UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Poultry layer manure use in raspberry production : soil nitrogen processes and crop nitrogen uptake Dean, Donna Marie 1996

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

Item Metadata

Download

Media
831-ubc_1997-0059.pdf [ 5.71MB ]
Metadata
JSON: 831-1.0099144.json
JSON-LD: 831-1.0099144-ld.json
RDF/XML (Pretty): 831-1.0099144-rdf.xml
RDF/JSON: 831-1.0099144-rdf.json
Turtle: 831-1.0099144-turtle.txt
N-Triples: 831-1.0099144-rdf-ntriples.txt
Original Record: 831-1.0099144-source.json
Full Text
831-1.0099144-fulltext.txt
Citation
831-1.0099144.ris

Full Text

POULTRY LAYER MANURE USE IN RASPBERRY PRODUCTION SOIL NITROGEN PROCESSES AND CROP NITROGEN UPTAKE by DONNA MARIE DEAN B.Sc, (Hons.) Trent University, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Soil Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1996 © Donna M. Dean, 1996 In presenting this thesis in partial fulfilment of the requirements for ah advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and* study. I, further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives, it is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. '. .,.-. :..?, r ' , \ ; ' Department of S>Q I StlfflOP, The University of British Columbia ' Vancouver, Canada Date j^a\A 3 J_Q] . :•• DE-6 (2/88) ABSTRACT Poultry manure application in excess of red raspberry (Rubus idaeus L.) nitrogen (N) requirement has been suggested as a significant contributor to elevated groundwater nitrate concentrations in the Abbotsford Aquifer. The purpose of this study was to examine the effect of poultry layer manure application on N processes in raspberry fields. An experiment was conducted at a site with no history of manure application in 1992 (C92) and 1993 (C93) and a site with a history of manure application in 1992 (H92) to determine the influence of rate of poultry layer manure application on the temporal and spatial variation in soil moisture, ammonium and nitrate concentrations, and on N uptake in raspberries. A comparison of soil inorganic N content over the growing season showed that inorganic fertilizer applied at a rate of 55 kg N ha"1 had similar crop N availability as poultry layer manure applied at a rate of 100 kg N ha'1, although small differences were observed at C93. Using manure as the sole source of N resulted in no adverse impact on estimated berry yield, primocane vigour, and crop N uptake, as compared to the inorganic fertilizer treatment. Greater depletion of soil inorganic N was measured near the crop row suggesting that manure and/or fertilizer should be placed there. Raspberry N uptake by the above-ground portion of the crop was relatively low (90-100 kg N ha'1), averaged across treatments, while N removal from the field was only in the form of berries (up to 20 kg N ha1) since raspberry canes remain in the field where the plant N is recycled. Soil nitrate in August in the control treatment at H92 was 51 and 83% higher than that at C92 and C93, respectively. A second experiment conducted at C92 and C93 found that manure N recovery as soil inorganic N was approximately 50% within 30 days of manure application in late ii February to early March. Mineralization and nitrification of manure N was rapid at C93 despite the fact that soil temperatures were 4 °C cooler than at C92. Additional experiments showed that the above-ground portion of a spring oats (Avena sativa L.) cover crop can be as high as 100 kg N ha"1 when seeded by early August with no adverse effect on primocanes, however when seeded two weeks later N uptake was only half that amount. TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES viii LIST OF TABLES xi ACKNOWLEDGMENTS xiv 1. INTRODUCTION. 1 2. LITERATURE REVIEW 3 2.1. Abbotsford Aquifer 3 2.2. Overview of the Nitrogen Cycle in Raspberry Production 7 2.2.1. Sources of Inorganic N in the Root Zone 7 2.2.1.1. Atmospheric Deposition and irrigation 8 2.2.1.2. Mineralization of soil organic N 9 2.2.1.3. Nitrification 10 2.2.1.4. Fertilizer Addition 11 2.2.2. Losses of Inorganic N From the Root Zone 13 2.2.2.1. Crop Uptake 13 2.2.2.2. Volatilization 15 2.2.2.3. Denitrification 17 2.2.2.4. Leaching 19 3. MATERIALS AND METHODS 23 3.1. Experimental sites 23 iv 3.1.1. Location and History 23 3.1.2. Soil Characterization 25 3.2. Climatic Parameters 28 3.3. Experiment 1: Poultry Manure Incorporation 29 3.3.1. Experimental Design 29 3.3.2. Soil Ammonium and Nitrate Concentrations 30 3.3.3. Statistical Analysis 32 3.4. Experiment 2: Influence of poultry manure application on soil inorganic N and raspberry N uptake 33 3.4.1. Experimental Design 33 3.4.2. Soil Ammonium and Nitrate Concentrations 36 3.4.3. Calculation of mean ammonium and nitrate concentrations across the raspberry inter-row 37 3.4.4. Plant Parameters 39 3.4.4.1. Floricanes 39 3.4.4.2. Primocanes 41 3.4.5. Statistical Analysis 42 3.5. Experiment 3: Influence of an Inter-row Cover Crop on Primocane Vigour and Soil Inorganic N Concentrations 43 3.5.1. Experimental design 43 3.5.2. Cover crop dry matter yield and N uptake 44 3.5.3. Soil inorganic N and primocane dry matter yield and N uptake 44 3.5.4. Statistical analysis 45 3.6. Experiment 4: Comparison of Date of Seeding and Cover Crop Species 45 v 3.6.1. Experimental Design 45 3.6.2. Cover Crop Dry Matter Yield and N Uptake 46 3.6.3. Statistical Analysis 48 4. RESULTS AND DISCUSSION 49 4.1. Climatic Conditions 49 4.1.1. Precipitation and Irrigation 49 4.1.2. Air and Soil Temperature 49 4.2. Experiment 1: Poultry Manure Incorporation 51 4.2.1. Changes in soil inorganic N 51 4.2.2. Distribution of inorganic N with soil depth 54 4.2.3. Apparent N recovery 54 4.2.4. Discussion 57 4.3. Experiment 2: Influence of poultry manure application on soil inorganic N and crop N uptake 61 4.3.1. Soil moisture content at six inter-row locations over time in 1993 61 4.3.2. Ammonium-N concentration at six inter-row locations over time in 1993 64 4.3.3. Nitrate-N concentrations at six inter-row locations over time in 1993 67 4.3.3.1. Soil ammonium and nitrate content over the growing season 74 4.3.4. Raspberry N Uptake 80 4.3.4.1. Floricanes 80 4.3.4.2. Primocanes 87 4.3.4.3. Crop N uptake and apparent N recovery 87 4.3.4.4. Discussion 89 vi 4.4. Experiment 3: Influence of an Inter-row Cover Crop on Primocane Growth and Vigour and Soil Inorganic N Content 97 4.4.1. Discussion 100 4.5. Experiment 4: Comparison of Cover Crop Species and Date of Seeding 102 4.5.1. Discussion 106 5. IMPLICATIONS OF THE RESEARCH 109 6. REFERENCES CITED 115 APPENDIX A 124 APPENDIX B 127 APPENDIX C 129 vii LIST OF FIGURES Figure 2.1. Location of the Abbotsford Aquifer 4 Figure 2.2. Mean monthly precipitation and potential evaporation at the Abbotsford Airport for 1961-1990 (AES unpublished data) 5 Figure 2.3. Timing of N uptake by raspberry floricanes and primocanes (from Kowalenko 1994a) .' 14 Figure 3.1. Location of experimental sites 23 Figure 3.2. Inter-row locations used for measuring soil parameters 27 Figure 4.1. Monthly total precipitation for 1992 and 1993 at the Abbotsford Airport compared to a 30-year (1961-1990) mean (AES unpublished data) 50 Figure 4.2. Soil ammonium- and nitrate-N distribution with depth, averaged across treatments which received manure: before manure application in A) 1992 and B) 1993; approximately one week after manure application in C) 1992 and D) 1993; and approximately one month after manure application in E) 1992 and F) 1993.... 53 Figure 4.3. Apparent manure N recovery as soil inorganic N to 60 cm depth following application of 400 kg total N ha"1 as poultry layer manure; 7 and 32 days following manure application in 1992, and 10 and 30 days following manure application in 1993 55 Figure 4.4. Soil moisture content (g g"1) distribution with depth and inter-row location averaged for the control and 200 kg N ha"1 as poultry layer manure treatments for four sampling dates in 1993 at the Columbia site (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations) 63 viii Figure 4.5. Water depth equivalent to 60 cm depth, averaged for the control and 200 kg N ha'1 as poultry layer manure treatments, for each inter-row location on four sampling dates at the Columbia site in 1993 64 Figure 4.6. Soil ammonium-N (mg L"1) distribution with depth and inter-row location averaged for the controland 200 kg N ha'1 as poultry layer manure treatments for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations) 66 Figure 4.7. Mean ammonium-N concentration for the control and 200 kg N ha"1 as poultry layer manure treatments, averaged across soil depth for each inter-row location on four sampling dates at the Columbia site in 1993 68 Figure 4.8. Mean ammonium-N concentrations for the control and 200 kg N ha'1 as poultry layer manure treatments, averaged across inter-row location for each depth interval on four sampling dates at the Columbia site in 1993 68 Figure 4.9. Soil nitrate-N (mg L'1) distribution with depth and inter-row location for the control treatment for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations) 69 Figure 4.10. Soil nitrate-N (mg L"1) distribution with depth and inter-row location for the 200 kg N ha'1 as poultry layer manure treatment for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations) 70 Figure 4.11. Nitrate-N concentration averaged across inter-row location for each depth interval for the control and 200 kg N ha'1 applied as poultry layer manure treatments on four dates at the Columbia site in 1993 72 ix Figure 4.12. Soil ammonium- and nitrate-N content to 60 cm over the growing season at the Columbia site in 1992 and 1993, and the Huntingdon site in 1992 76 Figure 4.13. Relationship between the berry yield estimate and the dry matter yield of the fruiting cluster on a per plot basis at three sites 81 Figure 4.14. Soil inorganic N content to 60 cm depth for August and October, and the difference in soil nitrate content to 60 cm depth between August and October, in two years as influenced by the application of poultry layer manure and the use of an oats inter-row cover crop (C-control, 200M-200 kg N ha'1 as poultry layer manure, CC-cover crop) 98 Figure 4.15. Influence of seeding date on cover crop a) dry matter yield, b) N concentration, and c) N uptake in two years (Dry matter yield and N uptake were modified assuming that the cover crop was present on only 75% of the field) 103 X LIST OF TABLES Table 3.1 Soil physical and chemical characteristics for the three soil depths at the experimental sites 26 Table 3.2 Soil bulk density (g cm3) for the three soil depths at the experimental sites..28 Table 3.3 Characteristics of the poultry layer manure used in 1992 and 1993 (Analysis performed by a commercial laboratory on an as received basis) 30 Table 3.4 Manure incorporation and soil sampling dates, and day numbers for Experiment 1 in1992 and 1993 31 Table 3.5 Contrast coefficients used to test for significant differences among treatments means at one week and one month for Experiment 1 in two years 33 Table 3.6 Soil test results for available nutrients (mg L'1) for the Columbia site in 1992 and 1993 and the Huntingdon site in 1992 35 Table 3.7 Soil sampling dates and plant growth stages at the Columbia site in 1992 and 1993 and the Huntingdon site in 1992 38 Table 3.8 Weighting factors used to test for significant differences among six procedures for estimating mean soil inorganic N on a kg N ha"1 basis (The first digit refers to the number of sampling locations, U-uniform, and N-non-uniform) 39 Table 3.9 Contrast coefficients used to test for significant differences among treatment means for the Columbia and Huntingdon sites for Experiment 42 Table 3.10 Contrast coefficients used to test for significant differences among treatment means for Experiment 3 45 Table 3.11 Cover crop seeding dates and day number for 1992 and 1993 for Experiment 4 47 Table 3.12 Contrast coefficients used to test for significant differences among treatment means in Experiment 4 in 1992 and 1993 48 xi Table 4.1. Mean monthly air temperature for 1992 and 1993 at the Columbia site as compared to the 30-year (1951-1980) mean (Environment Canada 1985), and mean monthly soil temperature at three depths at the Columbia site in 1992 and 1993 52 Table 4.2 Statistical significance of apparent manure N recovery (%) from the application of layer manure as influenced by time to manure incorporation in two years, and calculated approximately one week and one month following manure application 56 Table 4.3 Statistical significance of soil moisture content (g g~1), ammonium-N concentration (mg L1), and nitrate-N concentration (mg L'1) as influenced by rate of N application as poultry layer manure, inter-row location, depth and time 62 Table 4.4. Statistical significance of six different procedures for calculating soil inorganic N when more than one inter-row location was sampled 73 Table 4.5., Matrix of pairwise comparison probabilities using Fisher's least-significant-difference test 74 Table 4.6 Statistical significance of soil ammonium (kg ha"1) and nitrate (kg ha'1) content as influenced by rate of N application as poultry layer manure and inorganic fertilizer and sampling date at three sites 77 Table 4.7 Soil nitrate-N to 60 cm (kg ha1) on the August sampling date for the C, 100M, 200M, and 55F treatments 79 Table 4.8 Statistical significance of nitrate content (kg ha1) on the August sampling date for Columbia 1992, Columbia 1993 and Huntingdon 1992 for the C, 100M, 200M, and 55F treatments 80 xii Table 4.9 Dry matter yield, N concentration, and N uptake of the fruiting cluster, stem and whole floricane as influenced by the rate of N application as poultry layer manure and inorganic fertilizer at three sites 83 Table 4.10 Statistical significance of dry matter yield (t ha1), N concentration (%), and N uptake (kg ha'1) of the stem, fruiting cluster, and whole floricane as influenced by the rate of N application as poultry layer manure and inorganic fertilizer at three sites 85 Table 4.11 Statistical significance of primocane dry matter yield (t ha'1), N concentration (%), N uptake (kg ha1), and mean cane weight (g cane1) as influenced by the rate of N application as poultry manure and inorganic fertilizer for two years at the Columbia site 88 Table 4.12 Apparent N recovery (%) in the whole crop at the Columbia site in two years.88 Table 4.13 Statistical significance of soil ammonium- and nitrate-N (kg ha1) as influenced by N application as poultry layer manure and an inter-row cover crop (CC) for August and October, and the difference between August and October, in two years 100 Table 4.14 Statistical significance of dry matter yield (t ha'1), N concentration (%), and N uptake (kg ha'1) as influenced by date of seeding and cover crop species in two years 102 Table 4.15 Dry matter yield, N concentration and total N uptake for the oats, barley and spring wheat seeded in mid-August in two years 105 Table 4.16 Growth stage of cover crops at time of harvest expressed using Zadoks' decimal code (growth stage), and height of the above-ground portion of the cover crop in 1992 and 1993 106 xiii ACKNOWLEDGMENTS The research for this thesis would not have been possible without funding from the Canada-B.C. Soil Conservation Program, the B.C. Raspberry Growers' Association, and the FSAM II program. The co-operation of Mr. Gurmeit Brar and Mr. Gurmail Shidu for use of their land is greatly appreciated. I would like to thank, Dr. Grant Kowalenko and Dr. John Paul, (Pacific Agri-Food Research Centre, Agassiz, BC) for their input to the experimental design and sharing with me their expertise on the subjects of raspberries, nitrogen and manure. Dr. Art Bomke, Dr. Mike Novak (Soil Science Department, UBC), and Mr. Kevin Chipperfield (Sustainable Poultry Farming Group, Abbotsford, BC) also offered their support and direction. I would especially like to thank Dr. Bernie Zebarth, (Pacific Agri-Food Research Centre, Summerland, BC) for his role in experimental design, continued support and dedication. Several people assisted with soil and plant sampling and analysis and I would like to especially thank Mr. Brian Harding, and Mrs. Marianne Bickle (Pacific Agri-Food Research Centre, Agassiz, BC) for their enthusiasm, organizational skills. Finally I thank my friends and family, especially Bob, for their continued support and encouragement. xiv 1. INTRODUCTION Groundwater is an important natural resource. In British Columbia, approximately 22% of the population relies on groundwater for drinking water (Environment Canada 1990). Groundwater is also an important source of water for agricultural and industrial use. Groundwater supplies tend to be more dependable and uniform in quality than surface waters making them a desired natural resource (Foweraker 1994). Contamination of groundwater by nitrates and pesticides causes concern especially when alternative sources of water are not readily available. Nitrate contaminated drinking water is known to cause infantile methemoglobenemia (blue baby syndrome), and is also a suspected carcinogen (Shuval and Gruener 1977). Groundwater remediation is usually either very expensive or impractical. The Abbotsford Aquifer, located in the Lower Fraser Valley of British Columbia, is one example of a nitrate contaminated aquifer. Poultry manure nitrogen application to red raspberries {Rubus idaeus L.) in excess of the crop requirement has been suggested as a significant contributor to the elevated nitrate concentrations in the Abbotsford Aquifer (Liebscher et al. 1992; Wassenaar 1995; Zebarth and Paul 1995). British Columbia accounts for 89% of Canada's raspberry production, the majority of which is produced in the Lower Fraser Valley on or near the Abbotsford Aquifer (Zbeetnoff Consulting 1991). In addition, approximately 60% of British Columbia's poultry production is on or near the aquifer. Much of the manure produced by the poultry industry remains on the aquifer where it is used as an organic soil amendment for crops including raspberries. Using a nitrogen budget, Zebarth et al. (1994) estimated that N is applied to the cropped land over the Abbotsford Aquifer at an annual rate of 338 kg ha 1, 1 taking into account the addition of manure (71%), inorganic fertilizers (10%), and atmospheric deposition (19%). In comparison, the recommended rate for inorganic fertilizer application to raspberries is 55 kg N ha'1 for raspberries (BC Ministry of Agriculture Fisheries and Food (BCMAFF) 1994). The goal, therefore is to manage N such that optimum raspberry yields are obtained while minimizing nitrate leaching. However, research on N cycling in raspberry fields is limited, particularly where manure is used as a source of N. The purpose of this study was to examine the effect of poultry layer manure application to raspberry fields over the Abbotsford Aquifer on soil N processes and crop N uptake. Cover crops, which have been proposed as one approach to minimizing nitrate leaching, were also studied (Meisinger 1991). Specific objectives were: 1. to determine the influence of the time of poultry layer manure incorporation on the recovery of manure N as soil extractable inorganic N over a 30-day period, 2. to determine the influence of the rate of N application as poultry layer manure on the spatial and temporal variation in soil moisture, ammonium and nitrate concentrations and on N uptake by raspberries, 3. to determine the influence of the rate of poultry layer manure N application and a fall seeded inter-row cover crop on primocane vigour and soil inorganic N, and 4. to determine the influence of time of seeding and cover crop species on N uptake by inter-row cover crops in a raspberry field. 2 2. LITERATURE REVIEW 2.1. Abbotsford Aquifer The Abbotsford Aquifer is located in the Fraser Lowland in southwestern British Columbia and northwestern Washington State, approximately 75 km east of Vancouver (Figure 2.1). The aquifer is referred to as the Sumas Aquifer in Washington State. The Fraser Lowland, which is roughly triangular in shape, lies between the Coast Mountains to the north, the Cascade Mountains to the east and southeast and the Strait of Georgia to the west. The Abbotsford Aquifer is the largest of about 200 aquifers in the Fraser Lowland and covers an area of approximately 200 km2, of which 100 km2 is in British Columbia with the remainder in Washington State (Liebscher et al. 1992). The Abbotsford Aquifer will hereafter be taken to refer to the portion of the aquifer located in Canada. The glaciofluvial sands and gravels which form the aquifer were deposited during the glacioclimatic period of the last 2 million years (Liebscher et al. 1992). The lower boundary of the aquifer, which is underlain by glaciomarine and marine clays of low permeability, has not yet been defined but well records indicate that it is at least 70 m thick in places (Halstead 1986). The Abbotsford Aquifer is particularly susceptible to contamination by water soluble contaminants due to the unconfined nature of the upper boundary of most of the aquifer, the highly permeable sands and gravels within and overlying the aquifer, the mild, wet climatic conditions, and the intensive agricultural land use activities which occur over the aquifer. Abbotsford has a mean annual precipitation of 1596 mm, 94% of 3 Figure 2.1. Location of the Abbotsford Aquifer. 