0,1,0,1,0,1,0,1,0, 1,0,1,0,1,1,0,1,0,1, 1,1,0,1,1,0,1,1,0,1, 1,1,1,1,0,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,1}; struct ReadBlocklnput Datablock; struct ReadBlockOutput Value; void F A R P A S C A L LT_Cicon_init(fii_name,ptr,call_back) char *fh_name; C I C O N D A T A *ptr; Void *call_back(int fn, void far *inp_data, void far * outpdata); { inti; Ciconptr = ptr; /•initialize algorithm variables */ counter = 1; counter2 = 1; maxcount =10; max_count2 = 30; dcadjdelay = 60; tempinterval = 0 dutycountmin = 1; temp_sum = 0 temp_avg_prev = 0; duty_counr=floor((duty_percent/10)+0.5); /* duty cycle represented by integer eg 20% is 2*1 if (dutycount ==0) dutycount = dutycountmin; } void F A R P A S C A L LTCiconclosefhandle) int handle; { } void FAR P A S C A L LT_Cicon_Open(ptr) C I C O N D A T A *ptr; { int I; Ciconptr = [tr; } void F A R P A S C A L modeB (ptr, callback) C I C O N D A T A *ptr; Void*call_back(int fh, void far *inp_data, void far *outp_data); { /*run time*/ Ciconptr = ptr; Data_block.usNum=Xin; /*set X input block as the block to read*/ Call_back(READ_BLOCK)VAL,(void*)&Data_block,(void*)&Value);/*get current value*/ If(Value.error)/*check for error condition*/ { error_status=Value.error; Block ValReadErr; /*use L A B T E C H error macro to return correct Error code */ LTReturn; /* pass back error information and return to L A B T E C H * / } temp_curr=Value.dData; /*start algorithm*/ if( counter=max_count+l) counter = 1; /*reset at end of small cycle */ if (counter2==max_count2+l) counter2=l; /*reset at end of large cycle*/ if ( counter2<=30&&counter2>=2) /*add temp.value to sum*/ temp_sum=temp_sum+temp_curr; if (temp_curr>=temp_sp) temp_interval = 0; /*check for temp exceeding setpoint*/ else temp_interval++; /*increment temp interval if below setpoint*/ if(counter2==l) {/*check temp, averages and adjust duty cycle once per maxcount2 counts */ temp_avg_curr=temp_sum/29; /*calculated temp.averages*/ temp_change=temp_avg_curr-temp_avg_prev; temp_sum=0; /*reset temp.sum value*/ if(temp_change<-0.02&&temp_interval>dc_adj_delay) /*decreasing trend*/ duty_count=floor(((temp_avg_curr*0.0286-1.26) *10)+0.5); if(temp_change>0.2&&temp_interval>dc_adj_delay: /*increasing trend*/ duty_count=floor(((temp_avg_curr*0.00686+0.06)* 10+0.5); 334 if(duty_count<l) duty_count=duty_count_min; if(duty_count>10) duty_count=10; } if(temp_curr>temp_sp) air_flag=0; /*air on when temp exceeds setpoint*/ else air_flag=air_array[duty_count-l][counter-l]; /*air on when temp below setpoint based on duty cycle*/ dCalcResult = airflag; /*output air flag*/ temp_avg_prev=temp_avg_curr; counter++; counter2++; /*end algorithm*/ LTReturn; } 335 AIR O N OR OFF 2 MIN. A C C O R D I N G T O C U R R E N T D C 1 M I N U T E C Y C L E C A L C U L A T E 30 MIN RUNNING A V E R A G E T E M P • A V G T (CURR) >AVG T(PREV) A V G T (CURR) <AVG T(PREV) 9 NO N O C A L C U L A T E N E W D C B A S E D O N E Q . l ("INCREASING TEMP) C A L C U L A T E N E W D C B A S E D O N EQ.2 (DECREASING TEMP) Linear Temperature Feedback Flowchart. Note: "DC" indicates duty cycle. 30 M I N U T E C Y C L E APPENDLX F Sample Fertilizer Recipe of Growth Trial (Chapter VII) Concentrated (100x) Diluted Per Tank A Per 1000 litre No. of litres: 100 Per litre 15.5% TN Calcium Nitrate -1% Amm./14.5% Nitrate 125 kg 12.5 kg 1.25 g 13%TN Calcium Nitrate 0 kg 0.0 kg 0.00 g Calcium Chloride 0 kg 0.0 kg 0.00 g Iron chelate - solid 0 kg 0.0 kg 0.00 g Iron chelate - liquid 6 i 600 ml 0.06 ml Urea 1 kg 100.0 g 0.01 g Tank B Potassium Nitrate 25 kg 2.5 kg 0.25 g Mono-Potassium Phosphate 25 kg 2.5 kg 0.25 g Magnesium Sulphate (Epsom Salts) 50 kg 5.0 kg 0.50 g Potassium Chloride 15 kg 1.5 kg 0.15 g Potassium Bicarbonate 0 kg 0.0 kg 0.00 g Manganese Sulphate 175 g 17.5 g 0.0018 g Zinc Sulphate 200 g 20.0 g 0.0020 g Boron (Borax, 20.5% B) 300 g 30.0 g 0.0030 g Copper Sulphate 25 g 2.5 g 0.0003 g Sodium Molybdate 20 g 2.0 g 0.0002 g Urea 1 kg 100.0 g 0.01 g •337 APPENDIX G ANOVA-Plant Growth (for Chapter VII) Cases Included Excluded Total N J Percent N \ Percent N Percent 12 j 85.7% 2j 14.3% | 14 100.0% a Cumulative Growth by Medium, Nutrient Strategy, Row Number Cell Means(b,c) Cumulative Growth Medium Nutrient Strategy Row Number Mean N Conventional Total 563.97 3 Modified Total 601.33 3 Sawdust A 553.65 2 Total B 607.15 2 C 587.15 2 Total 582.65 6 Conventional Total 568.27 3 Modified Total 577.03 3 Sawdust A 568.40 2 +Amendment Total B 580.55 2 C 569.00 2 Total 572.65 6 A 546.90 2 Conventional B 580.55 2 C 570.90 •2 Total 566.12 6 A 575.15 2 Total Modified B 607.15 2 C 585.25 2 Total 589.18 6l A 561.03 4 Total B 593.85 4 C 578.08 4 Total 577.65(a) 12 a Grand Mean b Cumulative Growth by Medium, Nutrient Strategy, Row Number c 3-way and higher means are not computed due to the limit on maximum order interaction. 338 ANOVA(a) Experimental Method Sum of Squares df Mean Square F Sig. (Combined) 4052.26 4j 1013.07) 5.11 .17 Main Effects Medium 300.00 l | 300.00 1.51 .34 Nutrient Strategy 1596.21 lj 1596.21 8.05 .11 Row Number 2156.05 2, 1078.02 5.44 .16 (Combined) 1683.35 5| 336.67 1.70 .41 Cumulative Growth 2-Way Medium * Nutrient Strategy. 613.47 li 613.47 3.10 .22 Interactions Medium * Row Number 954.55 2) 477.27 2.41 .29 Nutrient Strategy * Row Number 115.33 2 57.67 .29 .78 Model 5735.61 9| 637.29 3.22 .26 Residual 396.47 2 198.23 Total 6132.07 11 557.46 a Cumulative Growth by Medium, Nutrient Strategy, Row Number Plant Growth ANOVA - Conventional Feed Descriptives . N Mean Std. Deviation Std. Error . ... 95% Confidence Interval for Mean Minimu m Maxim urn Lower Bound Upper Bound Cumulative Growth Medium Sawdust 3 563.97 24.53 14.16 503.04 624.90 538.80 587.80 Medium Sawdust+ Amendment 3 568.27 11.60 6.70 539.45 597.08 555.00 576.50 Medium Total 6 566.12 17.32 7.07 547.94 584.29 538.80 587.80 Leaf Length -mm Medium Sawdust 37.72 .83 .48 35.66 39.79 37.11 38.67 Medium Sawdust+ Amendment 3 38.59 1.05 .61 35.97 41.20 37.94 39.80 Medium Total 6 38.16 .97 .40 37.14 . 39.17 37.11 39.80 Stem Diameter -mm Medium Sawdust 3| 9.33 .44 .25 8.24 10.43 8.88 9.76 Medium Sawdust+ Amendment 3 9.91 1.07 .62 7.26 12.56 9.26 11.14 Medium Total ! 6! 9.62 .79 .32 8.79 10.46 8.88 11.14 339 A N O V A Sum of Squares df Mean Square F Sig. Between Groups 27.74 1 27.74 .08 .80 Cumulative Growth Within Groups 1472.29 4 368.07 Total 1500.03 5 Between Groups 1.12 1 1.12 1.24 .33 Leaf Length - mm Within Groups 3.60 4 .90 Total 4.71 5 Between Groups .50 1 .50 .75 .44 Stem Diameter - mm Within Groups 2.66 4 .67 Total 3.16 5 340 Plant Growth ANOVA - Modified Feed Sum of Squares df Mean Square F Sig. Cumulative Growth Between Groups 1138.82 2 569.41 1.06 .43 Within Groups 2150.09 4 537.52 Total 3288.91 6 Leaf Length - mm Between Groups 1.10 2 .55 1.03 .44 Within Groups 2.13 4 .53 Total 3.23 6 Stem Diameter - mm Between Groups .35 2 .18 1.06 .43 Within Groups .66 4 .17 Total 1.01 6 N Mean Std. Deviation Std. Error . 95% Confidence Interval for Mean Min. Max. Lower Bound Upper Bound Cumulative Growth Medium Sawdust 3 601.33 29.75 17.18 527.43 675.24 568.50 626.50 Medium Sawdust+ Amendment 3 577.03 13.78 7.96 542.79 611.27 561.50 587.80 Medium Amendment 1 572.00 572.00 572.00 Medium Total 7 586.73 23.41 8.85 565.08 608.38 561.50 626.50 Leaf Length -mm Medium Sawdust 3 38.28 1.3 .59 35.72 40.83 37.21 39.26 Medium Sawdust+ Amendment 3 37.60 9.29E-02 5.36E-02 37.37 37.83 37.49 37.66 Medium Amendment 1 37.25 37.25 37.25; Medium Total 7 37.84 .73 .28 37.16 38.52 37.21 39.26 Stem Diameter -mm Medium Sawdust 3 9.26 .56 .32 7.88 10.65 8.62 9.61 Medium Sawdust+ Amendment 3 9.41 .14 8.21E-02 9.06 9.77 9.25 9.51 Medium Amendment i ! 