4 which falls as rain, and 70% of which falls between October and March (Atmospheric Environment Service (AES) unpublished data). Based on 30 year mean (1961-1990) climatic data for Abbotsford (AES unpublished data), total precipitation exceeds potential evaporation from September to April leading to a high leaching potential during those months (Figure 2.2). Potential evaporation was estimated from maximum and minimum daily temperatures at the Abbotsford Airport and solar energy at the top of the atmosphere (Baier and Robertson 1965). Irrigation water is applied to most raspberry fields in the summer months to offset the water deficit. The net water surplus in winter, gives an estimated annual recharge rate of 960 mm over the Abbotsford Aquifer. 350 i Month Figure 2.2. Mean monthly precipitation and potential evaporation at the Abbotsford Airport for 1961-1990 (AES unpublished data). The groundwater flow patterns in the aquifer are complex, however flow is generally to the south (Liebscher et al. 1992). The southerly flow of the groundwater in the aquifer raises concerns regarding the transport of contaminated groundwater across the international boundary. 5 Land use over the Canadian portion of the Abbotsford Aquifer is primarily agricultural, including raspberry, forage grass (Dactylis glomerata L.), and other miscellaneous horticultural crops (Szeto et al. 1994). Approximately 20% of the land area is urban (Liebscher et al. 1992). Raspberries have been the dominant crop produced over the aquifer since the 1960's. The well drained soil over the aquifer, moderate winters and relatively cool humid summers make conditions ideal for raspberry growth (Schreier 1983). Kohut (1987) estimated that a total of 245 million m3 of groundwater is stored in the Abbotsford Aquifer. Groundwater withdrawal by industry (41%), municipal wells (34%), irrigation (21%), and domestic wells (4%) from the aquifer in 1985 was approximately 12 million m3. The Abbotsford Aquifer is one source of drinking water for the City of Abbotsford, and the only source of drinking water for some residents in the south part of the city. Nitrate-N concentrations above the 10 mg L'1 Canadian Drinking Water Guideline have frequently been measured in the Abbotsford Aquifer (Kwong 1986; Kohut et al. 1989; Liebscher et al. 1992; Gartner Lee Ltd. 1993). The highest nitrate-N concentrations were located in areas with a high density of poultry farms and raspberry fields (Liebscher et al. 1992; Gartner Lee Ltd. 1993). More detailed information on the geology and hydrogeology of the aquifer can be found in Armstrong (1959; 1984), Halstead (1986), Kohut (1987) and Kohut et al. (1989), while summaries are found in Liebscher et al. (1992), and Gartner Lee Ltd. (1993). 6 2.2. Overview of the Nitrogen Cycle in Raspberry Production Nitrogen is an essential plant nutrient involved in the production of amino acids, proteins and enzymes, which are fundamental to the function of plants. Nitrogen occurs in nature in a variety of oxidation states, however plant roots absorb N primarily as ammonium' and nitrate. The quantity of available soil N as ammonium and nitrate, is often small compared to the amount of N required for plant growth (Brady 1990). Soil N is largely part of organic matter and mineral material and in general, only a few kg N ha"1 exists in available forms compared to the total in soils, which often exceeds 4,000 kg N ha1 (Stevenson 1982). The N cycle is the transport of various forms of N in soil, plants, animals and the atmosphere, and is generally well understood (Stevenson 1982; Brady 1990). Each N transformation process responds differently to environmental factors and farm management practices, therefore N cycling is very site specific. Consequently the best N management recommendations are based on local research (Kowalenko 1987). It is convenient to discuss the N cycle in raspberry production over the Abbotsford Aquifer in terms of processes that add or remove inorganic forms of N to and from the soil root zone. 2.2.1. Sources of Inorganic N in the Root Zone Processes by which inorganic N is added to the soil root zone include atmospheric deposition, irrigation, mineralization of soil organic N, and fertilizer addition. 7 2.2.1.1. Atmospheric Deposition and Irrigation Atmospheric deposition can include both wet and dry deposition. Deposition of ammonium and nitrate can be significant, especially in areas of intensive agricultural activity. Welte and Timmerman (1987) estimated that 60-70% of the estimated 67 kg N ha"1 that is added to the atmosphere in Germany each year, by the burning of fossil fuels for cars and tractors, atmospheric drifts, sewage disposal, and livestock production is re-deposited on the earth again by wet and dry deposition. Much of the N volatilized from agricultural activities is deposited nearby (Berendse et al. 1988). Total losses of N as ammonia from farms in South Abbotsford over the Abbotsford Aquifer are estimated at 123 kg ha"1 (Brisbin 1996). No estimates of atmospheric dry deposition, and very few wet deposition measurements, have been made in the Lower Fraser Valley. Vet et al. (1983) studied the nitrate-N concentration of rainwater at the Vancouver International Airport between August 1977 and December 1979. The mean nitrate-N content of the rainwater was 0.3 kg ha"1 mo"1 for 19 of the 36 months in the study period; an average of approximately 3.6 kg ha'1 yr1 assuming the rate of deposition was uniform throughout the year. Data from the remaining 17 months were unusable. Irrigation of raspberry fields using nitrate enriched groundwater is another source of soil inorganic N. Using Kohut's (1987) estimated irrigation water use of between 20 and 30 cm during the summer months, and assuming a mean groundwater nitrate-N concentration of 10 mg L"1, the rate of nitrate-N added to the soil in this manner may be as high as 20 to 30 kg N ha"1. 8 2.2.1.2. Mineralization of soil organic N Mineralization of soil organic matter can be a significant source of inorganic N to the soil root zone. Since mineralization is a microbial process, soil and climatic factors that favour microbial growth enhance mineralization. The rate of mineralization increases with increasing soil oxygen concentrations and levels out as the percentage of oxygen in the soil air approaches 20%, the amount present in the atmosphere, although mineralization can be carried out at low oxygen levels. Mineralization of soil N is maximal at field capacity of a soil (Justice and Smith 1962; Cassman and Munns 1980). Cassman and Munns (1980) found the rate of soil N mineralization to increase linearly with increasing soil moisture when water was simply added to dry soil, and to increase exponentially with increasing soil moisture when water was drained from a saturated soil. Soil N mineralization is strongly dependent on soil temperature occurring between 0 and 40 °C (Reddy 1982; Stanford et al. 1973). Stanford et al. (1977) found maximum rates of mineralization between 30 and 35 °C and that in general the rate of nitrogen mineralization doubled for every 10 °C increase in temperature. Maidl and Fischbeck (1989) found a correlation between temperature and mineralization rate when they compared two sites with differing long-term applications of pig slurry. The site with a history of slurry application had significant mineralization at 0 °C whereas a second site with no history of slurry application had no mineralization until soil temperatures exceeded 4 °C. Stanford and Smith (1972) also found that soils with a history of manure application had a higher percentage of mineralizable N than non-manured soils when the net mineralization of 39 soils were compared by incubation at 35 °C over a 30 week period. 9 Kowalenko (1993) estimated that the organic N content of the soil over the Abbotsford Aquifer is 7000 kg N ha'1 to a depth of 30 cm, and that if only 1% of that organic N was mineralized in the growing season, it would amount to 70 kg N ha 1. Soil organic matter in the soils in raspberry production over the aquifer includes plant residues from the raspberry crop and from inter-row cover crops if they are planted. With the exception of the harvested berries, all plant residues from a raspberry crop are cultivated into the soil. Mineralization of soil organic N can occur in any month of the year over the Abbotsford Aquifer because of the mild winter temperatures. 2.2.1.3. Nitrification Nitrification is the transformation of ammonium to nitrate. Although nitrification does not increase inorganic N in the root zone, it is considered here because much of the nitrate in the root zone occurs as a result of nitrification of ammonium added to the soil through manure or fertilizer addition and through mineralization of soil organic N. Nitrifying bacteria are obligate autotrophic aerobes therefore adequate soil aeration, which is closely tied to soil moisture, is essential for nitrification to occur. Soils with high air-filled porosity provide good soil oxygen levels for the nitrification process, while restrictions to the flow of oxygen from the soil surface to the soil profile may result in delayed nitrification. Hadas et al. (1983) found a delay in nitrification at 14 and 35 °C manifested as an accumulation of ammonium at those temperatures, whereas at 25 °C, nitrification proceeded rapidly. Kowalenko and Cameron (1976) also found that nitrification was optimal at 25 °C. The range of soil pH values over which nitrification takes place is 10 between 5.5 and 10.0, the optimum being between 7.5 and 8.5. Nitrification can take place at pH values lower than 5.5, however the rates are significantly reduced. 2.2.1.4. Fertilizer Addition Fertilizer additions of N include both inorganic fertilizers and manure. The BC Ministry of Agriculture Fisheries and Food (1994) recommends an annual inorganic fertilizer N application rate of 55 kg N ha 1 for raspberry production. Chipperfield (1992a) estimated that actual rates of inorganic fertilizer N application are close to 50 and 70 kg N ha'1 for manured and non-manured raspberry fields, respectively, based on the results of a survey of 21 raspberry growers over the Abbotsford Aquifer. Manure N can be broadly divided into inorganic N and organic N. The inorganic N is primarily in the ammonium form, which is immediately available for plant growth. An estimated 50% of liquid cattle manure N is in the organic form and twenty percent of the organic N is assumed to be mineralized within the growing season (Beauchamp 1983). In comparison, poultry manure has a higher proportion of N in the organic form, however the organic N is more readily mineralizable (Chescheir et al. 1986; Chae and Tabatabai 1986). Bitzer and Sims (1988) found that 62% of poultry manure N was in the organic form based on 20 samples and predicted that 60% of the organic N would be plant available during the growing season. They found that when incorporated into a sandy loam soil, on average 66% of the organic N was mineralized within 140 days and found large quantities of inorganic N in the soil within two weeks of application. Sims (1987) found that 72% (based on nine samples) of the total N in poultry manure was in the organic form, and predicted that 60% would be plant available during 11 the growing season. In a 150 day study in which rates of mineralization of three poultry manures were compared, he found that most of the mineralization occurred in the first 90 days at 25 and 40 °C. For two of the manures, 30 to 60% of the organic N was mineralized when moisture was not limiting. Pratt et al. (1973) calculated a decay series, which expresses the rate of mineralization for successive years for poultry manure application, for the irrigated lands of California valleys. The decay series was 0.90, 0.10, and 0.05, indicating that on average 90% of the organic N would be mineralized in the year of application, and 10% and 5% of the remaining manure N would be available in the second and third years, respectively. The decay series illustrates that manure application in previous years must be taken into account when predicting the availability of soil N. The high rate of mineralization of organic N in poultry manure may be due in part to its low C:N ratio (Bitzer and Sims 1988). In a study which compared the characteristics of 20 poultry manures, Bitzer and Sims (1988) found that the C:N ratio ranged from one to 27 with an average of nine. Also, and perhaps more importantly, poultry manure may have a larger fraction of its organic N in easily mineralizable forms (Gale and Gilmour 1986). The soil type to which manure is applied can also affect how readily the organic N is mineralized. Chae and Tabatabai (1986) found that mineralization rates were related to soil texture when they compared mineralization rates in poultry manure treated soils. The apparent N recovery, during 26 weeks of incubation at 30 °C, taking into account mineralization of soil N in an untreated soil, ranged from 21 % in a fine silty soil to 67% in a fine montmorillonitic soil, indicating that the clay soil was more conducive to mineralization, probably due to its higher water holding capacity. 12 The recommended maximum rate of poultry manure application to raspberries is 100 kg N ha'1 when no other fertilizers are used (BC Ministry of Agriculture Fisheries and Food (BCMAFF) 1992). Actual rates of poultry manure N applied are probably significantly higher (Zebarth et al. 1994; Liebscher et al. 1992). 2.2.2. Losses of Inorganic N From the Root Zone Processes by which inorganic N is lost from the soil root zone include crop uptake, volatilization, denitrification, and leaching. 2.2.2.1. Crop Uptake Raspberry plants have perennial roots and biennial tops, with new above-ground canes developing each year (Crandall 1995). Each year, the previous season's canes called primocanes produce fruit and are then called floricanes. After fruit production the floricanes die. Primocanes grow and take up N later in the spring than the floricanes and continue growing until late in the fall, whereas floricanes grow leaves and laterals early in the season with dry matter and N uptake nearly complete by mid-summer when they begin to senesce, losing dry matter and N, mostly as leaves (Kowalenko 1994a) (Figure 2.3). Total N accumulation by the above-ground portion of a raspberry crop averaged 107 kg N ha'1 over four years in a study by Kowalenko (1994a) over the Abbotsford Aquifer. In comparison, Wood et al. (1962) and Wright and Waister (1990) measured total N accumulation in the above-ground portion of the raspberry crop to be 79 and 164 kg N ha"1, respectively. The inter-row spacing in the studies by Wood et al. (1962) in England, and Wright and Waister (1980) in Scotland, were 1.8 m as compared to 3 m in 13 the study by Kowalenko (1994a). The 3 m inter-row spacing accommodates the space required for machinery including tractors and mechanical harvesters and is standard practice in the Lower Fraser Valley. Kowalenko (1994b) found that only 15 kg N ha"1 was present in the fruit, the only portion of the crop removed from the field. May June July Aug Sept Oct Month Figure 2.3. Timing of N uptake by raspberry floricanes and primocanes (from Kowalenko 1994a). No research has been done in the Lower Fraser Valley on raspberry root distribution, although studies have been conducted in Washington State, the former Soviet Union and Britain. Root excavation studies have shown that raspberry roots are concentrated in the upper 30 cm of soil (Pehoto 1968; Voroncihina 1967), although some may penetrate to a depth of 70-80 cm (Voroncihina 1967). Crandall (1980) found that in a well drained soil, raspberry roots reach a depth of nearly 2 m. Christensen (1947), in a root excavation study of four raspberry varieties, found that an average 72, 14 21, 4, 2, and 1% of the roots were in the 0-25, 25-50, 50-75, 75-100 and >100 cm depth intervals, respectively. A study by Bristow and Brun (1987) showed that compaction of the raspberry inter-row by tractor wheels and mechanical harvester traffic resulted in higher root densities near the plant and that roots grew beneath the compacted band. The use of cover crops in raspberry inter-rows can also remove soil N from the root zone, although little research has been done on the quantity of N uptake. Bomke and Temple (1994), in a one year study in the Lower Fraser Valley, found that spring oats (Avena sativa L.) seeded in the third weeks of August and September, 1993, and harvested in November, had an above-ground N uptake of 42 and 23 kg N ha'1, respectively, in a cash crop field. Seeding of cover crops following raspberry harvest was practiced on two of the 21 raspberry farms over the Abbotsford Aquifer surveyed by Chipperfield (1992a). Although cover crops could be effective in taking up excess fall soil inorganic N, they may have a detrimental effect on primocanes, which continue to grow into the fall, by competing for nutrients and water. Zebarth et al. (1993) found that an inter-row groundcover of sheep's fescue (Festuca ovina L.) and perennial ryegrass (Lolium perenne L.) reduced raspberry crop vigour, as indicated by smaller cane diameter, but had only limited effects on berry yield, over a five year study. 2.2.2.2. Volatilization Following field application of manure and/or fertilizer, a significant loss of N may occur as a result of ammonia volatilization. Ammonia volatilization from N compounds already in the soil may occur simultaneously, however their contribution to the total 15 ammonia volatilized will likely be insignificant compared to that originating from the applied manure (Beauchamp et al. 1978). Soil properties, including elevated soil pH and calcium carbonate content, can increase the rate at which ammonia may be volatilized from surface applied manure by governing the adsorption of ammonium and conversion of ammonium to ammonia. High soil pH pushes the following reaction to the right: NH 4 + <=> NH3. Climatic factors relating to drying conditions at the soil surface such as high temperatures, rapid air movement and low humidity also enhance ammonia volatilization. The importance of temperature in controlling the rate of ammonia volatilization is illustrated by the diurnal pattern of volatilization, which has been observed by Beauchamp et al. (1982) and Van der Molen et al. (1990). The rate of ammonia volatilization is usually greatest at mid-day and least in the early hours of the morning. Many authors attribute this pattern to the higher temperatures, and hence the higher evaporative potentials, at mid-day (Adriano et al. 1974; Laurer et al. 1976; Hoff et al. 1981; and Beauchamp et al. 1982). Hoff et al. (1981) found that the rate of ammonia loss from liquid swine manure increased with increasing temperature and air movement. Brunke et al. (1988) suggested that ammonia volatilization potential may be estimated with the hay drying index, which is based upon mean temperature, wind speed, and relative humidity, and is available in some agrometeorological forecasts. Lauer et al. (1976) found that volatilization of ammonia from dairy manure spread on the soil surface was maximized under sustained drying conditions. Molloy and Tunney (1983) in a laboratory study of ammonia volatilization from cattle and pig slurry found a positive relationship between moisture loss and ammonia loss. 16 Losses of ammonia from manure have been found to occur relatively quickly after manure application. Lockyer and Pain (1989) found that the loss of ammonia was often in excess of 40% within six days following manure application. In that study, which compared various manure types, poultry slurry had the highest rate of loss at 83%, while air dried poultry manure had the lowest rate of volatilization at 21 %. They also found that in most cases 80% or more of the ammonia loss from land-applied manure occurred within 48 hours of application. High soil temperatures and dry air were thought to be contributing factors to the high ammonia volatilization. Rapid ammonia volatilization following manure application can be avoided by incorporation of the manure, by injection of manure into the soil or by tillage following manure application. Laurer et al. (1976) studied the effect of manure incorporation on ammonia volatilization. They found that plowing down of dairy manure within six days of application did not prevent loss of up to 83% of ammonia by volatilization, indicating that incorporation must follow application as soon as possible to avoid loss of ammonia. The soils over the Abbotsford Aquifer are acidic and have a low calcium carbonate content. Climatic and manure management factors may therefore have the greatest influence on volatilization rates of ammonia from manure applied over the aquifer. Most of the manure applied over the Abbotsford Aquifer is poultry, and is generally spread in late February or in March, months that are generally wet, therefore volatilization is probably low. 2.2.2.3. Denitrification Loss of soil nitrate by denitrification can occur under low soil oxygen concentrations. Sexstone et al. (1985) found elevated denitrification rates to be 17 associated with rainfall or irrigation events that exceeded 1 cm of water. They found that 38 and 55% of the total N loss by denitrification occurred within 48 h of rainfall or irrigation for a sandy loam and a clay loam soil, respectively. The maximum denitrification rate occurred within 1 to 2 hours in the sandy loam soil whereas it took 12 hours for maximum denitrification to occur in the clay loam soil. The characteristic lag period reflects the time necessary to establish conditions for denitrification following soil wetting (Sexstone et al. 1985). Soil amendment with manure supplies readily available carbon thereby providing energy to the denitrifying bacteria, and increasing the rate of microbial activity, which may result in depleted oxygen levels. In addition, manure application may result in surface sealing, which could result in anoxic conditions below the soil surface. Paul and Beauchamp (1989) compared nine different manures on the basis of their ability to supply carbon to denitrifying bacteria in a water logged silt loam soil. The highest rates of denitrification were from anaerobically decomposed liquid poultry and liquid swine manures and aerobically decomposed solid poultry manure. This was partially attributed to high volatile fatty acid (VFA) concentrations in the liquid manures compared to the solid manures, while the solid poultry manure contained fermentable carbon which resulted in accumulation of VFAs under anaerobic conditions. No direct measurements of denitrification over the Abbotsford Aquifer have been conducted. Zebarth and Paul (1995) in a N budget for the cropped land over the Abbotsford Aquifer, assumed that denitrification losses are low due to the well drained nature of the soils, and estimated the loss by denitrification to be equivalent to 5% of the total N applied from all sources. In addition, denitrification tends to be low in acidic soils with no significant loss of nitrate at pH values of less than 5. Most of the soils over the 18 aquifer tend to be acidic. Denitrification could however occur over the aquifer during intensive rainfall events that keep the soil saturated for extended periods of time. 2.2.2.4. Leaching Nitrate leaching is controlled primarily by the quantity of water moving out of the bottom of the soil root zone and the nitrate concentration in the soil through which the water moves. Since nitrate moves in the soil primarily by mass flow, one of the most significant factors affecting its leachability is the relationship between rainfall and potential evaporation (PE). The potential for nitrate leaching exists when rainfall exceeds PE by an amount greater than the water storage capacity of the soil. The nitrate concentration of soil water is dependent on a variety of factors including climate, cultural practices and soil characteristics. The water holding capacity of a soil, which is related to soil texture, has an inverse effect on the potential for nitrate leaching. Clay soils may delay leaching whereas a soil with a low water holding capacity has a greater probability of exceeding the field capacity resulting in a high hydraulic conductivity. When excess water accumulates above field capacity, hydraulic conductivity increases and leaching occurs. Elevated concentrations of groundwater nitrate are often associated with agricultural activities, particularly where manure N is applied at inappropriate times of year. This can occur in fall and winter when plant uptake is low and any application of N is in excess of plant requirements. In mild climates, if heavy rainfall occurs at the same time, leaching of nitrate will occur. In addition, rates of manure application are seldom calibrated and therefore it is easy to overapply. Also, manure application is often not considered in making inorganic N fertilizer recommendations. 19 The impact of precipitation on leachability of over-applied inorganic N fertilizers was demonstrated in a study by Schuman et al. (1975) in southwestern Iowa. Two rates of mineral N as anhydrous ammonia were applied to corn; one at the recommended rate of 168 kg N ha 1 and the second at 448 kg N ha'1. During the length of the three year study the accumulation of nitrate in the soil profile of the over-applied N moved from 1.0 to 3.1 m in depth. Three-quarters of the nitrate movement took place between June 1972 and April 1973 when 80 cm of precipitation resulted in 21 cm of water to be leached out of the root zone. Mean soil nitrate-N concentrations in the high rate of N plots at the water table depth were 3.7 and 12.9 mg N03"-N L"1 in April 1973 and April 1974, respectively, indicating it was reaching the saturated zone, compared to 2.0 and 4.5 mg N03"-N L'1 for same time period at the recommended rate of N. Spallacci (1981) compared N losses by leaching from different soils manured with four rates of application of pig slurry (0, 356, 711, and 1067 kg N ha1) on sand, sandy loam and sandy clay soils. The losses of N by leaching increased with increasing sand content. The average losses were mainly influenced by amounts and time of slurry applications and by the concentration of nitrate in the leachate. Vetter and Steffens (1981) also did a study on N leaching from pig slurry spread at various rates of application on sandy and loamy sand soils. At a rate of 540 kg N ha"1, approximately 225 kg N ha'1 yr'1 was leached from the sandy soil compared to 160 kg N ha 1 yr1 leached from the loamy sand. They attributed the difference to the higher storage capacity and lower percolation of water out of the root zone on the loamy sand. Sims (1987) found that soil inorganic N levels and N leaching were higher with ammonium nitrate than with poultry manure on a loamy sand in corn production in the early stage of growth. Nitrogen from the two sources was applied at rates designed to supply 0, 84, 168, and 252 kg N ha'1 based on predicted N availability for the poultry 20 manure. Soil inorganic N was measured at an early stage of growth just prior to the time when corn begins its most rapid uptake of N and at the mid-silking stage. At the early stage of growth, leaching potential appeared to be higher for the ammonium nitrate treatments compared to the poultry manure treatments. Leaching therefore may be especially a problem early in the season when uptake is low and broadcast application is used. Liebhart et al. (1979) found that poultry manure application exceeding that which can be utilized by a corn crop on a loamy sand in Delaware resulted in nitrate movement through the soil and into the groundwater. Nitrogen was applied as poultry manure at rates of 0, 325, 675, 1350 and 4475 kg N ha"1 over four years. Observation wells at 3, 4.5 and 6 m below the soil surface were monitored over time for inorganic N content. As the rate of poultry manure increased, the impact on the wells at the 3 m depth, just below the water table increased significantly. The groundwater nitrate-N concentration in the well at 3 m ranged from 65 to 174 mg L'1 at the high manure N rate, while it ranged from 1 to 15 mg L'1 at the control rate. Little impact was seen at the 4.5 and 6 m depths, which may be attributed to lateral movement of the nitrate or possibly denitrification. The county of Sussex in Delaware, has similar climate, intensity of poultry production, and coarse textured soils as the Abbotsford Aquifer. Robertson (1979) correlated land use in Sussex County to groundwater nitrate levels and found the lowest nitrate levels were associated with wooded areas while the highest nitrate levels were associated with barnyards, chicken or broiler raising operations, and in a few areas, high human population density. Over 20% of 800 wells sampled from a variety of land use areas had nitrate-N concentrations that exceeded 10 mg N L'1. Robertson also noted that the leaching of nitrate is especially prevalent in areas of high soil permeability and shallow water-table conditions. 21 A groundwater quality study in 1989 by Environment Canada (Liebscher 1992) over the Abbotsford Aquifer found that 46 of 73 (63%) of wells sampled had nitrate-N concentrations greater than 10 mg L'1. The mean nitrate-N concentration was 13 mg L'1 and ranged from 0 to 41.5 mg L"1. Samples were taken in areas of the aquifer believed to be the most severely impacted by nitrate contamination. Soil nitrate studies in south coastal British Columbia have shown that essentially all nitrate in the root zone is lost over the fall and winter, primarily by leaching (Kowalenko 1987; Zebarth et al. 1996). Zebarth and Paul (1996) showed some evidence of significant nitrate leaching on a sandy soil during a wet spring. A survey of soil nitrate concentrations with soil depth by Chipperfield (1992a) showed that 123 of 372 kg nitrate-N ha 1 was in the 60-90 cm depth interval at the end of the growing season, suggesting that growing season leaching may have occurred, possibly in response to irrigation. 22 3. MATERIALS AND METHODS 3.1. Experimental sites 3.1.1. Location and History Two experimental sites, referred to as the Columbia and Huntingdon sites, were selected over the Abbotsford Aquifer. The Columbia (lat 49°00'24"N, long 122°19'18"W) and Huntingdon (lat 49°00'54"N, long 122°20'48"W) sites are located approximately 75 km east of Vancouver in the Lower Fraser Valley (Figure 3.1). The sites were chosen from existing red raspberry (Rubus idaeus L.) fields where the raspberry stand was at least three years old, and where the surficial loess deposit was at least 50 cm deep. S C A N A D A u s' "A" J \ * ABBOTSFORD AQUIFER Figure 3.1. Location of experimental sites. 23 The soil at the experimental sites belong to the Marble Hill soil series, which is characterized by 50 cm or more of weakly structured stone-free silty eolian material overlying gravelly glacial outwash, and is classified as an Orthic Humo-Ferric Podzol (Luttmerding 1980; Luttmerding 1981). The sites are very gently sloping (1 to 5%) and well drained. Complete descriptions of the soil profiles at the Columbia and Huntingdon sites are presented in Appendix A. The stand of Meeker raspberries at the Columbia site was planted in 1979 (Brar, pers. comm.). Prior to that, the land was used as pasture for at least 10 years (Province of B.C. 1969; 1974; 1977). A 1930 air photo shows evidence of either active or recent logging at this site (Province of B.C. 1930), indicating that the site was broken near that time. The normal fertility program during raspberry production at the Columbia site consisted of a single application of inorganic fertilizer including 50 kg N ha 1, 55 kg P ha"1 and 80 kg K ha"1 applied mid-April or early May. Manure was last applied at this site in 1986 (Brar, pers. comm.). The field was rotovated to a depth of approximately 10 cm three to four times per year until 1988 and sub-soiling was carried out in 1988 and 1992 (Brar, pers. comm.). Small weak primocanes were pruned in the fall and winter along with the floricanes. The tops of the primocanes were woven at the top and tied. All other cultural practices including control of weeds, insects, and diseases, pruning and first shoot removal are carried out according to recommended practice (BCMAFF 1994). The stand of Skeena raspberries at the Huntingdon site was planted in 1985 (Sidhu, pers. comm.). Most of the trees at this site were cleared prior to 1930 (Province of B.C. 1930). The land may have been used as pasture in the early 1930's. The normal fertility program during raspberry production at the Huntingdon site consisted of a single application of inorganic fertilizer including 67 kg N ha"1, 94 kg P ha'1 and 94 kg K ha"1 applied mid April or early May (Sidhu, pers. comm.). Manure has been applied at the 24 site every year since 1976 (Sidhu pers. comm.). Normal practice is to roto-till twice per year (Sidhu pers. comm.). Small weak primocanes were pruned in the fall and winter along with the floricanes. The primocanes were pruned at the top and tied. All other cultural practices including control of weeds, insects, and diseases, pruning, and first shoot removal were carried out according to recommended practice (BCMAFF 1994). 3.1.2. Soil Characterization Soil characterization, including particle size distribution, soil pH, organic matter content and total N content, was done on soil samples collected prior to treatment application at the Columbia and Huntingdon sites in 1992 (Table 3.1). The samples were collected from the centre of the raspberry inter-row at the 0-15, 15-30, and 30-60 cm depth increments. The soil was air dried and ground to pass a 2 mm sieve. The pipette method (Gee and Bauder 1982) was used for particle size analysis with removal of organic matter. Soil pH was determined on a 1:1 (soikwater) suspension using a Corning Model 150 pH/ion analyzer (McLean 1982). The soil was sub-sampled and ground to pass a 100 mesh (250 ^m) sieve using a mortar and pestle for determination of organic matter and total N contents. Organic matter content was determined by weight loss during combustion in a muffle furnace for 12 hours at 450 °C of 5 g of previously oven dried soil (Goldin 1987). Total N of a 0.25 g sample was determined by a dry ash method on a LECO Model FP-428 Nitrogen Determinator. Soil bulk density was determined at four locations in the raspberry inter-row corresponding with soil sampling locations for ammonium and nitrate analysis. Inter-row 25 Table 3.1 Soil physical and chemical characteristics for the three soil depths at the experimental sites (Values are means (±1 SE) of 3 determinations). Depth (cm) Parameter 0-15 15-30 30-60 Columbia Particle size distribution sand (%) 21 (± 0.4) 23 (± 0.3) 26 (± 0.6) silt(%) 68 (+0.7) 69 (+0.4) 68 (+0.8) clay(%) 10 (±0.3) 7 (±0.6) 4 (±0.3) texture silt loam silt loam silt loam Soil pH 5.0 (± 0.06) 5.0 (± 0.10) 5.1 (± 0.12) Organic matter content (%) 10.1 (±0.34) 7.4 (±0.19) 4.9 (±0.25) Total N content (%) 0.32 (±0.014) 0.18 (±0.006) 0.09 (±0.004) Huntingdon Particle size distribution sand(%) 43 (±0.7) 45 (±0.7) 59 (±5.8) silt (%) 47 (± 1.2) 47 (± 0.8) 36 (± 4.9) clay(%) 9 (±0.7) 7 (±1.4) 3 (±1.1) texture loam sandy loam sandy loam SoilpH 6.0 (±0.29) 5.7 (±0.33) 5.8 (±0.25) Organic matter content (%) 8.5 (±0.30) 6.3 (±0.42) 3.0 (±0.12) Total N content (%) 0.25 (±0.012) 0.14 (±0.020) 0.03 (±0.006) locations were identified as the distance in cm from the centre of the raspberry row. The 20 cm location is within the mound, which is a raised area approximately 80 cm wide centred on the crop row, and approximately 15 to 20 cm high (Figure 3.2). The 40 cm location corresponds to the edge of the mound, while the 150 cm location corresponds to the location midway between raspberry rows and the 95 cm location is midway between the 40 and 150 cm locations. 26 Figure 3.2. Inter-row locations used for measuring soil parameters. Samples for bulk density determination were taken at the Columbia and Huntingdon sites in August 1993 in areas adjacent to the experimental plots. Soil bulk density was determined on nine 5.9 cm diam. by 13.9 cm long cores, three at each of the 40, 95, and 150 cm inter-row locations, at the Columbia site, and six cores, three at each of the 40 and 150 cm inter-row locations, at the Huntingdon site, from the middle of the 0-15, 15-30 and 30-60 cm depth increments (Table 3.2). In addition, bulk density was determined in the 0-15 cm layer only for the 20 cm inter-row location. The bulk density at the 20 cm inter-row location was significantly different (P<0.05) from the other locations at that depth and therefore are presented separately. The soil was dried to constant weight at 105 °C following removal of any large gravel. The mass and volume of the pebbles was subtracted from the core to obtain the bulk density of the soil matrix. 27 Table 3.2 Soil bulk density (g cm-3) for the three soil depths at the experimental sites (Values are means (+1SE)). Depth (cm) Site (Inter-row locations) n 0-15 15-30 30-60 Columbia (20 cm) 3 0.89 (+0.010) ND ND Columbia (40, 95 and 150 cm) 9 1.00 (+ 0.027) 1.06 (+0.031) 1.16(10.048) Huntingdon (40 and 150 cm) 6 1.05 (+ 0.023) 1.23 (+ 0.027) 1.32 (± 0.042) ND-not determined. Samples were taken at the Columbia site on 28 September 1992 and 28 September 1993 for determination of the population of nematodes. Three samples were collected using a 2.5 cm diameter Oakfield soil probe. Each sample was a composite of eight cores taken 30 cm from the raspberry row to a depth of 15 cm. Analysis was carried out at the Pacific Agriculture Research Centre (Vancouver). The mean nematode counts were 1475 (+346) and 958 (±57) per 50 mL of soil in 1992 and 1993, respectively. These numbers are extremely high and indicate the plants were under some stress and would be susceptible to winter damage (Vrain, pers. comm.). 3.2. Climatic Parameters Daily maximum and minimum soil and air temperatures were recorded at the Columbia site. The distance between the Columbia and Huntingdon sites is approximately 2 km therefore it was assumed that temperatures would be similar at the two sites. Soil temperature was measured at 5, 20, and 50 cm below the soil surface, while air temperature was recorded in a Stevenson Screen (1.2 m from the soil surface to the base of the screen) using thermisters. Data was stored on a 21X datalogger (Campbell Scientific) and downloaded to a cassette tape periodically and then 28 transferred to a computer file where it could be read. Rainfall data was obtained from the Abbotsford Airport (lat 49°02'N, long 122°22'W), located 4.2 km from the Columbia site and 2.4 km from the Huntingdon site (Figure 3.1). Comparisons were made between long term climate normals at the Abbotsford Airport (AES unpublished data) and 1992 and 1993 data. 3.3. Experiment 1: Poultry Manure Incorporation The objective of the experiment was to determine the influence of time of poultry manure incorporation on the recovery of poultry manure N as soil extractable inorganic N over a 30 day period. The experiment was conducted in 1992 and 1993 at the Columbia site at adjacent locations on the same field. 3.3.1. Experimental Design A randomized complete block design was used with three replications. The experimental unit was a 7 m long by 3 m (one raspberry inter-row) wide plot. Treatments included solid poultry layer manure applied at a rate of 400 kg total N ha"1 and incorporated within approximately 4 hours, 1 day, 1 week, and 1 month after application, and a control treatment, which received no manure. Incorporation was accomplished by roto-tilling to approximately 10 cm depth. The manure was broadcast by hand over the entire plot on 27 February 1992 and 9 March 1993. The total N content of the poultry layer manure was 2.7 and 1.4% in 1992 and 1993, respectively (Table 3.3). Although the total N content of the poultry layer manure differed between the two years, approximately 20% of the total N was in the 29 ammonium form in both years. The carbon to nitrogen ratio was 6.7 and 14.7 in 1992 and 1993, respectively. Table 3.3 Characteristics of the poultry layer manure used in 1992 and 1993 (Analysis performed by a commercial laboratory on an as received basis). Columbia Columbia Huntingdon Parameter 1992 1993 1992 Total nitrogen (g kg"1) 0.27 0.14 0.35 Ammonium-N (g kg1) 0.056 0.028 0.054 Dry matter (%) 47.8 44.6 57.3 Total Phosphorus (g kg"1) 0.025 ND 0.029 Total Potassium (g kg'1) 0.14 0.21 0.17 Total Carbon (%) 18.2 20.6 23.2 C:N Ratio 6.7 14.7 6.6 ND-not determined. 3.3.2. Soil Ammonium and Nitrate Concentrations Soil samples for determination of extractable ammonium- and nitrate-N concentrations were taken at the centre of the raspberry inter-row (i.e. the 150 cm location) at depth increments of 0-15, 15-30 and 30-60 cm. Soil samples were taken prior to manure application, and immediately after the 1 week and 1 month incorporation times (Table 3.4). One soil sample from each depth of each plot on each sampling date was collected using a 2.5 cm diameter Oakfield soil probe. For the 0-15 and 15-30 depth increments each sample was a composite of 10 cores, whereas 5 cores were sampled for the 30-60 cm depth increment. Soil samples were double bagged in plastic to prevent moisture loss, frozen within 12 hours, and kept frozen until analyzed. 30 Table 3.4 Manure incorporation and soil sampling dates, and day numbers for Experiment 1 in1992 and 1993. Date Event Day number 1992 February 27 February 28 March 5 March 30 1993 March 9 March 10 March 19 April 8 Soil sampled Applied manure Incorporated 4 h plots Incorporated 1 day plots Incorporated 1 week plots Soil sampled Incorporated 1 month plots Soil sampled Soil sampled Applied manure Incorporated 4 hr plots Incorporated 1 day plots Incorporated 1 week plots Soil sampled Soil sampled (too wet to incorporate) 0 1 7 32 1 10 30 Prior to analysis the soil samples were thawed and well mixed by hand. A 20 g sub-sample of field moist soil was extracted with 100 mL of 2M KCI, shaken for 1 hour and filtered with Whatman #40 filter paper into test tubes. Extracts were refrigerated for a maximum of two weeks or frozen until analyzed. Soil moisture content was determined gravimetrically at the time of extraction using approximately 40 g of soil (Gardner 1982). The extraction ratio was corrected for the moisture content of the soil. The ammonium- and nitrate-N concentrations of the soil extract were determined spectrophotometrically by flow injection analysis (Tecator Model 5020). The ammonium-31 N concentration was determined by a pH indicator after gaseous NH3 was produced and diffused through a teflon (PTFE) membrane. The nitrate-N concentration was determined by a diazo-based colour reaction with N02" after NOy reduction in a copperized cadmium column. Soil ammonium- and nitrate-N concentrations were converted to a kg ha'1 basis using the soil bulk density. After the soil ammonium- and nitrate-N concentrations for each depth and inter-row location were converted to kg ha 1, they were converted to a unit area basis by summing the three depth intervals. Apparent N recovery (ANR) of the applied manure was calculated as: ANR(%)= Treatment inorganic N (t1-t0) - Control inorganic N (t1-t0) x100 [3.1] Rate of total N applied where t1 = inorganic N content (kg ha1) at a specified time and tO = inorganic N content (kg ha"1) at the start of the experiment. ANR was calculated at one week after manure application for the 4h, 1 day, and 1 week treatments, and at one month for all four treatments. 