8.73 8.73 8.73 Medium Total 7 9.25 .41 .16 8.87 9.63 8.62 9.61 341 Correlations Cumulative Growth Leaf Length -mm Stem Diameter -mm Adjusted Marketable Yield - kg Medi um Pearson Correlation Cumulative Growth 1.00 .40 .29 .14 -.35 Leaf Length - mm .41 1.00 .81(**) .44 -.08 Stem Diameter - mm .29 .81(**) 1.00 .28 -.10 Adjusted Marketable Yield - kg .14 .44 .28 1.00 -.41 Medium -.35 -.08 -.10 -.41 1.00 Sig. (2-tailed) Cumulative Growth .15 .31 .64 .21 Leaf Length - mm .15 .00 .12 .78 Stem Diameter - mm .31 .00 • .33 .73 Adjusted Marketable Yield - kg .64 .12 .33 .14 Medium .21 .78 .73 .14 N Cumulative Growth 14 14 14 14 14 Leaf Length - mm 14 14 14 14 14 Stem Diameter - mm 14 14 14 14 14 Adjusted Marketable Yield - kg 14 14 14 14 14 Medium 14 14 14 14 14 ** Correlation is significant at the 0.01 level (2-tailed). 342 Plant Growth Data Summary by Row • / Cumulative Growth Leaf Length - mm Stem Diameter - mm Mean Mean Mean ; Sawdust Conventional A 538.80 37.11 8.88 B 587.80 37.39 9.36 C 565.30 38.67 9.76 Modified A 568.50 37.21 8.62 B 626.50 39.26 9.56 C 609.00 38.36 9.61 Sawdust +Amendment Conventional A 555.00 37.94 9.33 B 573.30 38.02 9.26 C 576.50 39.80 11.14 Modified A 581.80 37.66 9.25 B 587.80 37.49 9.48 C 561.50 37.64 9.51 Amendment Modified A 572.00 37.25 8.73 Modified 2 A 546.00 38.14 8.92 Plant Growth Summary by Group Cumulative Growth Leaf Length - mm Stem Diameter - mm Mean Mean Mean Sawdust 1 Conventional 563.97 37.72 9.33 | Modified 601.33 38.28 9.26 Sawdust 1 Conventional 568.27 38.59 9.91 +Amendment 1 Modified 577.03 37.60 9.41 Amendment | Modified 572.00 37.25 8.73 | Modified 2 546.00 38.14 8.92 343 APPENDIX H ANOVA - Test 3 (for Chapter VIII) Root Score Height Mean Range SD Mean Range SD Vermi- Inocula- F O R L - 1.00 .00 .00 55.11 36.50 12.26 Rockwool e *n culite tion F O R L + 1.87 4.00 1.06 48.94 40.50 9.79 > o Amend- Inocula-F O R L - 1.00 .00 .00 51.55 30.50 9.56 E U ment P6 tion F O R L + 1.40 2.00 .74 46.10 36.40 11.10 a •3 F O R L - 1.00 .00 .00 49.68 45.60 14.02 Amend-Covering Vermi- Inocula-ment P6 Covering culite tion F O R L + 1.13 1.00 .35 49.65 24.40 8.81 ANOVA - Root score, groups 1-6, last three dates Root Score Medium Covering Inoculation Mean N Rockwool Vermiculite Total 1.43 30 Amendment P6 Total 1.20 30 Total F O R L - 1.00 30 F O R L + 1.63 30 Total 1.32 60 Amendment P6 Vermiculite Total 1.07 30 Amendment P6 Total 0 Total F O R L - 1.00 15 F O R L + 1.13 15 Total 1.07 30 Total Vermiculite F O R L - 1.00 30 F O R L + 1.50 30 Total 1.25 60 Amendment P6 F O R L - 1.00 15 F O R L + 1.40 15 Total 1.20 30 Total F O R L - 1.00 45 F O R L + 1.47 45 Total 1.23(a) 90 a Grand Mean b Root Score by Medium, Covering, Inoculation c 3-way and higher means are not computed due to the limit on maximum order interaction. 344 ANOVA(a,b) Experimental Method Sum of Squares df Mean Square F Sig. (Combined) 6.97 3 2.32 7.36 .00 Main Effects Medium 2.02 1 2.02 6.39 .02 Covering .82 1 .82 2.59 .11 Inoculation 4.90 1 4.90 15.53 .00 Root Score Model 6.97 • 3 2.32 7.36 .00 Residual 27.13 86 .32 Total 34.10 89 .38 a Root Score by Medium, Covering, Inoculation b Due to empty cells or a singular matrix, higher order interactions have been suppressed. One way ANOVA - Root score, groups 1-6, last 3 dates, rockwool vs. amendment P6 medium • Descriptives , . ' 95% Confidence Interval for Mean N Mean Std. Deviation Std. Error Lower Bound Upper Bound Minim um Maxi mum Rockwool 60 1.32 .72 9.36E-02 1.13 1.50 1.00 5.00 Root Score Med -ium Amend-ment P6 i 30 1.07 .25 4.63E-02 .97 1.16 1.00 2.00 Total | 90! 1.23 .62 6.53E-02 1.10 1.36 1.00 5.00 A N O V A Sum of Squares df Mean Square F Sig. Root Score Between Groups 1.25 1 1.25 3.35 .07 Within Groups 32.85 88 .37 Total 34.10 89 345 T-Test - root score, groups 1-6, rockwool vs. amendment P6 Group Statistics Medium N Mean Std. Deviation Std. Error Mean Root Score Rockwool 60 1.32 .72| 9.36E-02 Amendment P6 30 1.07 ' .25 4.63E-02 Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Differe nee Std. Error Difference 95% Confidence Interval of the Mean Lower Upper Root Score Equal variances assumed 13.80 .00 1.83 88 • .07 .25 .14 -2.15E-02 .52 Equal variances not assumed 2.40 81.51 .02 .25 .10 4.2E-02 .46 Oneway ANOVA - groups 1-6, last 3 dates, rockwool vs. amendment p6, F+ Descriptives N. Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minim um Maxi mum Lower Bound Upper Bound Root Score Med ium Rockwool 30 1.63 .93 .17 1.29 1.98 1.001 5.00 Amendme ntP6 15 1.13 .35 9.09E-02 .94 1.33 1.00 2.00 Total 45 1.47 •81 .12 1.22 1.71 1.001 5.00 346 A N O V A Sum of Squares df Mean Square F Sig. Between Groups 2.50 1 2.50 4.03 .05 Root Score Within Groups 26.70 43 .62 Total 29.20 44 SPAD (leaf colour) groups 1-6 all dates Spad (colour) Mean Range Std Deviation Medium Rockwool Covering Vermiculite Inoculation F O R L - 38.86 36.40 10.65 F O R L + 36.461 30.00 8.55 Amendment P6 Inoculation F O R L - 39.56 39.40 9.21 F O R L + 39.55 31.70 8.45 Amendment P6 Covering Vermiculite Inoculation F O R L - 40.58 31.80 7.76 F O R L + 41.111 30.70 9.46 347 APPENDIX I ANOVA - Test 4 (for Chapter VIII) Test 4 First Scoring Summary Root Score Shoot Height - cm ] Leaf Colour S P A D Mean 1 Range Mean Range ] Mean ] Range Vermiculite F O R L (-) 1.62 1 4 47.55 55.801 37.191 47.70 Rock Wool FORL(+) 2.07 1 3 51.23 17.00 j 44.94 31.10 Amendment P6 F O R L (+) 1.00 0 56.49 39.101 40.27 J 30.30 Amendment P6 Vermiculite F O R L (+) 1.00 1 o. 44.82 25.70 j 36.441 22.20 Amendment P7 Vermiculite F O R L (+) 1.00 1 o 49.24 39.70 j 39.08 j 32.30 Test 4 Second Scoring Summary Root Score Shoot Height - cm Leaf Colour SPAD Mean J Range Mean Range Mean Range Vermiculite F O R L (-) 1.071 1 59.56 12.80 44.63 17.70 Rock Wool | F O R L (+) 2.21 j 4 56.13 46.50 38.96 52.60 Amendment P6 F O R L (+) 1.72 2 61.57 30.10 45.68 18.50 Amendment P6 Vermiculite F O R L (+) 1.00 0 . 48.40 32.00 37.85 26.30 Amendment P7 Vermiculite | F O R L (+) l.oo] 0 44.39 50.00 38.45 30.70 Test 4 Summary Root Score ] Shoot Height - cm Leaf Colour SPAD Mean Range| Mean Range Mean | Range Vermiculite | F O R L (-) 1.33 4 53.8 62.1 41.05 56.50 Rock Wool F O R L (+) 2.14 41 53.7 46.5 41.951 55.50 Amendment P6 F O R L (+) 1.35 2j 59.0 47.7 42.90 35.00 Amendment P6 Vermiculite F O R L (+) 1.00 01 • 46.5 32.0 37.12 j 26.30 Amendment P7 Vermiculite F O R L (+) 1.00 o] 46.8 51.2 38.77] 32.50 348 ANOVA - Overall : Case Processing Summary(a) Cases Included Excluded ] Total N | Percent N | Percent j N J Percent 1531 100.0% oj .o%| 1531 100.0%! a Root Score by Medium, Covering, Inoculation Cell Means(b,c) Root Score Medium Covering Inoculation Mean N Rock Wool Vermiculite Total 1.75 55 Amendment P6 Total 1.35 37 Total F O R L (-) 1.33 27 F O R L (+) 1.69 65 Total 1.59 92 Amendment P6 Vermiculite Total 1.00 27 Amendment P6 Total 0 Total F O R L (-) 0 FORL(+) 1.00 27 Total 1.00 27 Amendment P7 Vermiculite Total 1.00 34 Amendment P6 Total 0 Total F O R L (-) 0 F O R L (+) 1.00 34 Total 1.00 34 Total Vermiculite F O R L (-) 1.33 27 F O R L (+) 1.36 89 Total 1.35 116 Amendment P6 F O R L (-) 0 F O R L (+) 1.35 37 Total 1.35 37 Total F O R L (-) 1.33 27 F O R L (+) 1.36 126 Total 1.35(a) 153 a Grand Mean b Root Score by Medium, Covering, Inoculation c 3-way and higher means are not computed due to the limit on maximum order interaction. 349 ANOVA(a,b) Experimental Method Sum of Squares df Mean Square F Sig. (Combined) 25.08 4 6.27 12.91 .00000 Main Effects Medium 25.07 2 12.53 25.81 .00000 Covering 9.99 1 9.99 20.57 .00001 Inoculation 9.01 1 9.01 18.55 .00003 Root Score Model 25.08 4 6.27 12.91 .00000 Residual 71.86 148 .49 Total 96.94 152 .64 a Root Score by Medium, Covering, Inoculation b Due to empty cells or a singular matrix, higher order interactions have been suppressed. ANOVA - Controls Descriptives N Mean Std. Deviatio n Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Score Inocul -ation F O R L (-) 27 1.33 .96 .18 .95 1.71 1 5 F O R L (+) 28 2.14 1.11 .21 1.71 2.57 1 5 Total 55 1.75 1.11 .15 1.45 2.05 1 5| Shoot Height -cm Inocul -ation F O R L (-) 27 53.78 13.79 2.65 48.33 59.241 4.40 66.50 F O R L (+) 28 53.68 9.34 1.77 50.06 57.30 27.00 73.50 Total 