3.3.3. Statistical Analysis Treatment means were compared using the general linear model of Systat (Version 5.02; Systat Inc, Evanston, IL; P<0.05). Contrast coefficients were used to determine if the treatment means followed a linear or quadratic trend over time (Table 3.5). The contrast coefficients for 1992 and 1993 were different because the time intervals between soil sampling dates were not the same. 32 Table 3.5 Contrast coefficients used to test for significant differences among treatments means at one week and one month for Experiment 1 in two years. Treatments Contrast 4 hours 1 day 1 week 1 month 1992-One Week Linear -2.555 -1.722 4.278 NC Quadratic 2.457 -2.799 0.341 NC 1992- One Month Linear -9.373 -8.543 -2.543 20.458 Quadratic 54.970 30.043 -109.174 24.161 1993- One Week Linear -3.555 -2.722 6.278 NC Quadratic 3.719 -4.064 0.3442 NC 1993-One Month Linear -10.123 -9.293 -0.293 19.708 Quadratic 60.098 35.624 -141.290 45.569 NC-not determined. 3.4. Experiment 2: Influence of poultry manure application pn soil inorganic N and raspberry N uptake The objectives of the second experiment were to determine the influence of rate of N application as poultry layer manure on N utilization in raspberries, and the spatial and temporal variation of soil moisture, and ammonium and nitrate concentrations. 3.4.1. Experimental Design The experiment was conducted at the Columbia site in 1992 (C92) and 1993 (C93) at adjacent locations on the same field, and at the Huntingdon site in 1992 (H92). 33 A randomized complete block design with four treatments and three replications was used in each case. The experimental unit was a 10 m long by 6 m (two raspberry inter-rows) wide plot. Thus each experimental unit consisted of one row of raspberries and two inter-rows bordered by guard rows. Treatments included 0, 100 or 200 kg total N ha"1 as poultry layer manure (C, 100M, and 200M, respectively) or 55 kg N ha'1 as ammonium nitrate (55F). The poultry layer manure and fertilizer treatments were chosen to represent approximately 0, 1 and 2 times the maximum acceptable rate of poultry manure application (BCMAFF 1992), and the recommended rate of inorganic N fertilizer application (BCMAFF 1994), for raspberry production. The manure and inorganic fertilizer were broadcast by hand over the entire inter-row areas and incorporated within four hours by roto-tilling. Manure and inorganic fertilizer application took place on 27 February 1992 and 10 March 1993 at the Columbia site, and 25 February 1992 at the Huntingdon site. Characteristics of the manures used at the Columbia site in both years are presented in Table 3.3. At the Huntingdon site, sampling of three of the experimental plots (one from each block) was discontinued part way through the study due to an error in fertilizer application. Fertilizer application rates, other than N, in the experimental plots were chosen to meet or exceed the nutrient requirements of the crop based on the soil test (Table 3.6). Samples for nutrient analysis were collected at the Columbia and Huntingdon sites on 27 January 1992, and at the Columbia site on 23 February 1993. One composite sample of 25 cores from the 0-15 cm depth was taken using a 2.5 cm diameter Oakfield soil probe at each site and analyzed by Griffin Laboratories (Kelowna, B.C.). 34 Table 3.6 Soil test results for available nutrients (mg L"1) for the Columbia site in 1992 and 1993 and the Huntingdon site in 1992. Nutrient Columbia 1992 Columbia 1993 Huntingdon 1992 Phosphorus 89 54 599 Potassium 145 130 269 Calcium 757 651 •3981 Magnesium 73 70 249 Sulphate 17 16 37 Boron 0.6 0.56 1.05 Fertilizers were applied late in April or early May. The Columbia site received 60 and 40 kg P ha"1 as 0-45-0, and 80 and 120 kg K and S ha"1 and 40 and 60 kg Mg ha"1 as 0-0-22-22-11 in 1992 and 1993, respectively. Phosphorus and potassium were banded approximately 40 cm from the raspberry row. Boron was applied at 1 kg B ha"1 as sodium borate as a spray application in both years. Fertilizer was not applied at the Huntingdon site. Cultural practices including pruning, and weed, insect and disease control were performed by the farm operator according to recommended practice (BCMAFF 1994). The soil at the Columbia site was rotovated twice in 1993, which was the first time it had been rotovated since 1988. Rotovation was carried out once at the Huntingdon site. Irrigation water was applied with traveling gun irrigation systems by the farm operators. At the Columbia site in 1992 the experimental area was irrigated six times (25 May, 5 June, 10 June, 22 June, 13 August, and 24 August). Eight water collection pans were placed perpendicular to the path of the irrigation gun on 22 June and the mean depth of irrigation water collected was 1 cm. The concentration of nitrate-N in the irrigation water at the Columbia site on 10 June 1992 was 13.2 mg L'1. Irrigation was applied in the experimental area at the Columbia site three times in 1993 (2 September, 4 October, 35 and 12 October). At the Huntingdon site irrigation was applied in the experimental site on 30 July 1992. 3.4.2. Soil Ammonium and Nitrate Concentrations Soil samples for determination of extractable ammonium- and nitrate-N concentrations were taken at 40, 95 and 150 cm from the raspberry row at the Columbia site in both years, and at 40 and 150 cm from the raspberry row at the Huntingdon site, at depth increments of 0-15, 15-30 and 30-60 cm. Sampling at different inter-row locations was done because it was anticipated that significant spatial variation in soil inorganic N concentrations would be present. For depths 0-15 and 15-30 each sample was a composite of 12 cores (six from each side of the raspberry row) whereas six cores were sampled for the 30-60 cm depth. . Three additional inter-row locations of 20, 68, and 123 cm were added for the control and 200 kg N ha 1 of manure treatments on four sampling dates at the Columbia site in 1993. This was done in order to better define the spatial distribution of available inorganic N across the inter-row at different times during the growing season. Sample analysis for determination of soil ammonium and nitrate was done as described in Section 3.3.2. It was assumed that the 15-30 and 30-60 depth intervals at the 20 cm location had similar bulk densities to the 40, 95 and 150 cm inter-row locations and, the 68 and 123 cm inter-row locations had similar bulk densities to the 40, 95, and 150 cm inter-row locations for all three depth intervals. Soil samples were taken before and after treatment application, during the growing season, and in October at the Columbia site in both years (Table 3.7). Soil sampling dates during the growing season corresponded with fruit development, harvest and raspberry cane growth stages. The control plots 36 were not sampled prior to treatment application in both years. The extra inter-row locations (20, 68 and 123 cm) were sampled in May, June, August and October 1993. Soil samples at the Huntingdon site were taken before and after the application of manure and inorganic N, and during the growing season (Table 3.7). Soil sampling dates at the Huntingdon site corresponded to fruit development and berry harvest. The control plots were not sampled prior to treatment application. It was not possible to obtain soil samples early in October at the Huntingdon site. 3.4.3. Calculation of mean ammonium and nitrate concentrations across the raspberry inter-row Six procedures were compared for converting soil inorganic N concentrations (mg N kg 1 soil) to a soil inorganic N contents (kg N ha-1). Soil inorganic N concentrations measured on four dates (3 May, 10 June, 12 August, and 18 October) at C93 from all six inter-row locations for the control and 200M treatments were used for the comparison. The procedures included use of two, three or six inter-row locations, each with or without uniform weighting factors assigned to the different locations (Table 3.8). The non-uniform weighting factors were chosen such that each sampling location would represent the area between itself and either the mid-points of the two adjacent inter-row locations, the plant row, or the mid-row. For example, the 95 cm inter-row location was assumed to represent the area between 81.5 and 109 cm, or a weighting factor of 0.19 of the inter-row, while the 20 cm inter-row location was assumed to represent the area between the plant row and 30 cm or a weighting factor of 0.20 of the inter-row. 37 Table 3.7 Soil sampling dates and plant growth stages at the Columbia site in 1992 and 1993 and the Huntingdon site in 1992. Sampling Date Plant Growth Stage Columbia 1992 February 25-26 March 3-5 April 28-29 June 9-10 July 6-7 August 13-14 October 5-6 Columbia 1993 March 2 March 15-16 May 3-4* June 10* June 28 August 10-12* September 17 October 18-19* Huntingdon 1992 February 24 March 3 June 8 August 18-19 Leaf buds (pre-manure application) Leaf buds (post-manure application) Flower buds First ripe berries Near end of berry harvest After berry harvest Leaves senescing No leaf buds (pre-manure application) No leaf buds (post-manure application) Flower buds Immature fruit First ripe berries End of berry harvest Primocane growth Leaves senescing Leaf buds (pre-manure application) Leaf buds (post-manure application) First ripe berries End of berry harvest *Extra inter-row locations (20, 68 and 123 cm) sampled. Calculated soil available N contents from the six procedures were compared using the general linear model of Systat (Version 5.02; Systat Inc., Evanston, IL; P<0.05) where the procedures were treated as treatments and each plot on each sampling date was treated as a replicate. It was assumed that the procedure using six 38 inter-row locations and non-uniform weighting factors should yield the most representative estimate of soil inorganic N content. Table 3.8 Weighting factors used to test for significant differences among six procedures for estimating mean soil inorganic N on a kg N ha"1 basis (The first digit refers to the number of sampling locations, U-uniform, and N-non-uniform). Weighting factors used to calculate mean soil inorganic N Sample location 6U 6N 3U 3N 2U 2N 20 0.166 0.20 0 0 0 0 40 0.166 0.16 0.33 0.45 0.50 0.63 68 0.166 0.18 0 0 0 0 95 0.166 0.19 0.33 0.37 0 0 123 0.166 0.18 0 0 0 0 150 0.166 0.09 0.33 0.18 0.50 0.37 3.4.4. PLANT PARAMETERS Sampling of floricanes and primocanes was performed to determine the effect of the applied treatments on crop yield, vigour, and N uptake. 3.4.4.1. Floricanes The number of floricanes per plot was recorded in April. The mean number of floricanes per plot was 94, 85, and 71 for C92, C93 and H92 respectively, and was independent of the applied treatments Ten representative floricanes were removed at ground level from each plot at the time of first berry ripening. The berry ripening growth stage was chosen for dry matter and nitrogen uptake because floricane growth, which is most rapid in late spring, starts 39 to decline as fruit ripening progresses even though growing conditions may continue to be favourable (Kowalenko 1994a). The fruiting clusters, which are the combination of buds, flowers and fruit joined by small stems, were separated from the remainder of the sampled material, which included the cane laterals and leaves and is hereafter referred to as the stem. The number of flower buds, flowers and set fruit per cane were determined. This sampling took place 3-5 June 1992 and 24-25 June 1993 at the Columbia site and 1-2 June 1992 at the Huntingdon site. The stems and fruiting clusters were dried to constant weight at 60 °C and weighed. Individual floricanes were combined to form one composite sample for each crop component from each plot, and ground in a Wiley mill to pass a 2 mm screen. The total N content of a 0.2 g sub-sample of the ground material was determined by a dry ash method on a LECO Model FP-428 Nitrogen Determinator. Total cane and fruiting cluster dry matter yield on a kg ha"1 basis was calculated on a per plot basis. In addition, crop N uptake per plot on a kg ha"1 basis was calculated. Berry yield was estimated using the method of Daubney et al. (1986) as: Yield Estimate = (# set fruit/cane)(# canes/plot)(mean berry fresh wt.) [3.2] Plot area where the yield estimate, mean berry fresh weight and plot area have units of t ha 1, tonnes, and hectares, respectively. It was assumed that any set fruit at the time of first berry ripening would mature to ripe fruit whereas the buds and flowers at that time would not mature before the end of harvest. The fresh weight of an individual berry per plot was determined by averaging the weight of 50 ripe berries, measured six times during the harvest period in 1992 at both sites and five times during the harvest period in 1993 at the Columbia site, and dividing by 50. 40 3.4.4.2. Primocanes Primocane harvest took place on 6-7 October 1992 and 7-8 October 1993 at the Columbia site when the leaves began to senesce. Primocane harvest could not be done at the Huntingdon site. The time of sampling was chosen to measure maximum primocane dry matter production and N uptake (Kowalenko 1994a). All primocanes within three 0.5 m portions of row were removed at ground level in each plot. The primocanes were sorted into greater than 130 cm length which might be chosen as floricanes for the following year, and "suckers" less than 130 cm long which would automatically be pruned out. The samples were dried to constant weight at 60 °C and weighed. All dried samples from each plot including primocanes and suckers were ground in a Wiley mill with a 2 mm screen. Total N of a 0.2 g sub-sample was determined by a dry ash method on a LECO Model FP-428 Nitrogen Determinator. Total primocane dry matter yield and N uptake per plot on a kg ha"1 basis was calculated. In addition, mean cane weight of the greater than 130 cm length primocanes was calculated. It was assumed that mean cane weight would provide an index of primocane vigour by assuming that the higher the mean cane weight, the higher the yield potential in the following growing season. Primocane length and diameter have been used previously (Zebarth et al. 1993), however mean cane weight was used in the present study because it was thought to be a more sensitive index than length or diameter. The apparent N recovery (ANR) of applied N by the entire plant and by the fruiting cluster was calculated for C92 and C93 as: ANR(%)= Nuptake treatment - Nuptake control x100 [3.3] total N applied as manure or fertilizer 41 where N uptake is the sum of the N content of the primocanes as determined in late fall and 0.9 times the N content of the floricanes as determined at harvest measured in kg ha 1 and the unit of N application is kg ha'1. It was assumed that 10% of the N content in the primocane was taken up in the previous growing season (Kowalenko 1994a). 3.4.5. STATISTICAL ANALYSIS Means were compared using the general linear model of Systat (Version 5.02; Systat Inc. Evanston, IL; P<0.05). Contrast coefficients were used to determine if a linear or quadratic relationship existed among the three manure treatments (Table 3.9). The third contrast was used to determine if there was a significant difference between the ammonium nitrate and low rate of manure treatment. Table 3.9 Contrast coefficients used to test for significant differences among treatment means for the Columbia and Huntingdon sites for Experiment 2 (C-control, 55F-55 kg N ha'1 as ammonium nitrate, 100M-100 kg N ha"1 as manure, 200M-200 kg N ha"1 as manure). Contrast Treatments C 100M 200M 55F Linear -1 0 +1 0 Quadratic +1 -2 +1 0 55F vs 100M 0 +1 0 -1 42 3.5. Experiment 3: Influence of an Inter-row Cover Crop on Primocane Vigour and Soil Inorganic N Concentrations The objective of the third experiment was to determine the influence of rate of nitrogen application and a fall seeded cover crop (CC) on primocane vigour and soil inorganic N. The experiment was conducted at the Columbia site in 1992 and 1993 at adjacent locations on the same field. 3.5.1. Experimental design A randomized complete block design with four treatments and three replications was used in each of two years. The experimental unit was a 10 m long and 6 m (two raspberry inter-rows) wide plot. Thus each experimental unit consisted of one row of raspberries and two inter-rows. The four treatments included two rates of poultry layer manure (0, and 200 kg total N ha1), with and without a spring oats (Avena sativa L. 'Saia') cover crop (CC). The treatments with no cover crop were the same as those in Experiment 2. Following roto-tilling, the cover crop was seeded on 14 August 1992 and 13 August 1993 at a rate of 100 kg ha"1 using a cone seeder with a row width of 15 cm. In 1992 the width of the seeded area was 150 cm, centred in the raspberry inter-row, and extending to approximately 75 cm from the raspberry row. In 1993, the width of the seeded area was 220 cm, and extended to approximately 40 cm from the raspberry row. The 1993 seeding was designed to more effectively study the influence of the cover crop on primocane vigour. Normal practice would be to seed the cover crop to within 40 cm of the crop row over the entire inter-row area, which would cover approximately 75% of the field. 43 3.5.2. Cover crop dry matter yield and N uptake The above-ground portion of the cover crop was removed at ground level with shears 21 to 22 October 1992 and 15 October 1993. Plant growth stage and height was determined on three representative plants for each plot. Growth stage was classified using Zadoks' decimal code system (Zadoks et al. 1974; Tottman et al. 1979), a brief explanation of which is given in Appendix C. A harvested area of at least 3.75 m2 and 3.6 m2 in size was used in 1992 and 1993, respectively. The fresh weight of the entire harvest and an approximately 650 g sub-sample was determined. The sub-sample was oven dried to constant weight at 60 °C and weighed to calculate dry matter yield. The dried samples were ground in a Wiley mill to pass a 2 mm screen. Total N concentration was determined on a 0.2 g sub-sample by a dry ash method on a LECO Model FP-428 Nitrogen Determinator. Total cover crop dry matter yield and N uptake per plot on a kg ha"1 basis were calculated. Dry matter yield and N uptake were multiplied by 0.75 because it was assumed that under normal farming practices 75% of the inter-row area would be seeded with a cover crop. 3.5.3. Soil inorganic N and primocane dry matter yield and N uptake Soil samples were collected as described previously (Section 3.4.2). Sampling locations were 40, 95, and 150 cm from the centre of the raspberry row. Soil samples were taken on the last two soil sampling dates in 1992 and the last three soil sampling dates in 1993 used for Experiment 2 (Table 3.7). Procedures for sampling the primocanes were as described in Section 3.4.4.2. 44 3.5.4. Statistical analysis Means were compared using the general linear model of Systat (Version 5.02; Systat Inc, Evanston, IL; P<0.05). Contrasts were used to test for significant differences among treatments (Table 3.10). Table 3.10 Contrast coefficients used to test for significant differences among treatment means for Experiment 3 (C-control, 55F-55 kg N ha 1 as ammonium nitrate, 100M-100 kg N ha 1 as manure, 200M-200 kg N ha'1 as manure). Contrast Treatments C 200M C+CC 200M+CC Manure -1 1 -1 1 Cover crop -1 -1 1 1 Manure x Cover crop 1 -1 -1 1 3.6. Experiment 4: Comparison of Date of Seeding and Cover Crop Species The objective of the fourth experiment was to determine the influence of cover crop species and date of seeding on N uptake by inter-row cover crops in a raspberry field. The experiment was conducted in 1992 and 1993 at the Columbia site on adjacent locations in the same field. 3.6.1. Experimental Design A randomized complete block design with seven treatments and three replications was used. Treatments included Saia spring oats seeded on five dates (Table 3.11), and spring barley (Hordeum vulgare L. 'Winchester') and spring wheat (Tritucum aestivum L. 'Max') seeded on the second date only. Winchester barley was 45 replaced with Verden barley in 1993 because the former was not available. The experimental unit was a 10 m long by 3 m (one raspberry inter-row) wide plot. The first seeding date represents a seeding date immediately after crop harvest in an early year such as 1992. The second date represents a more realistic seeding date given that harvest typically extends into early August. The last three dates were included for comparison. The experimental plots were fertilized as described in Section 3.4.1 with an additional 55 kg N ha 1 as ammonium nitrate in both years. Seeding was accomplished with a cone seeder over the entire plot length in the centre of the inter-row following roto-tilling. The width of the seeded area was approximately 150 cm or about 50% of the inter-row in both years. The oats, barley and wheat were seeded at a rate of 100 kg ha"1 in 1992. In 1993 the oats were seeded at a rate of 100 kg ha"1, however the barley and wheat were seeded at 216 kg ha"1 and 275 kg ha"1 respectively to give the same number of seeds per unit area for all three cover crop species. 3.6.2. Cover Crop Dry Matter Yield and N Uptake The above-ground portion of the cover crop was removed at ground level with shears on 20 to 21 October 1992 and 15 October 1993. Plant growth stage and height was determined on three representative plants for each treatment. Growth stage was classified using Zadoks' decimal code system (Zadoks et al. 1974; Tottman et al. 1979), a brief explanation of which is given in Appendix C. In 1992 a representative area of at least 1.2 m2 was harvested with most harvest areas 3 m2 or greater. In 1993 harvest areas were 3 m2 with the exception of one plot with an area of 1.8 m2. 46 Table 3.11 Cover crop seeding dates and day number for 1992 and 1993 for Experiment 4. Seeding date Day number 1992 1. August 5 0 2. August 14 9 3. August 25 20 4. September 4 30 5. September 15 41 1993 I.July 30 0 2. August 13 14 3. August 26 27 4. September 10 42 5. September 24 56 In 1992, and for the last seeding date in 1993, the entire sample of plant material for each plot was dried to constant weight at 60 °C and weighed. For the remaining sampling dates in 1993, the fresh weight of entire harvest, and an approximately 500 g sub-sample was determined. The sub-sample was oven dried to constant weight at 60 °C and weighed to calculate dry matter yield. The dried samples were ground in a Wiley mill with a 2 mm screen. Total N concentration was determined on a 0.2 g sub-sample by a dry ash method on a LECO Model FP-428 Nitrogen Determinator. Total cover crop dry matter yield and N uptake per plot on a kg ha 1 basis were calculated. Dry matter yield and N uptake were multiplied by 0.75 because it was assumed that under normal farming practices 75% of the inter-row area would be seeded with a cover crop. 47 3.6.3. Statistical Analysis Treatment means were compared using the general linear model of Systat (Version 5.02; Systat Inc, Evanston, IL; P<0.05). Contrast coefficients were used to test for significant differences among treatment means in both years (Table 3.12). Table 3.12 Contrast coefficients used to test for significant differences among treatment means in Experiment 4 in 1992 and 1993. Treatments Day 1 Day 2 Day 3 Day 4 Day 5 Contrast Oats Oats Oats Oats Oats Barley Wheat 1992 Oats linear -20 -11 0 +10 +21 0 0 Oats quadratic +205.1 -81.8 -212.4 -121.2 +210.2 0 0 Barley vs wheat 0 0 0 0 0 +1 -1 Oats vs barley and 0 +2 0 0 0 -1 -1 wheat 1993 Oats linear -27.8 -13.8 -0.8 +14.2 +28.2 0 0 Oats quadratic +397.4 -193.4 -391.0 -199.0 +386.2 0 0 Barley vs wheat 0 0 0 0 0 +1 -1 Oats vs barley and 0 +2 0 0 0 -1 -1 wheat 48 4. RESULTS AND DISCUSSION 4.1. Climatic Conditions 4.1.1. Precipitation and Irrigation Total precipitation at the Abbotsford Airport during the study period was near normal. Total annual precipitation was 1422 and 1260 mm in 1992 and 1993, respectively, compared to the 30-year (1961-1990) mean of 1596 mm (Figure 4.1). Climate normals for the Abbotsford Airport for a 30-year period (1951-1980) are shown in Appendix B for air temperature, precipitation, wind speed and direction, and sunshine hours. Growing season (March to October) precipitation was 99 mm below the 30-year mean of 713 mm in 1992, while precipitation during the same period in 1993 was 71 mm above the 30-year mean. In 1993 precipitation during March, April and May was 172 mm above the 30-year long term mean of 320 mm, increasing the potential for nitrate leaching to occur that spring. 4.1.2. Air and Soil Temperature The difference in mean monthly air temperature between the Columbia site and the Abbotsford Airport averaged 0.3 °C, and was never greater than 1.0 °C. Therefore, the air temperatures measured at the two sites were considered to be the same for the purposes of this thesis. The mean growing season (March to October) air temperature 49 at the Columbia site was 14.5 and 14.0 °C for 1992 and 1993, respectively, compared to the 30-year (1951-1980) mean of 12.4 °C (Table 4.1). 3501 E 300 £ c 250 o .•t- 200 Q. 0 2> 150 o. 1 100 F 50 • 1992 • 1993 • Mean M M J J Month 0 N Figure 4.1. Monthly total precipitation for 1992 and 1993 at the Abbotsford Airport compared to a 30-year (1961-1990) mean (AES unpublished data). The early spring of 1992, was very mild (Table 4.1), which may have been responsible for the early leaf budding and the earlier than normal berry harvest observed in 1992. The higher than normal temperatures continued into the summer months and the mean air temperature for June, July and August was 18 °C, 1.8 °C above the 30-year mean. The winter of 1992/93 was more severe than normal with eleven consecutive days from late December to early January when the temperature did not rise above freezing. The cooler temperatures may have caused the damage to the floricane tips 50 observed in the spring of 1993. The mean temperature for June, July and August was 16.6 °C, 0.4 °C above the 30-year mean. On average, mean monthly soil temperature did not differ from the mean monthly air temperature by more than 1.2, 1.3, and 1.7 °C for the 5, 20 and 50 cm depths, respectively, for 1992 and 1993 (Table 4.1). Soil temperature decreased with depth during spring and summer, and increased with depth during fall and winter. Mean growing season (March to October) soil temperatures were 14.7, 14.5 and 14.1 °C for the 5, 20 and 50 cm depths, respectively, for 1992, as compared to 13.5, 13.3 and 12.8 °C for the 5, 20 and 50 cm depths, respectively, in 1993 (Table 4.1). 4.2. Experiment 1: Poultry Manure Incorporation 4.2.1. Changes in soil inorganic N Soil inorganic N to 60 cm depth in the control treatment increased from 24 to 67 kg N ha"1 over the duration of the experiment in 1992. In contrast, soil inorganic N to 60 cm depth decreased from 32 to 26 kg N ha'1 during a similar time period in 1993. There was a net increase of inorganic N, averaged across treatments which received manure, between day 7 (174 kg N ha'1) and day 32 (250 kg N ha'1) in 1992, while it changed little between day 10 (197 kg N ha1) and day 30 (193 kg N ha'1) in 1993. For treatments receiving manure, the proportion of soil inorganic N in nitrate form increased between the one week and one month sampling dates in both years (Figure 4.2). On average, 27% of the inorganic N was in nitrate form 7 days following manure application in 1992, whereas 72% was in nitrate form 32 days following application. In 1993, 52% of the inorganic N was in the nitrate form 10 days following manure application compared to 83% 30 days following application. 51 CO CO o o •*—• 03 k _ CO O -E o CM CO CO O O o 3 to CD Q. E 03 CD I O CO E o o to E o o CM E o LO E o o LO E o o CM E o LO CO CO CO CM CO CO c 03 E r-- CM CM CO fc o Q Q Q Q Q Q co CM Q Q co CD C C | P f-m fc LO CO CM LO LO CO CO CM LO •xt d CO CO LO 03 c 03 —5 03 CD LL CO CO CO CO LO CO CO N CO (O CO (O O) r ; O0 CO CO> CD Is-' -xt 00 CO 00 O LO O - T - CD CT) O0 CD 00 O O co in cn in co CD CD Is- ^ CM 00 Is- T - ; -r-C0 LO Is- Is-CD LO CM CM LO Is-00 CO 00 fc i - fc CD CM O O LO LO CD 00 LO LO LO LO CO 00 T-m CD 00 (ji T - ^" CD CO S ^ co cn LO CM CM -st CO d T - : -st fc- O Is- o 00 CM "xt fc CO CO LO i -d Q. < CO CD x — i —) CO < CD n E CD +-» Q. CD CO Q Q o O o O Q Q Z Z Q Q Z Z 00 LO T - ; co 0) iri O O CM 00 00 CO CM CM 00 CO •xf CM CO T - 1 Q Q LO O CO CM O -xT fc od od -xt CM cb CD .a E CD > o CO CD CM ib co CD - O E CD o CD Q 52 A 0 0-15 15-30 30-45 45-60 1992 50 100 kg ha"1 B 150 0 45-60 1993 50 100 150 I-. I Nitrate-N I I Ammonium-N C 0 0-15 15-30 30-45 45-60 E 0 0-15 15-30 30-45 45-60 50 100 150 50 100 150 F 0 0-15 15-30 30-45 45-60 50 100 150 50 100 150 re 4.2. Soil ammonium- and nitrate-N distribution with depth, averaged across treatments which received manure: before manure application in A) 1992 and B) 1993; approximately one week after manure application in C) 1992 and D) 1993; and approximately one month after manure application in E) 1992 and F) 1993. / 53 4.2.2. Distribution of inorganic N with soil depth The distribution of soil inorganic N with depth differed between 1992 and 1993 at the one month sampling date for the treatments which received manure (Figure 4.2). Prior to manure application, there was less than 20 kg N ha"1 soil inorganic N in the 0-15 cm depth interval and soil inorganic N decreased with soil depth in both 1992 and 1993. By one week following manure application, soil inorganic N in the 0-15 cm depth increment increased to 115 and 146 kg N ha"1 in 1992 and 1993, respectively. Increases in soil inorganic N of up to 54 kg N ha"1 were also observed in the 15-30 and 30-60 cm depth increments. In 1992, there was a net increase of soil inorganic N to 60 cm from 201 to 250 kg N ha 1 between days 7 and 32, while there was a decrease from 203 to 193 kg N ha 1 in 1993. In 1992, 50, 35 and 15% of the inorganic N was in the 0-15, 15-30 and 30-60 depth intervals, respectively at Day 32. In comparison, soil inorganic N was more evenly distributed through the soil profile in 1993 with 37, 34 and 29% of the inorganic N in the same depth intervals, respectively, at Day 30. 4.2.3. Apparent N recovery Apparent recovery of the 400 kg total poultry manure N as soil inorganic N to 60 cm was corrected for any increase in soil inorganic N in the control treatment which had no manure applied, and for inorganic N present in the soil at the beginning of the experiment. By day 7 in 1992, the apparent N recovery in the 4 hr to incorporation treatment was 42%, almost twice as high as the mean apparent N recovery of 22% in the 1 and 7 d treatments (Figure 4.3). By day 32, however, the apparent N recovery in the 4 hr, 1 and 7 d treatments were similar and averaged 52%, almost twice as much as 54 601 1992 rs- 50 40 30 20 10-4 hours 1 day 7 days Time after application on which calculation was based: • At 7 days • At 32 days 32 days 60 ^ 50 cu > o o CD c cu 1_ CO Q . Q . < 401 30 20 10 1993 4 hours 1 day 10 days Time to incorporation Time after application on which calculation was based: • At 10 days • At 30 days 30 days Figure 4.3. Apparent manure N recovery as soil inorganic N to 60 cm depth following application of 400 kg total N ha'1 as poultry layer manure; 7 and 32 days following manure application in 1992, and 10 and 30 days following manure application in 1993. 55 the 28% for the 32 day treatment. Apparent N recovery decreased curvilinearly at 7 days, and linearly at 32 days, with increased time to incorporation following manure application in 1992 as indicated by contrast analysis (Table 4.2). Table 4.2 Statistical significance of apparent manure N recovery (%) from the application of layer manure as influenced by time to manure incorporation in two years, and calculated approximately one week and one month following manure application (Values are mean squares). Source df 1992" 1993 One week after application Block Treatment Linear Quadratic Error One month after application Block Treatment Linear Quadratic Error *P<0.05. In 1993, apparent N recovery by day 10 was 53% in the 4 hr treatment compared to an average of 33% in the 1 and 10 d treatments (Figure 4.3). By day 30, the apparent N recovery in the 4 hr treatment was 49% compared to 38% averaged across the 1, 10 and 30 d treatments. Although the treatment main effect was not significant, apparent N recovery decreased curvilinearly with increased time to manure 2 146.6 240.4 2 421.3* 377.6 1 273.4* 75.5 1 569.2* 679.7* 4 28.0 62.4 2 659.9* 151.3 3 432.0 93.4 1 1137.7* 31.3 1 143.3 120.3 6 108.9 73.1 56 incorporation at 10 days following manure application in 1993 as indicated by contrast analysis (Table 4.2). There was no significant effect of time to incorporation on apparent N recovery 30 days following manure application. 4.2.4. Discussion The increase in soil inorganic N in the control treatment in 1992 indicates that significant mineralization can occur early in the growing season at this site. The apparent net mineralization rate of 1.34 kg N ha"1 day1 averaged over the duration of the experiment in the control treatment in 1992 was high compared to an average of 1.06 kg ha"1 day'1 in a broccoli study by Zebarth et al. (1995) in Agassiz, BC. Their estimate took into account plant N uptake, and atmospheric and irrigation sources of N and was averaged over the growing season. Apparent net mineralization calculated in a similar manner for broccoli was 1.4 kg N ha*1 day1 for another Fraser Valley soil, averaged over the growing season (Kowalenko and Hall 1987). The high rate of mineralization in this study was probably due in part to the warm soil temperatures and near field capacity moisture conditions, conditions known to enhance the mineralization process (Cassman and Munns 1980; Justice and Smith 1962). Mineralization increases with increasing soil temperature up to a maximum temperature of 40 °C (Stanford et al. 1973; Reddy 1982), while mineralization is believed to be low for soil temperatures below 4 °C where there is no long term history of manure application as was the case in this study (Maidl and Fischbeck 1989). Lower mineralization would be expected in 1993 compared to 1992 due to the cooler soil temperatures. However lower net mineralization would not account for the decrease of 8 kg N ha"1 during the experimental period in the control treatment in 1993. 57 This loss likely resulted from leaching of nitrate below the 60 cm depth due to high rainfall during that time period. Total rainfall over the experimental period was 30.8 mm in 1992, compared to 233.5 mm in 1993; 82% of which fell between days 10 and 30. Some leaching loss may be expected where manure is applied early in the spring on shallow, medium coarse textured soils such as those over the Abbotsford Aquifer. Mineralization of the organic N in the poultry layer manure, which made up 80% of the total N applied, proceeded rapidly in both 1992 and 1993 as indicated by the increase in soil inorganic N in the treatment plots by one week following manure application. The proportion of organic N of the layer manure used in this study was similar to other poultry manures studied: 88% in one poultry layer manure (Bitzer and Sims 1988), 72% in three poultry manures (Sims 1987), 78% in three broiler litters (Westerman et al. 1988), and 75% in eight poultry layer manures (Chipperfield 1992b). Several studies have shown that the organic N in poultry manure mineralizes more rapidly than other manures and sewage sludge (Pratt et al. 1973; Castellanos and Pratt 1981; Chae and Tabatabai 1986; Chescheir et al. 1986; Sims 1986; Bitzer and Sims 1988). In a study by Bitzer and Sims (1988) large amounts of inorganic N were found in the soil within two weeks of poultry manure application. Early spring nitrification of soil and manure N may lead to an increased risk of nitrate leaching in wet springs. Such leaching appears to have occurred in 1993. The decrease in soil inorganic N in the control treatment in 1993, and the downward re-distribution of inorganic N with depth on day 30 in 1993 suggest that leaching did occur. There was a net decrease in soil inorganic N content to 60 cm depth between days 10 and 30. This represents a minimum estimate of the loss of soil inorganic N during this period because the net mineralization of soil and manure N is unknown. Given the rapid mineralization measured in 1992, and the rapid nitrification measured in 1993, 58 suggesting significant microbial activity, significant mineralization can be expected to have occurred during this period. Leaching may account for the loss, although denitrification may also account for some of the nitrate loss during short intervals of time when the soil would have been saturated, particularly where manure was applied. However the relatively cool soil temperatures in 1993 would have minimized denitrification. Plant N uptake during this time period is assumed to be minimal (Kowalenko 1994a). Bitzer and Sims (1988) found early spring leaching with poultry manure application on a coarse textured soil in Delaware. Mean apparent N recovery one month after manure application, averaged across treatments, was 46 and 41 % for 1992 and 1993 respectively. The lower N recovery in 1993 may be partly a result of nitrate leaching. The calculated recovery rates found are similar to those found in other studies. In an incubation experiment, Castellanos and Pratt (1981) found that approximately 48% of the organic N in poultry manure was mineralized in 10 weeks at 23 °C. Within one week, Hadas et al. (1983) found that 36% of the total organic N in ground poultry manure mixed with soil was mineralized in an incubation experiment at 14 °C. In an aerobic incubation experiment using broiler litter at 25 °C, Westerman et al. (1988) found an apparent N recovery of 52% after four weeks. The trend of decreasing apparent N recovery with increasing time to incorporation suggests that greater ammonia volatilization may have occurred in those treatments where manure incorporation was delayed. Lockyer and Pain (1989) found that an average of 29% of the ammonium-N applied in poultry manure was volatilized within 6 days following application and the pattern of loss over time showed that the rate of volatilization was most rapid during the first 6 hr. Rapid ammonia volatilization in the first 4 hours following manure application, followed by mineralization after manure incorporation, may explain why there was little difference between the 4 hour, 1 day and 59 7 day incorporation times on N recovery calculated at 32 days after manure application, in 1992. Some aspects of the results may be related to the climatic conditions at the time of manure application and shortly after manure application. Manure application was completed by 12 noon and it was warm and sunny on both days of application with a daytime high air temperatures of 19 °C and 16 °C in 1992 and 1993, respectively. This situation may be typical of most years because a warm sunny day would generally be chosen for manure application. Under these conditions, rapid volatilization of manure ammonia would be expected to occur. This is consistent with the similar N recovery obtained in both years of the study. The difference in results when apparent net mineralization was calculated seven or 30 days following manure application in 1992 may have been a result of surface drying of the manure. There was little precipitation during the experimental period in 1992 so mineralization may have been delayed until the manure was incorporated into the top 10 cm of soil thus providing more ideal moisture conditions. In contrast, wetter conditions in 1993 would have allowed mineralization of manure N to occur even without incorporation. This is consistent with the similar estimates of manure N recovery calculations 10 or 30 days after manure application in 1993. 60 4.3. Experiment 2: Influence of poultry manure application on soil inorganic N and crop N uptake 4.3.1. Soil moisture content at six inter-row locations over time in 1993 Soil moisture content between raspberry rows varied in both space and time. Significant main effects were observed for inter-row location, soil depth, and time, but not for treatment (Table 4.3). Only the location x time interaction was significant. On 3 May the mean soil moisture content, averaged across depth and inter-row locations, was 0.41 g g 1 . Soil moisture content was highest at shallow depth and close to the crop row (Figure 4.4). There was a small net increase in the mean soil moisture content from 0.41 to 0.42 g g"1 between 3 May and 10 June. A total of 189 mm of rain fell in the same time period while the potential evaporation was 133 mm. The increase in soil moisture content was uniform over most sampling locations and depths, resulting in a moisture distribution similar to that observed on 3 May, except that the moist area near the crop row extended deeper into the soil (Figure 4.4). Between 10 June and 12 August, 107 mm of rain fell and the potential evaporation was 224 mm. There was a net decrease in mean moisture content from 0.42 to 0.35 g g'1. The decrease in moisture content occurred uniformly over most depths and locations, resulting in a similar pattern to that observed on 3 May and 10 June (Figure 4.4). There was a small net increase in the mean moisture content from 0.35 to 0.36 g g~1 between 12 August and 18 October. During this period there was 55 mm of precipitation, an unknown amount of irrigation and the potential evaporation was 183 mm. Although there was an overall increase in soil moisture, small decreases in 61 Table 4.3 Statistical significance of soil moisture content (g g'1), ammonium-N concentration (mg L"1), and nitrate-N concentration (mg L~1) as influenced by rate of N application as poultry layer manure, inter-row location, depth and time (Values are mean squares). Moisture Ammonium-N Nitrate-N Source df content concentration concentration (x103) Block 2 2.41 18.57 49.2 Treatment (Tr) 1 0.09 2.96 4350.6 Error 2 1.89 6.00 397.4 Location (L) 5 8.02* 29.78* 669.7* Tr xL 5 0.58 4.74 75.0 Error 20 2.94 6.31 79.0 Depth (D) 2 141.17* 201.42* 1298.1* Tr xD 2 0.51 7.23 117.8 LxD 10 0.68 3.63 68.0 Tr x L x D 10 0.96 0.75 4.4 Error 48 4.32 8.58 156.9 Time (T) 3 146.28* 114.30* 2317.3* Tr x T 3 1.38 4.38 112.0* L x T 15 3.37* 28.49* 130.0* DxT 6 1.10 21.62* 831.3* Tr x L x T 15 0.50 2.88 15.8 Tr x D x T 6 1.46 1.79 34.2* L x D x T 30 0.75 1.12 36.6* Tr x L x D x T 30 0.87 1.47 7.4 Error 216 0.91 2.02 12.9 *P<0.05. 62 Inter-row location (cm from raspberry row) E o C L CO T3 O 00 3 May 0 20 40 60 80 100 120 140 160 T • 1 • T • r 0 -j E 15-Q. CD 30-T3 O 45-CO 60-10 June 0 20 40 60 80 100 120 140 160 ^7 i i •— i—.—.