55 53.73 11.63 1.57 50.59 56.87 4.40 73.50 Leaf Colour SPAD Inocul -ation F O R L (-) 27 41.05 10.80 2.08 36.77 45.32 .00 56.50 F O R L (+) 28 41.95 11.65 2.20 37.43 46.47 .00 55.50 Total 55 41.51 11.15 1.50 38.49 44.52 .00 56.50 350 A N O V A Sum of Squares df Mean Square F Sig. Root Score Between Groups 9.01 1 9.01 8.31 .006 Within Groups 57.43 53 1.08 Total 66.44 54 Shoot Height - cm Between Groups .15 1 .15 .0011 .97 i Within Groups 7300.25 53 137.74 Total 7300.39 54 Leaf Colour SPAD Between Groups 11.18 1 11.18 .09 .77 Within Groups 6702.08 53 126.45 Total 6713.26 54 ANOVA - RockwooH- vs Rockwool P6 Covering+ Descriptives N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Score Cove ring Vermicu lite 2 8 2.14 1.11 .21 1.71 2.57 1 5 Amendm ent P6 3 7 1.35 .63 .10 1.14 1.56 1 3 Total 6 5 1.69 .95 .12 1.46 1.93 1 5 Shoot Height - cm Cove ring Vermicu lite 2 8 53.68 9.34 1.77 50.06 57.30 27.00 73.50 Amendm entP6 3 7 58.96 8.32 1.37 56.19 61.74 27.40 75.10 Total 6 5 56.69 9.10 1.13 54.43 58.94 27.00 75.10 Leaf Colour SPAD Cove ring Vermicu lite 2 8 41.95 11.65 2.20 37.43 46.47 .00 55.50 Amendm ent P6 3 7 42.90 7.34 1.21 40.45 45.35 18.50 53.50 Total 6 5 42.49 9.37 1.16 40.17 44.81 .00 55.50 351 A N O V A Sum of Squares df Mean Square F Sig. Between Groups 9.99 1 9.99 13.14 .001 Root Score Within Groups 47.86 63 .76 Total 57.85 64 Between Groups 444.94 1 444.94 5.78 .020 Shoot Height - cm Within Groups 4849.11 63 76.97 Total 5294.06 64 Between Groups 14.38 1 14.38 .16 .69 Leaf Colour SPAD Within Groups 5605.47 63 88.98 Total 5619.85 64 ANOVA - Rockwoolj- vs Amendment P6+ Descriptives N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Score Medium Rock Wool 28 2.14 1.11 .21 1.71 2.57 1 5 Medium Amend-ment P6 27 1.00 .00 .00 1.00 1.00 1 1 Medium Total 55 1.58 .98 .13 1.32 1.85 1 5 Shoot Height - cm Medium Rock Wool 28 53.68 9.34 1.77 50.06 57.30 27.00 73.50 Medium Amend-ment P6 27 46.54 8.80 1.69 43.06 50.03 30.00 62.00 Medium Total 55 50.18 9.69 1.31 47.56 52.80 27.00 73.50 Leaf Colour SPAD Medium Rock Wool 28 41.95 11.65 2.20 37.43 46.47 .00 55.50 Medium Amend-ment P6 27 37.12 7.33 1.41 34.22 40.02 24.10 50.40 Medium Total 55 39.58 9.99 . 1.35 36.88 42.28 .00 55.50 352 ANOVA Sum of Squares df Mean Square F Sig. Root Score Between Groups 17.95 1 17.95 28.46 .000 Within Groups 33.43 53 .63 Total 51.39 54 Shoot Height - cm Between Groups 699.59 1 699.59 8.48 .005 Within Groups 4370.91 53 82.47 Total 5070.50 54 Leaf Colour SPAD Between Groups 320.86 1 320.86 3.36 .073 Within Groups 5065.19 53 95.57 Total 5386.05 54 ANOVA - RockwooH- v s Amendment P7+ Descriptives N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Score Medium Rock Wool 28 2.14 1.11 .21 1.71 2.57 1 5 Medium Amend-ment P7 34 1.00 .00 .00 1.00 1.00 1 1 Medium Total 62 1.52 .94 • 12 1.28 1.75 1 5 Shoot Height - cm Medium Rock Wool 28 53.68 9.34 1.77 50.06 57.30 27.00 73.50 Medium Amend-ment P7 34 46.81 13.11 2.25 42.24 51.39 14.50 65.70 Medium Total 62 49.91 11.98 1.52 46.87| 52.96 14.50 73.50 Leaf Colour SPAD Medium Rock Wool 28 41.95 11.65 2.20 37.43 46.47 .00 55.50 Medium Amend-ment P7 34 38.77 9.41 1.61 35.49 42.05 21.40 53.90 Medium Total 62 40.20 10.51 1.34 37.53 42.87 .00 55.50 353 A N O V A Sum of Squares df Mean Square F Sig. Root Score Between Groups 20.06 1 20.06 36.00 .000 Within Groups 33.43 60 .56 Total 53.48 61 Shoot Height - cm Between Groups 724.03 1 724.03 5.41 .02 Within Groups 8030.90 60 133.85 Total 8754.93 61 Leaf Colour SPAD Between Groups 155.50 1 155.50 1.42 .24 Within Groups 6586.98 60 109.78 Total 6742.49 61 354 APPENDLX J ANOVA - Test 6 (for Chapter VIII) Correlations j Mean Std. Deviation N J Root Score 1.27 .67 168 J Root Health 1.27 .52 168 ] Stem Diameter - mm 6.01 .65 168 ] Shoot Height - cm 50.45 8.86 168 Correlations Root Score Root Health Stem Diameter -mm Shoot Height - cm Pearson Correlation Root Score 1.00 .25(**) -.12 .07 Root Health .25(**) 1.00 -.35(**) -.20(**) Stem Diameter -mm -.12 -.35(**) 1.00 -.03 Shoot Height - cm .07 -.20(**) -.03 1.00 Sig. (2-tailed) Root Score .001 .14 .38 Root Health .001 .00! .01 Stem Diameter -mm .14 .000 .73 Shoot Height - cm .37 .01 .73 N Root Score 168 168 17 168 Root Health 168 168 168 168 Stem Diameter -mm 168 168 168 168 Shoot Height - cm 168 168 168 168 ** Correlation is significant at the 0.01 level (2-tailed). Test 6 Summary 355 Root Health Root Score Shoot Height - cm Stem Diameter - mm Leaf Colour SPAD Root Index Mean Range Mean Range Mean Range Mean Range Mean Range Mean | Range Rock Wool Vermi-culite | F O R L <r) 1.18 2 1.00 0 49.7 33.50 6.3 1.70 45.55 26.80 1.18 2 F O R L (+) 1.50 2 2.18 3 55.1 36.00 5.8 3.30 42.77 32.53 3.36 8 Amend -ment i P6 F O R L (+) 1.14 1 1.25 2 52.5 35.50 6.0 2.00 42.71 25.10 1.50 5 Amend -ment j 5A F O R L (+) 1.11 1 1.00 0 46.9 21.50 6.3 2.10 45.97 21.80 1.11 1 Amend -ment I 5C F O R L (+) 1.36 2 1.11 2 52.4 33.00 5.9 3.30 43.19 30.87 1.57 5 356 ANOVA - Controls Descriptives N Mean Std. Deviatio n Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Health Inocul ation F O R L (-) 28 1.18 .48 8.99E-02 .99 1.36 1 3 F O R L <-+) 28 1.50 .64 .12 1.25 1.75 1 3 Total 56 1.34 .58 7.76E-02 1.18 1.49 1 3 Root Score Inocul ation F O R L 28 1.00 .00 .00 1.00 1.00 1 1 F O R L (+> 28 2.18 1.02 .19 1.78 2.57 1 4 Total 56 1.59 .93 .12 1.34 1.84 1 4 Root Index Inocul ation F O R L (-) 28 1.18 .48 8.99E-02 .99 1.36 1 3 F O R L < + > 28 3.36 2.34 .44 2.45 4.27 1 9 Total 56 2.27 2.00 .27 1.73 2.80 1 9 Shoot Height - cm Inocul-ation F O R L (-) 28 49.70 7.41 1.40 46.83 52.57 35.50 69.00 F O R L (+) 28 55.09 8.57 1.62 51.76 58.41 33.50 69.50 Total 56 52.39 8.39 1.12 50.15 54.64 33.50 69.50 Stem Diameter -mm Inocul-ation F O R L (-) 28 6.29 .43 8.09E-02 6.12 6.46 5.30 7.00 F O R L <•+) 28 5.78 .58 .11 5.55 6.00 3.90 7.20 Total 56 6.03 .57 7.57E-02 5.88 6.18 3.90 7.20 Leaf Colour SPAD Inocul-ation F O R L (-) 28 45.55 6.52 1.23 43.02 48.08 30.27 57.07 F O R L (+) 28 42.77 8.89 1.68 39.33 46.22 23.40 55.93 Total 56 44.16 7.85 1.05 42.06 46.27 23.40 57.07 35,7 ANOVA Sum of Squares df Mean Square F Sig. Between Groups 1.45 1 1.45 4.57 .037 Root Health Within Groups 17.11 54 .32 Total 18.55 55 Between Groups 19.45 1 19.45 37.36 .000 Root Score Within Groups 28.11 54 .52 Total 47.55 55 Between Groups 66.45 1 66.45 23.22 .000 Root Index Within Groups 154.54 54 2.86 Total 220.98 55 Between Groups 406.08 1 406.08 6.33 .015 Shoot Height - cm Within Groups 3464.19 54 64.15 Total 3870.28 55 Between Groups 3.70 1 3.70 14.35 .000 Stem Diameter - mm Within Groups 13.94 54 .26 Total 17.64 55 Between Groups 108.00 1 108.003 1.778 .19 Leaf Colour SPAD Within Groups 3279.85 54 60.738 Total 3387.86 55 358 ANOVA-Ctrl+vs P6+ Descriptives N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Health Covering Vermiculite 28 1.50 .64 .12 1.25 1.75 1 3 Covering Amendment P6 28 1.14 .36 6.73E-02 1.00 1.28 1 2 Covering Total 56 1.32 .54 7.26E-02 1.18 1.47 1 3 Root Score Covering Vermiculite 28 2.18 1.02 .19 1.78 2.57 1 4 Covering Amendment P6 28 1.25 .59 .11 1.02 1.48 1 3 Covering Total 56 1.71 .95 .13 1.46 1.97 1 4 Root Index Covering Vermiculite 28 3.36 2.34! .44 2.45 4.27 1 9 Covering Amendment P6 28 1.50 1.14 .22 1.06 1.94 1 6 Covering Total 56 2.43 2.05) .27 1.88 2.98 1 9 Shoot Height - cm Covering Vermiculite 28 55.09 8.571 1.62 51.76 58.41 33.50 69.50 Covering Amendment P6 28 52.48 7.94 1.50 49.40 55.55 34.00 69.50 Covering Total 56 53.78 8.29 1.11 51.56 56.001 33.50 69.50 Stem Diameter -mm Covering Vermiculite 28 5.78 .58! .11 5.55 6.00 3.90 7.20 Covering Amendment P6 28 6.00 .52 9.78E-02 5.80 6.20 4.80 6.80 Covering Total 56 5.89 .56 7.42E-02 5.74 6.04 3.90 7.20 Leaf Colour SPAD Covering Vermiculite 28 42.77 8.89 1.68 39.33 46.22 23.40 55.93 Covering Amendment P6 28 42.71 6.58 1.24 40.16 45.25 28.37 53.47 Covering Total 56 42.74 7.75 1.04 40.67 44.81 23.40 55.93 359 ANOVA Sum of Squares df Mean Square F Sig. Between Groups 1.79 1 1.79 6.68 .01 Root Health Within Groups 14.44 54 .27 Total 16.21 55 Between Groups .12.07 1 12.07 17.45 .00 Root Score Within Groups 37.36 54 .69 Total 49.43 55 Between Groups 48.29 1 48.29 14.22 .00 Root Index Within Groups 183.43 54 3.40 Total 231.71 55 Between Groups 95.42 1 95.42 1.40 .24 Shoot Height - cm Within Groups 3684.19 54 68.23 Total 3779.61 55 Between Groups .73 1 .73 2.44 .13 Stem Diameter - mm Within Groups 16.22 54 .30 Total 16.95 55 Between Groups 6.79E-02 1 6.790E-02 .001 .974! Leaf Colour SPAD Within Groups 3300.35 54 61.118 Total 3300.41 55 360 ANOVA - Amendments+ Case Processing Summary(a) Cases Included Excluded Total N | Percent N Percent N | Percent 1121 66 .7% 56 33.3% 168 J 100.0% a Root Index by Covering Cell Means(b) Root Index Covering Mean J N Amendment P6 1.501 28 Amendment L5A 1.111 28 Amendment L5C 1.571 28 Total 1.42(a)! 112 a Grand Mean b Root Index by Covering ANOVA(a) Experimental Method Sum of Squares df Mean Square F Sig. ! Root Index Main Effects | Covering 3.74 3 1.25 1.30 .28 Model 3.74 3 1.25 1.30 .28 Residual 103.54 108 .96 Total 107.28 n i .97 a Root Index by Covering 3 6 1 A N O V A - P 6 + v s 5 A + Descriptivcs N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Health s u oi > o U Amend-ment P6 28 1.14 .36 6.73E-02 1.00 1.28 1 2 Amend-ment L 5 A 28 1.11 .31 5.95E-02 .99 1.23 • lj 2 Total 56 1.13 .33 4.46E-02 1.04 1.21 l ! 2 Root Score on B oi > o u Amend-ment P6 28 1.25 .59 .11 1.02 1.48 i 3 Amend-ment L 5 A 28 1.00 .00 .00 1.00 1.00 1 1 Total 56 1.13 .43 5.73E-02 1.01 1.24 l 3 Root Index OS s <u > o U Amend-ment P6 28 1.50 1.14 .22 1.06 1.94 1 6 Amend-ment L 5 A 28 1.11 .31 5.95E-02 .99 1.23 l 2 Total 56 1.30 .85 .11 1.08 1.53 l 6 Shoot Height -cm en s u Oi > i o U Amend-ment P6 28 52.47 7.94 1.50 49.40 55.55 34.00 69.50 Amend-ment -5A 28 46.87 5.03 .95 44.92 48.83 35.50 57.00 Total 56 49.67 7.17 .96 47.76 51.59 34.00; 69.50 Stem Diamete r - mm en B Oi > O U Amend-ment P6 28 6.00 .52 9.78E-02 5.80 6.20 4.80 6.80 Amend-ment L 5 A 28 6.33 .55 .10 6.12 6.54 5.20 7.30 Total 56 6.17 .55 7.39E-02 6.02 6.32 4.80 7.30 Leaf Colour SPAD en B I. > o y Amend-ment P6 28 42.70 6.58 1.24 40.16 45.26 28.37 53.47 Amend-ment L 5 A 28 45.97 4.87 .92 44.09 47.86 31.27 53.07 Total 56 44.33 5.96 .80 42.74 45.94 28.37 53.47 362 ANOVA Sum of Squares df Mean Square F Sig. Between Groups 1.79E-02 1 1.79E-02 .16 .69 Root Health Within Groups 6.10 54 .11 Total 6.13 55 Between Groups .88 1 .88 5.11 .03 Root Score Within Groups 9.25 54 .17 Total 10.13 55 Between Groups 2.16 1 2.16 3.10 .084 Root Index Within Groups 37.68 54 .70 Total 39.84 55 Between Groups 439.04 1 439.04 9.93 .003 Shoot Height - cm Within Groups 2387.05 54 44.21 Total 2826.09 55 Between Groups 1.51 1 1.51 5.34 .025 Stem Diameter - mm Within Groups 15.29 54 .28 Total 16.80 55 Between Groups 149.54 1 149.54 4.47 .039 Leaf Colour SPAD Within Groups 1806.66 54 33.46 Total 1956.19 55 363 A N O V A - L5A+ v s L5C+ Descriptives N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Root Health on e u 41 > O U Amendment L5A 28 1.11 .31 5.95E-02 .99 1.23 1 2 Amendment L5C 28 1.36 .56 .11 1.14 1.57 1 3 Total 56 1.23 .47 6.24E-02 1.11 1.36 1 3 Root Score B La cu > o U Amendment L5A 28 1.00 .00 .00 1.00 1.00 1 1 Amendment L5C 28 1.11 .42 7.87E-02 .95 1.27 1 3 Total 56 1.05 .30 3.96E-02 .97 1.13 1 3 Root Index wo B ' i. > O <J Amendment L5A 28 1.11 .31 5.95E-02 .99 1.23 1 2 Amendment L5C 28 1.57 1.14 .21 1.13 2.01 1 6 Total 56 1.34 .86 .11 1.11 1.57 l ! 