—J- 7 E o Q . ~U O CO 12 August 0 20 40 60 80 100 120 140 160 E o C L 0 •o O CO 18 October 0 20 40 60 80 100 120 140 160 Figure 4.4. Soil moisture content (g g1) distribution with depth and inter-row location averaged for the control and 200 kg N ha"1 as poultry layer manure treatments for four sampling dates in 1993 at the Columbia site (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations). 63 moisture content were observed at the 20, 68 and 123 cm inter-row locations (Figure 4.4) . Moisture content decreased with soil depth as on the three previous dates. There was 28 cm of water in the top 60 cm of soil at the 20 cm inter-row location on 10 June as compared to an average of 25 cm for the other inter-row locations (Figure 4.5) . Soil moisture was more evenly distributed across the inter-row on the May, August and October sampling dates. 0 20 40 60 80 100 120 140 160 Inter-row location (cm from raspberry row) Figure 4.5. Water depth equivalent to 60 cm depth, averaged for the control and 200 kg N ha"1 as poultry layer manure treatments, for each inter-row location on four sampling dates at the Columbia site in 1993. 4.3.2. Ammonium-N concentration at six inter-row locations over time in 1993 Soil ammonium-N concentration also varied in space and time. Significant main effects were observed for inter-row location, soil depth, and time, but not for treatment (Table 4.3). There were significant interactions for location x time and depth x time. 64 On 3 May the mean ammonium-N concentration, averaged across depth and inter-row location, was 4.6 mg L"1. Ammonium-N concentrations decreased with increasing soil depth and were relatively uniform across inter-row location (Figure 4.6). Between 3 May and 10 June there was a net 47% decrease in the mean ammonium-N concentration to 2.7 mg L"1. Ammonium-N concentrations decreased at all depths and inter-row locations except the 15-30 and 30-60 cm depth intervals at the 20 cm inter-row location where ammonium-N concentrations increased (Figure 4.6). The largest decreases in ammonium-N concentrations occurred near the soil surface away from the crop row. The mean ammonium-N concentration on 12 August was 2.8 mg L'1, a net increase of 0.2 mg L'1 from 10 June. The ammonium-N concentration was highest near the soil surface, close to the raspberry row (Figure 4.6). There was a small net increase in the mean ammonium-N concentration between 12 August and 18 October from 2.8 to 3.0 mg L"1. There was a more uniform distribution of ammonium-N with soil depth than on previous sampling dates and the highest concentrations were near the crop row at all soil depths (Figure 4.6). Ammonium-N concentrations, averaged across depth, varied widely among inter-row locations on 3 May, but were similar for all inter-row locations on the other three sampling dates, resulting in a significant location x time interaction (Figure 4.7). There is no obvious explanation for the pattern of ammonium-N concentrations observed at the different inter-row locations observed on 3 May. Ammonium-N concentration, averaged across inter-row location, became more uniform with soil depth over time resulting in a significant depth x time interaction (Figure 4.8). Higher ammonium-N at the surface in the spring is likely a result of enhanced mineralization of soil organic matter and manure. A more uniform distribution of 65 Inter-row location (cm from raspberry row) 3 May 0 20 40 60 80 100 120 140 160 10 June 0 20 40 60 80 100 120 140 160 12 August 0 20 40 60 80 100 120 140 160 Figure 4.6. Soil ammonium-N (mg L"1) distribution with depth and inter-row location averaged for the control and 200 kg N ha"1 as poultry layer manure treatments for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations). 66 ammonium-N concentration with depth would be expected through time due to nitrification. 4.3.3. Nitrate-N concentrations at six inter-row locations over time in 1993 Soil nitrate-N concentrations also varied in space and time. Significant main effects were observed for inter-row location, soil depth, and time, but not treatment(Figure 4.9, Figure 4.10, Table 4.3). Significant interactions included treatment x time, location x time, depth x time, treatment x depth x time, and location x depth x time. On 3 May the mean nitrate-N concentration in the control treatment averaged across inter-row locations and soil depths, was 7.6 mg L"1. Nitrate-N concentrations were highest near the raspberry row in the 0-15 cm depth interval and in the centre of the inter-row, and gradually decreased with soil depth (Figure 4.9). By 10 June the mean nitrate-N concentration in the control treatment increased to 12.1 mg L 1. The highest nitrate-N concentrations were found mid-way through the soil profile furthest from the raspberry row and decreased with horizontal and vertical distance away from that area (Figure 4.9). Between 10 June and 12 August there was a further net increase in the mean nitrate-N concentration in the control treatment to 15.1 mg L"1. On 12 August, the highest nitrate-N concentrations were found furthest from the raspberry row near the soil surface, and concentrations decreased with soil depth and with closer proximity to the raspberry row (Figure 4.9). 67 Figure 4.7. Mean ammonium-N concentration for the control and 200 kg N ha"1 as poultry layer manure treatments, averaged across soil depth for each inter-row location on four sampling dates at the Columbia site in 1993. £ o. •a o Ammonium-N (mg L'1) 2 4 15 30 1 45 60 J • 3-May • 10-June • 12-Aug O 18-Oct Figure 4.8. Mean ammonium-N concentrations for the control and 200 kg N ha'1 as poultry layer manure treatments, averaged across inter-row location for each depth interval on four sampling dates at the Columbia site in 1993. 68 Inter-row location (cm from raspberry row) E o C D -C D T3 O 00 3 May 0 20 40 60 80 100 120 140 160 E o Q . C D TD 6 CO 10 June 0 20 40 60 80 100 120 140 160 E o Q . C U T3 O CO 12 August 0 20 40 60 80 100 120 140 160 E o C L C D T3 O CO 18 October 0 20 40 60 80 100 120 140 160 Figure 4.9. Soil nitrate-N (mg L"1) distribution with depth and inter-row location for the control treatment for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations). 69 Inter-row location (cm from raspberry row) 3 May 0 20 40 60 80 100 120 140 160 E 0 -15-x: -*-< C L CD 30-" O '5 45-CO 60-10 June 0 20 40 60 80 100 120 140 160 (cm) o-(cm) 15-depth 30-depth Soil 45-60-_1 . L 12 August 0 o--E 15-JC. Q. CD 30-T J O 45-CO 60-: 18 October 0 20 40 60 80 100 120 140 160 0-E 15-Q. CD 30-T3 O 45-CO 60-Figure 4.10. Soil nitrate-N (mg L"1) distribution with depth and inter-row location for the 200 kg N ha"1 as poultry layer manure treatment for four sampling dates at the Columbia site in 1993 (the soil surface at the 20 and 40 cm inter-row locations is approximately 20 cm higher than the remaining inter-row locations). 70 There was a net increase in the mean nitrate-N concentration in the control treatment from 15.1 to 19.6 mg L"1 between 12 August and 18 October. On 18 October the nitrate-N concentration decreased with soil depth, however there was a more uniform distribution of nitrate-N over soil depth toward the centre of the inter-row (Figure 4.9). Similar patterns of nitrate-N distribution with inter-row location and soil depth were observed in the 200M treatment although the concentrations were higher at all depths and inter-row locations (Figure 4.10). In addition, the pattern of soil nitrate concentration with depth tended to be uniform over a wider area (70 to 150 cm) on 10 June than for the control treatment. The mean nitrate-N concentrations, averaged across inter-row locations and soil depths, were 16.7, 19.3, 22.2 and 23.5 mg L'1 on 3 May, 10 June, 12 August and 18 October, respectively, in the 200M treatment. The distribution of nitrate-N with depth varied among treatments and sampling dates (Figure 4.11). The difference in nitrate-N concentration between the C and 200M treatments decreased over the growing season from 9.1 on 3 May to 3.9 mg L"1 on 18 October. Nitrate-N concentration for both treatments was uniform with depth on 3 May, highest in the 15-30 cm depth interval on 10 June, and decreased with depth on 12 August and 18 October indicating that downward migration of nitrate-N had taken place between 3 May and 10 June. 71 Figure 4.11. Nitrate-N concentration averaged across inter-row location for each depth interval for the control and 200 kg N ha"1 applied as poultry layer manure treatments on four dates at the Columbia site in 1993. 72 content on a unit area basis. The calculated soil inorganic N varied little among the six procedures. The 611, 6N, 3U, 3N, 2U and 2N calculation procedures gave mean inorganic N contents of 131, 129, 133, 127, 133 and 126 kg N ha 1, respectively, averaged across treatments and the four sampling dates when used. The calculation procedure did however have a significant effect on the calculated soil inorganic N content (Table 4.4). Table 4.4. Statistical significance of six different procedures for calculating soil inorganic N when more than one inter-row location was sampled. Source df Mean square Block 23 4950.4* Treatment 5 234.4* Error 115 79.7 *P<0.05. A comparison of means showed that with six inter-row locations it made no difference whether uniform or non-uniform weighting factors were used (Table 4.5). Where two or three inter-row locations were used, estimated soil inorganic N content was significantly lower using non-uniform than uniform weighting factors, however, the difference in estimated soil inorganic N content was small. The uniform weighting factors for both two and three inter-row locations provided an estimate of soil inorganic N content that was more similar to six inter-row locations. As a result, all calculations of soil inorganic N contents were done using the number of inter-row locations available and uniform weighting factors for those locations. 73 Table 4.5., Matrix of pairwise comparison probabilities using Fisher's least-significant-difference test (the number indicates the number of inter-row locations, II-uniform, N-non-uniform). Calculation procedure 611 6N 3U 3N 2U 2N 611 1.000 6N 0.468 1.000 3U 0.464 0.147 1.000 3N 0.104 0.363 0.019* 1.000 2U 0.525 0.175 0.923 0.025* 1.000 2N 0.029* 0.140 0.004* 0.567 0.005* 1.000 *P<0.05. 4.3.3.1. Soil ammonium and nitrate content over the growing season Soil ammonium-N contents were 21, 42 and 15 kg ha 1 prior to treatment application at C92, C93 and H92, respectively (Figure 4.12). At all three sites the highest soil ammonium-N contents were observed on the first soil sampling date after treatment application; 7, 13 and 7 days after treatment application for C92, C93, and H92, respectively. On subsequent sampling dates soil ammonium contents were similar to those observed before treatment application and remained relatively constant over the remainder of the growing season. There was a significant effect of treatment and date, and a significant date x treatment interaction on soil ammonium content at C92 and C93 (Figure 4.12, Table 4.6). Ammonium content differed between treatments for the 3 and 15 March sampling dates at C92 and C93, respectively, otherwise, the differences among treatments were very small. Soil ammonium content was similar for the 55F and 100M treatments at C92 or C93. 74 There was a significant effect of date, but not treatment, on soil ammonium content at H92 (Figure 4.12, Table 4.6). Similar to C92 and C93, soil ammonium-N was highest on the first sampling date following treatment application, and there was no significant difference in soil ammonium contents in the 55F and 100M treatments. Soil nitrate-N contents prior to treatment application were 6, 28 and 8 kg ha"1 at C92, C93 and H92, respectively (Figure 4.12). The net change in soil nitrate between the soil sampling dates before and after treatment application was very small in the control and manure treatments compared to the fertilizer treatment at all three sites. On the first sampling date following treatment application, nitrate-N made up 58% of the soil inorganic N in the fertilizer treatments, compared to 37% in the control and manure treatments, averaged across sites. On subsequent dates, nitrate-N made up the largest portion of the soil inorganic N in all four treatments. There was a significant effect of treatment and date, and a significant date x treatment interaction, on soil nitrate content at C92 (Figure 4.12, Table 4.6). On sampling dates prior to 9 June, the 55F treatment had higher soil nitrate than the 100M treatment, whereas on 9 June and subsequent sampling dates the reverse was true. Soil nitrate content increased linearly with increasing rate of poultry layer manure application. Soil nitrate content increased over the growing season and was highest on the 13 August sampling date when soil nitrate-N contents were 119, 142, 215 and 163 kg ha"1 in the C, 100M, 200M and 55F treatments, respectively. There was a net decrease of soil nitrate in all treatments between the 13 August and 5 October sampling dates. 75 3001 250-200 150-100 50 0 Ammonium-N Columbia 1992 • Control o 100M • 200M • 55F F' M A M' J J ' A' S ' O 300 250 200 150 100 50 0 Nitrate-N Columbia 1992 F M A M J J A S O 300-1 250-200-150-100-50-0 Ammonium-N Columbia 1993 F M A M J J A S O 300 250 200 150 100 50 0 Nitrate-N Columbia 1993 F M A M J 300 250 200 150 100 50-I 0 Ammonium-N Huntingdon 1992 F M A M J J A S O Month 300 250 200 150 100 50 0 Nitrate-N Huntingdon 1992 F M A M J J A S O Month Figure 4.12. Soil ammonium- and nitrate-N content to 60 cm over the growing season at the Columbia site in 1992 and 1993, and the Huntingdon site in 1992. 76 Table 4.6 Statistical significance of soil ammonium (kg ha'1) and nitrate (kg ha1) content as influenced by rate of N application as poultry layer manure and inorganic fertilizer and sampling date at three sites (55F-55 kg N ha'1 as ammonium nitrate, 100M-100 kg N ha"1 as poultry layer manure) (Values are mean squares). Source df Ammonium Nitrate Columbia 1992 Block Treatment Linear Quadratic 55F vs 100M Error Date Date x treatment Error Columbia 1993 Block Treatment Linear Quadratic 55 F vs 100M Error Date Date x treatment Error 2 3 1 1 1 6 5 15 40 2 3 1 1 1 6 6 18 48 39.1 566.7* 1656.5* 34.9 33.8 41.8 3236.8* 515.0* 80.3 365.2* 392.1* 1062.0* 104.2 10.9 23.8 6053.4* 364.7* 51.9 988.4* 18759.8* 56208.5* 1.6 56.8 868.4 24458.2* 1096.4* 163.6 7.3 6476.4* 15737.4* 1422.8 3681.1* 660.2 17700.1* 419.3* 172.0 cont. 77 Table 4.6 (continued) Source df Ammonium Nitrate HUNTINGDON 1992 Block 2 146.4 657.3 Treatment 3 300.0 3276.8* Linear 1 1.8 8331.1* Quadratic 1 649.8 506.9 55F vs 100M 1 614.9 382.7 Error 6 145.8 270.3 Date 2 3874.0* 81209.0* Date x treatment 6 187.8 939.0 Error 10 222.3 358.8 *P<0.05. Like C92, there was a significant effect of treatment and date, and a significant date x treatment interaction on soil nitrate content at C93 (Figure 4.12, Table 4.6). On sampling dates prior to 10 June, the 55F treatment had higher soil nitrate-N than the 100M treatment, whereas on 10 June and subsequent sampling dates the reverse was true. Soil nitrate content increased linearly with increasing rate of poultry layer manure application. Soil nitrate content increased over most of the growing season, similar to C92, however the rate of increase was not as rapid as observed for C92. The highest nitrate-N contents were measured on the 17 September sampling date at 155, 170, 205 and 178 kg ha 1 for the C, 100M, 200M and 55F treatments, respectively. There was a net decrease in soil nitrate between the 17 September and 18 October sampling dates in all four treatments. Soil nitrate was significantly higher in the 55F treatment than the 100M treatment at C93. 78 There was a significant effect of treatment and date, but no interaction between the two, on soil nitrate-N content at H92 (Figure 4.12, Table 4.6). Soil nitrate increased more rapidly over the growing season at H92 than at C92 or C93. Soil nitrate-N was highest on the 10 August sampling date when 181, 224, 262 and 225 kg ha"1 was measured in the C, 100M, 200M and 55F treatments, respectively. Similar to C92 and C93 there was a linear relationship between rate of N applied as poultry layer manure and soil nitrate-N. Soil nitrate content in the 55F and 100M treatments did not differ significantly. On the August sampling date after the berry harvest was complete there was a significant effect of site and treatment but no interaction of site and treatment, on soil nitrate at all three sites (Table 4.7; Table 4.8). Soil nitrate contents were higher, and generally less responsive to the applied treatments at H92 as compared to C92 and C93. Table 4.7 Soil nitrate-N to 60 cm (kg ha'1) on the August sampling date for the C, 100M, 200M, and 55F treatments (C-control, 100M-100 kg N ha"1 as poultry layer manure, 200M-200 kg N ha"1 as poultry layer manure and 55F-55 kg N ha"1 as ammonium nitrate). Site Treatment Columbia 1992 Columbia 1993 Huntingdon 1992 C 120 99 181 100M 163 112 225 200M 215 143 262 55F 142 124 224 79 Table 4.8 Statistical significance of nitrate content (kg ha1) on the August sampling date for Columbia 1992, Columbia 1993 and Huntingdon 1992 for the C, 100M, 200M, and 55F treatments (C-control, 100M-100 kg N ha"1 as poultry layer manure, 200M-200 kg N ha"1 as poultry layer manure and 55F-55 kg N ha"1 as ammonium nitrate). Source df Error mean squares Block 2 813.1 Treatment 3 6476.5* Site 2 27254.6* Site x treatment 6 511.5 Error 19 343.1 *P<0.05. 4.3.4. Raspberry N Uptake 4.3.4.1. Floricanes The applied treatments had no effect on mean berry weight or estimated mean berry yield. Mean berry weight averaged 2.8, 3.3 and 3.4 g, and mean berry yield averaged 13.4,15.3 and 15.01 ha'1, for C92, C93 and H92, respectively. The number of floricanes plot"1 averaged 94, 85, and 71 for C92, C93, and H92, respectively, and was independent of the applied treatments. The berry yield estimate increased linearly with increasing dry matter yield of the fruiting cluster at each site (Figure 4.13). The regression equations accounted for 81 and 87% of the variability in estimated yield at C92 and C93, respectively, compared to 56% of the variability at H92. 80 ro CD -i—« ro E co CD TJ CD > 22 20 18 16 " 14 -12 " 10 Columbia 1992 y=18.0x + 0.493,r^O.81 1 1 1 1 1 1 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 ro CD ro E .»-» co CD 35 CD 22 20 18 16 14 12 10 Columbia 1993 y=19.4x-1.04, ^=0.87 1 1 1— 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 ro CD ro £ co CD 32 CD 22 20 18 16 14 12 10 Huntingdon 1992 y=6.92x + 9.96, r^O.56 1 1 1 1 1 1 1 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Fruiting cluster dry matter yield (t ha"1) Figure 4.13. Relationship between the berry yield estimate and the dry matter yield of the fruiting cluster on a per plot basis at three sites. 81 Dry matter yield of the floricanes (stem plus fruiting cluster), averaged across treatments, was 3.54, 3.64 and 3.34 t ha 1 at C92, C93, and H92, respectively (Table 4.9). On average, 78% of the floricane dry matter was in the stems with the remaining 22% in the fruiting clusters. There was no significant effect of treatment on dry matter yield of the floricane or floricane components at C92 or H92 whereas dry matter yield of the stem portion of the floricane increased linearly with increasing rate of poultry layer manure application at C93 (Table 4.10). The contrast comparing the 55F and 100M treatments was significant for the stem and whole floricane dry matter yield at C93 and H92. Dry matter yield was higher for the 100M treatment than for the 55F treatment for the stems and total floricane at C93, whereas the reverse was true at H92. The mean N concentration of the floricanes (stem plus fruiting cluster) averaged across treatments was 1.42, 1.46 and 1.55% at C92, C93, and H92, respectively (Table 4.9). On average, the fruiting cluster N concentration was 46 and 30% greater than the stem portion of the floricane at C92 and C93, respectively, compared to 60% at H92. Nitrogen concentration of the whole floricane increased linearly with increasing rate of poultry layer manure application at all three sites (Table 4.10). Similarly, N concentration of the floricane stem increased linearly at C93 and H92, and curvilinearly at C92, with increasing rate of manure application. Nitrogen concentration of the fruiting cluster increased linearly with increasing manure application rate at C92, but was not influenced by manure application rate at C93 or H92. 