6 Shoot Height - cm 01 B u 4> > o U Amendment L5A 28 46.88 5.03 .95 44.92 48.83 35.50 57.00 Amendment L5C 28 52.36 7.20 1.36 49.57 55.16 27.50 60.50 Total 56 49.62 6.75 .90 47.81 51.43 27.50 60.50 Stem Diame ter-mm 01 B L . > o U Amendment L5A 28 6.33 .55 .10 6.12 6.54 5.20 7.30 Amendment L5C 28 5.94 .69 .13 5.68 6.21 3.80 7.10 Total 56 6.14 .64 8.62E-02 5.96 6.31 3.80 7.30 Leaf Colour SPAD Of B V. o U Amendment L5A 28 45.97 4.87 .92 44.09 47.86 31.27 53.07 Amendment L5C 28 43.19 7.95 1.50 40.10 46.27 22.73 53.60 Total 56 44.58 6.68 j .89 42.79 46.37 22.73 53.60 364 ANOVA Sum of Squares df Mean Square F Sig. Between Groups .88 1 .88 4.25 .04 Root Health Within Groups 11.11 54 .21 Total 11.98 55 Between Groups .16 1 .16 1.86 .18 Root Score Within Groups 4.68 54 8.66E-02 Total 4.84 55 Between Groups 3.02 1 3.02 4.34 .04 Root Index Within Groups 37.54 54 .70 Total 40.55 55 Between Groups 421.85 1 421.85 10.93 .002 Shoot Height - cm Within Groups 2085.04 54 38.61 Total 2506.89 55 Between Groups 2.12 1 2.12 5.52 .02! Stem Diameter - mm Within Groups 20.75 54 .38 Total 22.87 55 Between Groups 108.64 1 108.64 2.50 .12 Leaf Colour SPAD Within Groups 2347.00 54 43.46 Total 2455.65 55 365 APPENDIX K ANOVA - Disease suppression yield test (for Chapter VIII) UBC disease suppression yield ANOVA - FORL-Descriptives N Mean Std. Dev. Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Yield - kg E •5 s Sawdust 9 10.69 1.23 .4088 9.75 11.64 8.43 11.89 Sawdust/Amend-ment Mix 9 9.79 1.74 .5787 8.45 11.12 7.05 13.23 Total 18 10.23 1.53 .3609 9.48 11.00 7.05 13.23 Average Fruit Size E a •5 s Sawdust 9 .15 1.67E-02 5.565E-03 .14 .16 .13 .18 Sawdust/Amend-ment Mix 9 .16 1.69E-02 5.633E-03 .15 .17 .14 .19 Total 18 .15 1.66E-! 02 3.918E-03 .15 .16 .13 .19 ANOVA Sum of Squares df Mean Square F Sig. Yield-kg Between Groups 3.70 1 3.70 1.64 .22 Within Groups 36.15 16 2.26 Total 39.85 17 Average Fruit Size Between Groups 1.84E-04 1 1.84E-04 .65 .43 Within Groups 4.51E-03 16 2.82E-04 Total 4.70E-03 17 366 Yield ANOVA - FORL+ Descriptives N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Yield - kg Medium Sawdust 9 5.39 5.36 1.79 1.27 9.50 .00 11.64 Medium Sawdust/Amend-ment Mix 9 8.34 3.32 1.11 5.78 10.89 .00 10.95 Medium Total 18 6.86 4.58 1.08 4.58 9.14 .00 11.64 Average Fruit Size Medium Sawdust 5 .15 5.70E-03 2.55E-03 .14 .15 .14 .15 Medium Sawdust/Amend-ment Mix 8 .16 1.22E-02 4.31E-03 .15 .17 .14 .17 Medium Total 13 .15 1.09E-02 3.02E-03 .15 .16 .14 .17 ANOVA Sum of Squares df Mean Square F Sig. Yield - kg Between Groups 39.22 1 39.22 1.97 .18! Within Groups 317.89 16 19.87 Total , 357.11 17 Average Fruit Size Between Groups 2.58E-04 1 2.58E-04 2.43 .15 Within Groups 1.17E-03 11 1.06E-04 Total 1.43E-03 12 367 Yield ANOVA - FORL+ without plant 4A-1 Descriptives N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Mini mum Maxi mum Lower Bound Upper Bound Yield -kg Medium Sawdust 9 5.39 5.36 1.79 1.27 9.50 .00 11.64 Medium Sawdust/Amend -ment Mix 8 9.38 1.20 .43 8.37 10.39 7.49 10.95 Medium Total 17 7.27 4.38 1.06 5.01 9.52 .00 11.64 Average Fruit Size Medium Sawdust 5 .15 5.69E-03 2.55E-03 .14 .15 .14 .15 Medium Sawdust/Amend -ment Mix 8 .16 1.22E-02 4.31E-03 .15 .17 .14 .17 Medium Total 13 .15 1.09E-02 3.02E-03 .15 .16 .14 .17 ANOVA Sum of Squares df Mean Square F Sig. Yield - kg Between Groups 67.58 1 67.58 4.23 .06 Within Groups 239.66 15 15.98 Total 307.24 16 Average Fruit Size Between Groups 2.58E-04 1 2.58E-04 2.43 .15 Within Groups 1.17E-03 11 1.06E-04 Total 1.43E-03 12 368 UBC disease suppression yield ANOVA - overall Cases Included Excluded | Total N 1 Percent N ] Percent N 1 Percent 36] 100.0% 0] .0%] 36 j 100.0% a Yield - kg by Medium, Inoculation Yield - kg Medium Inoculation Mean N Sawdust FORL (-) 10.69 9 FORL (+) 5.39 9 Total 8.04 18 Sawdust/Amendment Mix FORL (-) 9.79 9 FORL (+) 8.34 9 Total 9.06 18 Total FORL (-) 10.24 18 FORL (+) 6.86 18 Total 8.55(a) 36 a Grand Mean ~ b Yield - kg by Medium, Inoculation Experimental Method Sum of Squares df Mean Square F Sig. (Combined) 112.00 2 56.00 5.06 .012 Main Effects Medium 9.42 1 9.42 .85 .36 Inoculation 102.58 1 102.58 9.27 .005 Yield - kg 2-Way Interactions Medium * Inoculation 33.51 1 33.51 3.03 .09 Model 145.50 3 48.50 4.38 .01 Residual 354.04 32 11.06 Total 499.54 35 14.27 a Yield - kg by Medium, Inoculation 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0058673/manifest

Comment

Related Items