82 Table 4.9 Dry matter yield, N concentration, and N uptake of the fruiting cluster, stem and whole floricane as influenced by the rate of N application as poultry layer manure and inorganic fertilizer at three sites. (C-control; 55F-55 kg N ha 1 as ammonium nitrate; 100M-100 kg N ha 1 as poultry layer manure; 200M-200 kg N ha"1 as poultry layer manure). Treatment Site and floricane component C 100M 200M 55F Mean Dry matter yield (t ha1) Columbia 1992 Stem 2.63 3.06 2.83 2.74 2.82 Fruiting cluster 0.66 0.77 0.72 0.73 0.72 Whole floricane 3.29 3.83 3.55 3.47 3.54 Columbia 1993 Stem 2.57 3.08 3.08 2.44 2.79 Fruiting cluster 0.84 0.94 0.91 0.73 0.86 Whole floricane 3.40 4.02 3.99 3.16 3.64 Huntingdon 1992 Stem 2.81 2.47 2.11 3.10 2.62 Fruiting cluster 0.72 0.65 0.66 0.82 0.71 Whole floricane 3.53 3.13 2.77 3.93 3.34 N Concentration (%) Columbia 1992 Stem 1.16 1.33 1.35 1.33 1.29 Fruiting Cluster 1.77 1.88 1.97 1.93 1.89 Whole floricane 1.28 1.44 1.48 1.46 1.42 Columbia 1993 Stem 1.33 1.35 1.47 1.32 1.37 Fruiting Cluster 1.70 1.78 1.82 1.78 1.77 Whole floricane 1.42 1.45 1.55 1.42 1.46 cont. 83 Table 4.9 (continued) Treatment Site and floricane component C 100M 200M 55F Mean Huntingdon 1992 Stem 1.25 1.36 1.51 1.38 1.38 Fruiting Cluster 2.11 2.24 2.04 2.36 2.19 Whole floricane 1.43 1.54 1.64 1.57 1.55 N Uptake (kg N ha1) Columbia 1992 Stem 30.4 40.6 38.3 36.3 36.4 Fruiting Cluster 11.6 14.4 14.1 14.0 13.5 Whole floricane 42.0 55.1 52.4 50.3 50.0 Columbia 1993 Stem 34.0 41.5 45.2 32.1 38.2 Fruiting Cluster 14.2 16.6 16.6 12.9 15.1 Whole floricane 48.2 58.1 61.8 45.0 53.3 Huntingdon 1992 Stem 35.1 33.6 32.0 42.8 35.9 Fruiting Cluster 15.2 14.6 13.6 19.1 15.6 Whole floricane 50.4 48.2 45.5 61.9 51.5 Nitrogen uptake by the floricanes (stem plus fruiting cluster) was 50, 53 and 52 kg N ha"1 at C92, C93 and H92, respectively (Table 4.9). On average, the stem portion of the floricane accounted for 71 % of the N uptake for all three sites, while the fruiting cluster accounted for the remaining 29%. Nitrogen uptake by the floricane and floricane components was not influenced by treatment at C92 (Table 4.10). At C93, N uptake by the stem and whole floricane increased linearly with increasing rate of N application as poultry layer manure. Nitrogen uptake by the stem and whole floricane for the 84 co i2 c £ CO d « O O O x: 'co CO ->£ CD _ CO Q . 3 E CD CO I -I o -x - C c o CO •g O o c o O E 0 CO _CD O CO 3 3 CD >» CD ts CO E Q o LL E CD 4—• CO CD o k 3 o CO CO •xf 00 CD CD CO CO CM O X O X. O X LO cn CD O LO r--LO o o CM o> T— cu !5 E 3 o o CM CD cq c- CO CM CD CO CM CO CO 06 CO 1 "~ "xf cn 0 CM CO CM CO cn cri LO 0 C\i CO CO r f CD CD CO CM * CD * 0 •xf 1— r f CM CD CD CO CD CM CO * O CM O r f CM CM CD CD O CO * CM * CM O * CM T _ r f CM CO LO CD 1— CO CO r f LO T— 0 CD r f CO CD CD CD 0 CO r- r f CD CD CD* LO CO o o CQ CO CM CD CD CM CM O LO CM CM CD r f LO O 1— X o CD $r <°. o O T— LO 1— CTi r f T- CM CD CD CM CM CM CO CM CD CJi r f O LO CD CM CO CO r f CD CM LO LO CM 0 CD CO r-- CO CD CD r f ! CD CD LO T CO CO O CO CM LO CO CM " r _ T— CM CM CO 1— 1 _^ CD CM c CD E •a CO CO CO c g 'to k-TD CO o o o 1— CO > LL LO LO * CO r--CM LO CD T- r f rf rf O LO LO co""* b 0 co T— CO Is- CM * CD CD r f CO CO o> cn (0 c !5 CO Error olum lock reatm Error O CQ 1-CD CD CM r f CD CO 00 * LO 00 LO CM CD CO CD CD LO CO 06 CO CD CM * CO * 00 00 CO CO * "xf LO LO CM CM T— CD CD CM T -O CM O r f LO O 1^ C r 1 - LO U •xf 0 CO LO LO T- T- O LO CM T -00 c 0 CD cn T — CO T c c * LO T - CO CD CD CO T— C CO 1 - CD 1— T - 1— C CO CD C o 'to T3 CO O o o CO > LO LO I 85 o .5 "I CO 4—» Q . 3 C CD O c o IO O > i CO I o c o CO CO O IX. E CD +-» CO 0 o 12 CD > . CO E IQ >. o E CD CO lo X lo lo X l o lo X lo * fc r- CD - r - * r- CD fc T— CM cb o LO d 00 CM d CM LO CO fc •xt y— O oo •xt fc 00 d •xt d •xt d d ib * o CO CO CO LO CO LO 00 * cn CD CM d CO LO cb d CM CM 1— ib 00 CM o •xt * fc LO O) fc o> •xt CO 00 LO d •xt CO o o CM 00 o 00 00 GJ 00 00 •xt LO CO d •xt •xt •xt 00 •xt •xt CM fc * o CM 00 fc p 00 •xt •xt CM 00 r-cb •xt LO CO LO * CM •xt CO •xt CO CM LO CO fc LO d ib CM 00 fc fc-CO •xt 00 LO 00 00 00 00 00 00 ob cb CM 00 CO •xt CD O * p LO CO 00 CM •xt 00 CM fc^ CM ib cb LO LO d 00 CM CO CO T3 CD C V-< c o o •xt 0 - Q CO I-0 O * O CO c i U) c c 3 O o CQ c 0 E to 0 CO 0 c o to •o CO n O o o 1— co > UL LO LO O i— UJ I LO o d 86 100M treatment was greater than for the 55F treatment at C93, whereas the reverse was true at H92. 4.3.4.2. Primocanes Mean primocane dry matter yield, averaged across treatments, was 3.36 and 3.97 t ha 1 at C92 and C93, respectively and there was no significant effect of treatment at either site (Table 4.11). Mean primocane N concentration was not affected by treatment, and was 1.33 and 1.24% at C92 and C93, respectively. Primocane N uptake was significantly higher in the 55F treatment (54 kg ha1) compared to the 100M treatment (43 kg ha1) at C92. Treatment did not influence primocane N uptake at C93 and averaged 49 kg N ha 1. Mean cane weight was 52.6 and 64.0 g cane1 at C92 and C93, respectively, and was not influenced by treatment. 4.3.4.3. Crop N uptake and apparent N recovery Total N uptake by the floricanes (adjusted for N present in the woody portion of the stem in the spring) and the primocanes, averaged across treatments, was 90 and 97 kg ha'1 at C92 and C93, respectively. Apparent N recoveries in the whole crop were low, particularly at C93, and varied widely among treatments at C92 (Table 4.12). Apparent N recoveries for the fruiting clusters were less than 5% at all three sites (data not shown). 87 Table 4.11 Statistical significance of primocane dry matter yield (t ha1), N concentration (%), N uptake (kg ha"1), and mean cane weight (g cane1) as influenced by the rate of N application as poultry manure and inorganic fertilizer for two years at the Columbia site (55F-55 kg N ha'1 as ammonium nitrate, 100M-100 kg N ha"1 as poultry manure; values are error mean squares). Source df Dry matter yield N Cone. N uptake Mean cane weight COLUMBIA 1992 (x103) (x10"3) Block 2 68.2 0.16 17.7 62.9 Treatment 3 534.0 9.25 145.3* 88.3 Linear 1 47.0 20.42 50.2 248.7 Quadratic 1 12.7 2.45 8.1 8.6 55F vs 100M 1 900.2 0.82 202.4* 0.3 Error 3 189.8 5.45 24.7 48.5 COLUMBIA 1993 (x10"3) (x10"3) Block 2 327.2 0.08 54.8 98.7 Treatment 3 205.0 3.25 19.5 24.2 Linear 1 177.2 0.27 38.1 45.5 Quadratic 1 158.5 5.69 5.0 4.3 55F vs 100M 1 437.4 8.82 20.2 7.2 Error 3 570.2 3.59 61.2 49.3 *P<0.05. Table 4.12 Apparent N recovery (%) in the whole crop at the Columbia site in two years. Rate of N applied 1992 1993 55F 24 4 100M 28 9 200M 7 8 Mean 20 7 88 4.3.4.4. Discussion Soil moisture content was relatively high on all four sampling dates when the six inter-row locations were sampled. The decrease in soil moisture content with depth was likely due to the lower water holding capacity deeper in the soil profile as a result of higher sand content and lower clay and organic matter contents (Table 3.1). The mound present at the base of the crop row formed in a deeper A horizon, and therefore had a higher water holding capacity due to increased organic matter content, resulting in higher moisture contents near the crop row (ie. 20 cm location) as compared to the other inter-row locations. No pattern of reduction in water content between sampling dates, which could be used to identify the pattern of root uptake, was observed possibly because of the high soil moisture contents, and frequent rainfall and irrigation events. The relatively high ammonium-N concentrations on 3 May may indicate a period of rapid mineralization since soil moisture conditions were near ideal and temperatures were relatively warm compared to early spring. Simultaneously, conditions may have been less favorable for nitrification and caused ammonium accumulation. Kowalenko (1989), in a study under similar climatic conditions, found ammonium accumulation in May and attributed it to the fact that nitrification did not keep up with mineralization due to cooler soil temperatures. Since ammonium did not accumulate in the soil profile on the next three sampling dates, nitrification must have followed mineralization rapidly. There was surplus of precipitation over potential evaporation between 3 May and 10 June. The occurrence of maximum nitrate concentration in the 15-30 cm depth rather than the 0-15 cm depth, which would be expected from mineralization/nitrification of soil organic matter and manure, suggests downward movement of nitrate occurred in June. This is consistent with the water surplus between 3 May and 10 June. In the four days 89 preceding the 10 June sampling date 52 mm of rain fell, over 50% of which fell on 9 June. There was a net increase in nitrate in the 30-60 cm depth in this period as well, indicating the possibility that some leaching of nitrate below 60 cm depth did occur. Zebarth and Paul (1996) showed evidence of significant nitrate leaching on a sandy soil in the Fraser Valley'during a wet spring. The 42 mm decrease in soil moisture content to a depth of 60 cm between 10 June and 12 August did not account for the amount that potential evaporation exceeded precipitation so irrigation of approximately 75 mm of water could have taken place or, actual evaporation was much less than potential evaporation. A decrease in soil moisture content in that time interval would be expected since the rate of water usage by raspberry plants is greatest during berry enlargement and harvest (Crandall 1980). The raspberry roots may have taken up water below 60 cm. Studies of raspberry root distribution have shown that the majority of roots are concentrated in the upper 20 to 30 cm of soil (Christensen 1947; Pehoto 1968; Seipp 1986), although some roots may penetrate deeper into the soil profile. Christensen (1947), in detailed excavations in a sandy loam soil, found raspberry roots as deep as 175 cm while Crandall (1980) found raspberry roots as deep as 120 cm in a well drained soil. Since potential evaporation was greater than precipitation between 12 August and 18 October, a decrease in moisture would have been expected. A possible explanation for the small net increase in soil moisture content in that time interval is that irrigation provided the extra water. Due to the more uniform distribution of nitrate with depth it appears as though the fall leaching may have started by 18 October. The similarity of nitrate concentrations in the 0-15 cm depth interval on 10 June and 18 October in the C and 200M treatments 90 gives further evidence for leaching since leaching of the readily available nitrate would reduce the difference in nitrate concentrations between the two treatments. Nitrate-N concentration, averaged across depth and inter-row location, increased between 10 June and 12 August. During this period decreases in nitrate concentration were observed at the 20, 40 and 68 cm inter-row locations in the 15-30 and 30-60 cm depth increments in both the C and 200M treatments and increases were observed in the 0-15 cm depth increment. The lack of root uptake in the 0-15 cm interval may be attributed to rotovation to 10 cm below the soil surface between the 40 and 150 cm inter-row location. There is no evidence of a similar pattern of. increased water usage at the 15-30 and 30-60 cm depth increments near the crop row. The absence of a pattern of water usage indicates that water was kept in abundant supply by rainfall and/or irrigation. Root growth beyond the 68 cm inter-row location may have been impeded by the compacted zone of the tractor wheel path centred at 68 cm. Bristow and Brun (1987) found fewer raspberry roots beyond the band of compaction, caused by tractor wheels centred at 69 cm from the plant row, however roots did pass under the compacted zone. The pattern of nitrate removal from the root zone appears to be more obvious in the control where nitrate is more limited than the 200M treatment, although there was no significant effect of treatment. This suggests that N uptake was generally closer to the crop row for the manure treatment than for the control. Nitrification occurred rapidly in the spring as indicated by the low soil ammonium by the third sampling date at all three sites. The mean soil temperature in March 1992 was 10°C at 5 cm compared to only 6°C in March 1993. Despite the lower soil temperatures in 1993 nitrification still proceeded rapidly. Rapid nitrification in this experiment is consistent with the results obtained from Experiment 1. 91 The low soil ammonium content prior to treatment application at all three sites was likely due to a combination of limited soil N mineralization in early spring due to the cool soil temperatures, and loss of nitrate from the root over the fall and winter by leaching. Essentially all nitrate is leached from the root zone over the winter months in the lower Fraser Valley because of the high rainfall and low evapotranspiration (Kowalenko 1987; Zebarth et al. 1996). Similar soil nitrate contents were measured for the 55F and 100M treatments at C92 and H92. This is consistent with the approximately 50% recovery of manure N as soil inorganic N reported previously (Section 4.2). In contrast, soil nitrate content was higher for the 55F than the 100M treatment at C93. This may have resulted from higher N losses from the 100M than the 55F treatment, possibly as a result of denitrification. Loss by denitrification would be enhanced where manure was applied since manure would provide the necessary carbon source to fuel the process. Poultry manure is very high in volatile fatty acids, which are important for denitrification in manure amended soils (Paul and Beauchamp 1989). Rolston et al. (1978) found that at water contents very close to saturation, approximately 73, 14, and 3% of the applied KN0 3 fertilizer was lost from manured, cropped and uncropped treatments, respectively. The N loss cannot be attributed to leaching, which would favour greater nitrate loss for the 55F treatment due to higher nitrate contents after treatment application relative to the 100M treatment. Alternatively, the lower soil nitrate content for the 100M treatment at C93 may have been the result of reduced manure N availability in 1993. However in Experiment 1 (Section 4.2.3) similar manure N recovery as soil inorganic N was obtained in 1992 and 1993 despite the cooler soil temperatures in 1993. 92 The higher soil nitrate contents at H92, as compared to C92 and C93, were likely due to the history of manure application at H92. Maidl and Fischbeck (1989) compared soil nitrate levels at two sites; one with a long term history of swine slurry application and one with no history of slurry application. They found that the increase of soil nitrate between February and May was twice as high at the site with a history of slurry application compared to the site with no slurry application. In a mineralization rate prediction model proposed by Pratt et al. (1973) in California for poultry manure, 90% of the applied manure would be available in the year of application, while 10, and 5%, of the remaining N would be available in the second year, and the third years, respectively. Soil nitrate concentrations in August or September may provide a good indication of the potential for nitrate leaching from raspberry over the fall and winter, provided that sampling is done before any significant leaching has occurred. Soil nitrate contents are at or near a maximum at this time. The decrease in soil nitrate between August and October at C92 and September and October at C93 indicate that either leaching had occurred or primocane uptake continued into the fall. It is assumed that loss of nitrate would have taken place at H92 in the same time interval. Leaching could have occurred in response to either rainfall or irrigation. At C92 and H92, it is not known whether soil nitrate levels had reached their peak in August or if a further net increase had occurred by September since no samples were taken in September 1992. Total rainfall in September 1992 was close to the long term normal whereas in September 1993 rainfall was exceptionally low which could account for the fact that no losses of nitrate were observed between the August and September sampling dates in 1993. Berry weights measured in this study were similar to those reported by others (Hill 1958; Kowalenko 1981; Papp 1984; and Martin et al. 1980). The lack of response of berry weight and estimated berry yield to various rates of N application may be 93 attributed to factors other than the addition of N in the year of study. Ljones (1965) suggested that berry yield may be less influenced by fertility in the year of fruit development than in the previous year when the primocane developed. Also, mineralization of soil organic N appears to have provided enough N early in the spring (Section 4.2) for lateral and fruiting cluster growth and development where no N was applied. Translocation of N from other parts of the plant to the fruiting cluster could also have occurred (Kowalenko pers. comm.), minimizing any influence of treatment. Life stage of the raspberry plant may also influence fruit production and yield. Lawson and Waister (1972) showed a positive response to N addition in the first and second year of raspberry production, no response in the third year, and a negative response in the fourth year. Also, in a four year trial, Kowalenko (1981) found a relatively small response of yield to 134 kg N ha 1 compared to 1 kg B ha 1 in a planting that was two years old when the study began. He concluded that the response to N was somewhat luxuriant and largely due to increased berry size. Seipp (1986) in a study in Germany found that a higher N supply increased fruit size but slightly reduced yields. Due to the strong correlation between dry matter yield of the fruiting cluster and estimated raspberry yield by counting individual fruits and flowers, the dry matter yield of the fruiting cluster may be used to provide relative yield estimates within an experiment. This method would eliminate counting individual fruits and flowers (Daubeny et al. 1986), which is very labour intensive. The dry matter yield method does not give an estimate of yield but can measure differences between treatments. The poorer correlation between estimated berry yield and fruiting cluster dry matter yield at H92 compared to C92 and C93 could be partially due to the more limited range of yield estimate values. 94 Floricane dry matter yield in the present study was similar to that found by Wood et al. (1962) in a two year study in Scotland. They measured a floricane dry matter yield of 3.64 t ha"1 averaged across six varieties, although fallen primocane leaves were included in the total. The inter-row spacing in that study was 1.8 m compared to 3.0 m in the present study. Kowalenko (1994a) found a lower floricane dry matter yield of 2.61 ha"1 averaged over 4 years for the Willamette variety under similar climatic and soil conditions and a similar inter-row spacing, to the present study. There is no apparent reason why dry matter yield of the floricane stem and whole cane was higher in the 100M treatment than the 55F treatment at C93 and lower in the 100M treatment at H92 since soil inorganic N contents were similar in both treatments at the two sites at the time of berry harvest. A N concentration of 1.47% for the whole floricane was found by Wood et al. (1962), similar to the floricane N concentration found in this study. Fruiting cluster N concentrations found in other studies were higher than measured in this study: 2.6% N in the fruiting cluster reported by Wright and Waister (1980) and an average of 2.3% N over June and July reported by Kowalenko (1994b). There was a lack of measurable crop response to N addition, therefore the apparent linear increase in N concentration in the stem and fruiting cluster at C92 and C93 and whole floricane and stem at H92, with increased rate of N application may have been luxury uptake since there was no measurable effect of treatment on berry or dry matter yield. Nitrogen uptake by the floricane and its components followed similar patterns to dry matter yield and N concentration. Since the influence of N application did not affect yield the greater N uptake at higher rates of N application may simply have been luxury uptake. Wood et al. (1962) reported 54 kg N ha"1 was taken up by the floricanes, which 95 is very similar to this study. In comparison, Kowalenko (1994b) reported N uptake of 34 kg ha 1 averaged over 4 years. It is important to note that with the exception of less than 16 kg N ha-1 removed from the field in the fruit, all the floricane components are returned to the soil. Similar to the dry matter yield, there is no apparent reason for the difference between the 55F and 100M treatments for N uptake. Primocane samples taken late in August by Wright and Waister (1980), in Scotland in a one year study, contained 67 kg N ha'1 which was 20 kg ha"1 greater than measured in this study in October. However, their study included leaves that had fallen to the ground whereas the present study did not. Also the inter-row spacing was 1.8 m in the Wright and Waister study compared to 3.0 in this study. Kowalenko (1994a) found a N uptake of 37 kg N ha'1 by primocanes which is more similar to this study. Luxury uptake of N by primocanes may be detrimental to the crop by increasing the risk of winter injury (Seipp 1986). Primocane N uptake is not necessarily all from the soil as some translocation from the floricane may occur (Kowalenko pers. comm.). The apparent plant N recovery of manure and inorganic fertilizer N was low, while there was minimal N recovered by the fruiting cluster. A vegetable crop, with good N uptake and low soil N at the end of the growing season, typically has an apparent N recovery of 50% (Greenwood et al. 1989). Plant populations are higher for vegetable crops than for the widely spaced raspberry crop, whose inter-row spacing is largely determined by the machinery used for harvesting, and may partially explain the lower crop N recovery found in this study. Alternatively, the low N recovery suggests that crop N demand is low relative to soil N supply. 96 4.4. Experiment 3: Influence of an Inter-row Cover Crop on Primocane Growth and Vigour and Soil Inorganic N Content Inter-row cover crop dry matter yield was not influenced by rate of N applied as poultry layer manure and averaged 2.63 and 2.291 ha"1 for 1992 and 1993, respectively. Similarly, neither the N concentration of the cover crop, average of 3.86 and 3.74% in 1992 and 1993, respectively, or cover crop N uptake, average of 102 and 85 kg N ha"1 in 1992 and 1993, respectively, were influenced by manure application. The cover crops reached the stem elongation growth stage (GS #3) according to Zadoks' code (Zadoks et al. 1974; Tottman et al. 1979; Appendix C). The cover crops reached a mean height of 85 and 80 cm in 1992 and 1993, respectively. Average primocane dry matter yield was 3.17 and 3.81 t ha"1 in 1992 and 1993, respectively, and was not significantly affected by manure application or the presence of a cover crop. Primocane N concentration averaged 1.33 and 1.22%, and primocane N uptake averaged 42 and 46 kg N ha'1, in 1992 and 1993, respectively, and was not significantly affected by manure application or the presence of a cover crop. Similarly, primocane vigour, estimated as the mean cane weight, was 53.0 and 65.2 g cane"1, in 1992 and 1993, respectively, but was not influenced by manure application or the presence of a cover crop. Based on soil samples taken at the time of cover crop seeding in August, application of 200 kg N ha"1 as poultry manure increased soil nitrate content from 115 to 205 kg ha 1 in 1992, and from 96 to 140 kg ha"1 in 1993 (Figure 4.14; Table 4.13). Mean soil ammonium content in mid-August was 15 and 20 kg ha'1 in 1992 and 1993, respectively and was not influenced by treatment. 97 200M co sz CO Z o "c CO E ? o c o CO August 1992 200M + CC • NH/-N 200M August 1993 200M+CC C+CC CO sz CO Z o 'c CO E> o _c 'o CO 200M October 1992 2 5 0 "i 200 200M +CC October 1993 50 CO CO o 'c C O E° o _c o CO -50 -100 August-October 1992 50 -i c August-October 1993 200M 0 C+CC ^ ™ " 5 0 200M+CC -100 200M+CC Figure 4.14. Soil inorganic N content to 60 cm depth for August and October, and the difference in soil nitrate content to 60 cm depth between August and October, in two years as influenced by the application of poultry layer manure and the use of an oats inter-row cover crop (C-control, 200M-200 kg N ha"1 as poultry layer manure, CC-cover crop). 98 By 5 October 1992, 52 days after cover crop seeding, soil nitrate was still significantly higher in the treatments where manure was applied (160 kg ha1), compared to those where no manure was applied (88 kg ha1) (Figure 4.14; Table 4.13). Soil nitrate was also higher where no cover crop was seeded (149 kg ha'1), compared to where a cover crop was present (98 kg ha'1). Soil ammonium was affected by treatment, however the difference in ammonium content among treatments was less than 4 kg ha 1. The net decrease in soil nitrate between 13 August and 5 October 1992, was significantly higher where manure was applied (45 kg ha1) than where no manure was applied (27 kg ha'1) (Figure 4.14; Table 4.13). The net decrease in soil nitrate-N was also higher in the cover crop treatments (55 kg ha1) compared to the treatments with no cover crop (17 kg ha1). The difference in ammonium-N in the same time interval was not affected by treatment. In 1993 the cover crop significantly reduced soil nitrate, but not ammonium contents by 18 October (Figure 4.14; Table 4.13). Soil nitrate contents were 144 and 39 kg ha'1 for the non-cover crop and cover crop treatments, respectively. Soil nitrate content was also higher in the manured (104 kg ha'1) compared to the non-manured (79 kg ha'1) treatments. In the 66 days between the August and October soil sampling dates the change in nitrate was affected by the presence of a cover crop. There was a net increase in nitrate of 24 kg ha"1 where no cover crop was present compared to a net decrease in soil nitrate of 76 kg ha 1 where a cover crop was present. 99 Table 4.13 Statistical significance of soil ammonium- and nitrate-N (kg ha1) as influenced by N application as poultry layer manure and an inter-row cover crop (CC) for August and October, and the difference between August and October, in two years (Values are mean square errors). August October - August-October -Source df NH4+-N N03--N NH4+-N NO3--N NH4+-N NO3 -N 1992 Block 2 20.296 240.6 43.68* 268.6 111.4* 100.1 Treatment 3 3.581 8143.1* 9.02 7798.4* 8.0 1696.5* Manure 1 0.001 23751.4* 16.59* 15480.8* 16.7 881.8* CC 1 2.803 561.8 10.10 7839.2* 2.3 4203.4* ManurexCC 1 7.938 116.1 0.36 75.1 4.9 4.4 Error 6 5.371 438.6 2.40 606.4 6.1 122.7 1993 Block 2 71.075* 84.48 18.49* 19.8 34.04* 182.6 Treatment 3 1.284 1988.31* 4.71 11767.8* 5.64 10327.9* Manure 1 1.665 5872.42* 6.69 1951.3* 1.66 1053.6 CC 1 2.108 92.52 5.07 33323.3* 13.67 29902.1* ManurexCC 1 0.078 0.01 2.38 29.0 1.59 28.0 Error 6 2.852 104.18 3.00 260.4 3.38 295.7 * P< 0.05. 4.4.1. Discussion The mean N uptake of 93 kg ha 1 of the cover crop does not include the roots. Meisinger et al. (1991) in a review of cover crop research stated that roots may contain an additional 30% of the plant dry matter, suggesting that the N uptake in this study may be greater than 93 kg N ha"1 for Saia oats. Since cover crop dry matter yield was not influenced by the addition of N as manure in the spring, the N content of the soil must not have been limiting. 100 Mineralization of N over the growing season in the control treatment provided sufficient N for cover crop growth. The cover crop did, however, have higher N uptake when additional N was applied. The presence of the Saia oats cover crop had no effect on primocane growth and resulted in a decline in soil nitrate. Therefore the cover crop can be effective in taking up excess soil N at the end of the growing season without adversely influencing growth and vigour of the primocanes. This result is consistent with a four year study by Sanderson and Cutcliffe (1988), which compared clean cultivation; cultivation till harvest with oats seeded after; and permanent sod in raspberries in a study on Prince Edward Island. They found that fruit yields were 19 to 35% lower in the sod treatments by the third year of a four year study, however yields were not decreased where oats were seeded after harvest. Characteristics affected by the permanent sod were cane height and number of buds per cane. Zebarth et al. (1993) in a four year study of inter-row groundcovers in raspberries in the Lower Fraser Valley found that crop vigour, as indicated by cane diameter, was significantly higher with a barley inter-row cover crop compared to perennial ryegrass and sheep's fescue in two years and total berry yield was higher in the barley treatment. Few studies which involve the measurement of cover crop N uptake, also examine soil N concentrations. Brinsfield and Staver (1991) in a study in Maryland found that a rye cover crop seeded 1 October, immediately after corn harvest, yielded 57 kg N ha-1 at 48 days after seeding. In the same time period soil nitrate in the top 30 cm of soil decreased by approximately 34 kg ha 1. The lower N uptake by the cover crop and the smaller decrease in soil N in their study compared to the present study is probably due to the later time of seeding. With raspberries it is possible to seed cover crops earlier relative to other crops thereby maximizing N uptake in the fall. 101 4.5. Experiment 4: Comparison of Cover Crop Species and Date of Seeding Dry matter yield of the oats inter-row cover crop decreased linearly in 1992, and curvilinearly in 1993, with delayed seeding date (Table 4.14; Figure 4.15). Dry matter yield was greater than 2 t ha"1 in both years for cover crops seeded by mid-August, whereas those seeded in the last week of August or later yielded less than half that amount. Table 4.14 Statistical significance of dry matter yield (t ha"1), N concentration (%), and N uptake (kg ha"1) as influenced by date of seeding and cover crop species in two years (Values are error mean squares). Source df Dry matter yield N concentration N uptake 1992 (x103) (x10"3) Block 2 93.4 1088.8* 202.4 Treatment 6 1833.0* 4986.8* 1837.2* Linear (date) 1 9155.1* 21009.1* 9760.6* Quadratic (date) 1 274.3 45.5 14.1 Barley vs Wheat 1 14.6 100.0 64.2 Oats vs Barley + 1 1048.2* 104.3 1966.7* Wheat Error 10 61.5 243.7 161.5 1993 (x10"3) (x10"3) Block 2 67.1 5.7 34.4 Treatment 6 3788.7* 6388.9* 3055.1* Linear (date) 1 17431.7* 16894.9* 16417.7* Quadratic (date) 1 1362.9* 477.8 270.7 Barley vs Wheat 1 26.8 27.6 37.9 Oats vs Barley + 1 28.5 911.7* 711.3* Wheat Error 10 65.7 226.2 98.8 * P<0.05. 102 CD 33 co •>. (D t -CC E c o « CD i— -*—' c CD o c o o - • 1992 • 1993 August September 120 100 'ro CO 80 ro 60 "Q. Z 40 20 0 August September Date of seeding Figure 4.15. Influence of seeding date on cover crop a) dry matter yield, b) N concentration, and c) N uptake in two years (Dry matter yield and N uptake were modified assuming that the cover crop was present on only 75% of the field). 103 Nitrogen concentration in the oats cover crop increased linearly with delayed seeding date in both years (Table 4.14; Figure 4.15). Cover crops seeded on the earliest date in 1992 and 1993 had an average N concentration of 3.2%, whereas cover crops seeded on the latest date had an average N concentration of 6.5%. Nitrogen uptake by the oats cover crop decreased linearly with delayed seeding date in 1992 and 1993 (Table 4.14; Figure 4.15). Cover crops seeded in the last week of August took up an average of 43 kg N ha 1 for 1992 and 1993, whereas cover crops seeded approximately two weeks earlier took up 80 kg N ha 1, almost twice as much. Dry matter yields of barley and spring wheat seeded on 14 August 1992 were not significantly different, and averaged 1.3 t ha 1; however their yield was significantly lower than the 2.1 t ha"1 yielded by the oats seeded on the same date (Table 4.14; Table 4.15). In 1993 when all three cover crop species were seeded with the same number of seeds per unit area, dry matter yield averaged 2.2 t ha"1 and was independent of crop species. Nitrogen concentration of the cover crops were similar for all three species in 1992 and averaged 3.8%. In 1993 the N concentration of the barley and spring wheat did not differ and averaged 2.8%, but was significantly less than the N concentration of 3.5% measured for the oats. Nitrogen uptake for the barley and spring wheat did not differ and averaged 49 and 60 kg N ha"1 in 1992 and 1993, respectively, which was less than the 80 and 78 kg N ha"1 for the oats in 1992 and 1993, respectively. 104 Table 4.15 Dry matter yield, N concentration and total N uptake for the oats, barley and spring wheat seeded in mid-August in two years. Cover crop species Year Oats Barley Spring wheat Dry matter yield (t ha"1) 1992 2.1 1.3 1.3 1993 2.3 2.2 2.1 Mean 2.2 1.8 1.7 N concentration (%) 1992 3.8 3.4 4.0 1993 3.5 2.8 2.7 Mean 3.7 3.1 3.4 N uptake (kg ha'1) 1992 80.4 43.8 53.3 1993 78.5 62.2 57.2 Mean 79.5 53.0 55.3 The oats and spring wheat seeded by mid-August reached the stem elongation growth stage (GS # 33) according to Zadoks' code (Zadoks et al. 1974; Tottman et al. 1979) by harvest in both years and were greater than 80 cm in height (Table 4.16; Appendix C). Inflorescence (GS # 59) was completed in both years for the barley, which reached a height of 58 and 98 cm in 1992 and 1993, respectively. Oats seeded late in August 1992 reached the stage of stem elongation (GS # 32, 70 cm tall), while in 1993 they only reached the tillering stage (GS # 22, 40 cm tall). The oats seeded early in September reached the tillering stage in both years (GS # 22, 40 cm tall). In 1992 the oats seeded in mid-September reached the tillering stage (GS # 21, 35 cm tall), while in 1993 they only reached the seedling growth stage (GS # 12, 21 cm tall). 105 Table 4.16 Growth stage of cover crops at time of harvest expressed using Zadoks' decimal code (growth stage), and height of the above-ground portion of the cover crop in 1992 and 1993. Date of seeding and cover crop species Growth stage Height (cm) 1992 5 August—Oats 34 105 14 August—Oats 33 85 25 August—Oats 32 70 4 September—Oats 22 40 15 September—Oats 21 35 14 August—Barley 59 58 14 August—Wheat 33 83 1993 30 July—Oats 34 117 13 August—Oats 33 80 26 August—Oats 22 62 26 September—Oats 22 41 10 September—Oats 12 21 13 August—Barley 59 98 13 August—Wheat 33 96 4.5.1. Discussion In this study, seeding an oats cover crop by the second week in August yielded twice as much dry matter and nitrogen as those seeded two weeks later. Therefore cover crops grown to capture residual soil nitrogen at the end of the growing season should be seeded as soon as possible after harvest of the raspberry crop. Similar results were found in a study by Bomke and Temple (1994) in the Lower Fraser Valley where spring oats were seeded in the third weeks of August and September, 1993, and harvested in November to determine dry matter yield and N uptake. The dry matter yield 106 was 2.0 and 0.4 t ha"1 for the August and September seeding dates, respectively, while N uptake of those same crops was 42 and 23 kg N ha 1, respectively. The oats cover crop had greater N uptake compared to spring wheat and barley in the two years of study, however for different reasons in each year. While the N concentrations were similar between the three crops in 1992, the higher dry matter yield by the oats resulted in a higher overall N uptake. This may have been due in part to the lower number of seeds per hectare for the spring wheat and barley in 1992. In 1993 the opposite was true since dry matter uptake was similar between the three crops and N concentration was highest in the oats, even though the seeding rate was similar for all three. The later seeded crops did not have sufficient time to develop structurally as indicated by their higher N concentrations and lower dry matter yields. The inorganic N in the less developed late seeded crops could lead to a greater potential for leaching of cover crop N during warm spells in the winter, however measurements would be required to confirm this. The late seeded oats in the study by Bomke and Temple (1994) had accumulated 23 kg N ha"1 in the above-ground portion of the crop by the time of harvest in November, however only 14 kg N ha"1 remained in the soil after the winter. The N content of the early seeded oats in the spring was not available. Another factor which has not been explored in this study is the significance of growth stage in determining how tightly N is bound in the cover crops. Barley, which reached the inflorescence growth stage both years, may be able to retain N more efficiently over the winter than the spring wheat or oats, which only reached the stem elongation stage. Nafuma (pers. comm.) studied the N fractionation of various cover crops at different stages of growth and found that as crops matured a greater portion of N existed in insoluble forms. In another study by Nafuma (pers. comm.), early seeded 107 barley, spring wheat, and oats retained 21, 17 and 17%, respectively, of the N over the winter, although the majority of the N that was lost in the plants was retained in the upper 60 cm of soil. Other factors, such as resistance to disease and nematodes, and cost may have to be taken into account when choosing a cover crop species. Decomposition of cover crop organic residues results in an increase in the numbers of fungi, nematodes, mites and other organisms that are parasitic or predacious on plant parasitic nematodes (Patrick et al. 1965). In the same study, Patrick et al. (1965) found that substances produced during the decomposition of rye residues in soil are directly toxic to some plant parasitic nematodes. 108 5. IMPLICATIONS OF THE RESEARCH Apparent manure N recovery as soil inorganic N was calculated to be approximately 50% within 30 days of manure application in late February to early March when manure was incorporated within four hours of application. Mineralization and nitrification of manure N was rapid at C93 despite the fact that soil temperatures were 4 °C cooler than at C92. The small differences among treatments with varying times to incorporation suggests that most ammonia volatilization occurred within four hours of application. Based on previous research on ammonia volatilization from surface applied manures (Laurer et al. 1976) a more substantial decrease in apparent N recovery with increasing time to manure incorporation would have been expected. Significant manure N recovery may have been delayed until it was incorporated into the soil. In 1992 the soil moisture content at the surface would have been low for mineralization to occur, however incorporation into the 0-10 cm depth interval would have provided more ideal moisture conditions. A comparison of soil inorganic N content over the growing season showed that the inorganic fertilizer applied at a rate of 55 kg N ha'1, had a similar N availability to the poultry layer manure applied at a rate of 100 kg N ha'1, although some differences were observed at C93. In addition, fruiting cluster N uptake was similar between the two treatments and there was no consistent effect on N uptake of the remaining plant components. This is consistent with the 50% apparent manure N recovery within 30 days of manure application. Using manure as the sole source of N resulted in no adverse impact on estimated berry yield, primocane vigour, or crop N uptake, as compared to the inorganic 109 fertilizer treatment. Although there was no significant difference in N uptake of the fruiting cluster, the 55F and 100M treatments did show a significant difference in the stem and whole floricane components at C92 and H92, which perhaps resulted from random variability of the canes. The use of manure as a sole source of N for annual crops is not recommended because it is assumed that the manure N continues to mineralize late into the fall with little or no plant N uptake (BCMAFF 1992). However in this study soil nitrate contents in August or later all three sites were similar in the 55F and 100M treatments indicating that the potential for leaching was no greater in the manure treatments compared to the inorganic fertilizer treatments. Raspberry N uptake by the above-ground portion of the crop was relatively low (90 to 100 kg N ha1), while N removal from the field is only in the form of berries (20 kg N ha"1) since raspberry canes remain in the field where the plant N is recycled. The apparent N recovery of manure and inorganic fertilizer N averaged only 20 and 7%, averaged across treatments, in 1992 and 1993, respectively, providing evidence of a sufficient supply of soil N for the raspberries. Soil nitrate content in the control treatments increased throughout the growing season while raspberry yield and primocane vigour were similar to those where N had been applied, providing evidence of sufficient N mineralization in the control treatment to meet crop N demand. What appears to be an abundant supply of N for raspberries may be due in part to N additions from other sources. Other sources of N to soil include atmospheric deposition, which may be as high as 123 kg ha'1 in the area of the Abbotsford Aquifer (Brisbin 1996), and irrigation water which may account for an additional 20 to 30 kg N ha'1, as discussed previously (Section 2.2.1.1). The history of manure use should be taken into account when considering rates of N application. Soil nitrate in the control treatment at H92 was 51 and 83% higher than 110 at C92 and C93, respectively, in August. This suggests that there may be a higher risk of nitrate leaching from fields where there is a history of high manure application rates and that those fields may require little or no manure or fertilizer N in some years to obtain optimum yield. Due to the rapid manure N availability, poultry manure N can be applied in early to mid April; roughly the same time as fertilizer N would normally be applied. Floricane uptake is highest in May and June therefore manure application in early April would provide sufficient time for mineralization and nitrification to occur. However a standard broadcast application could not be done because the manure could contaminate the raspberry leaves. Early spring leaching of nitrate, as was found in Experiment 1 when manure was applied in late February to early March, could be minimized. Early spring leaching may be a problem particularly for coarse textured soils. Fall application of manure N is not recommended since soil inorganic N is already highest late in the growing season when crop uptake is low. Other studies have shown that essentially all nitrate is leached from the root zone over the winter months in the Lower Fraser Valley because of the high rainfall and low evapotranspiration (Kowalenko 1987; Zebarth et al. 1996). Also, due to rapid mineralization and nitrification with to the warm fall soil temperatures, much of the plant available N would be converted to nitrate rapidly in the fall. Increased soil inorganic N contents in fall may enhance primocane growth late into the fall, and may have an effect of crop dormancy, increasing the risk of winter injury (Seipp 1986). Manure and/or fertilizer N should be placed near the crop row since N uptake appears to be higher there. Less manure N would be required if it was applied in this manner. Farm machinery used to cultivate and harvest in the raspberry inter-row may 111 create a band of compacted soil near the plant impeding the penetration of raspberry roots beyond that zone as was found by Bristow and Brun (1987) in Washington State. Cover crops may be used to capture excess fall soil inorganic N with no adverse effect on the primocanes. The above-ground portion of cover crops can take up nearly 100 kg N ha'1 when seeded by early August, however excess N may remain in the soil even in control treatments in November where a cover crop has been seeded. Nitrogen uptake by the cover crop may reduce the risk of winter injury. There was a linear decrease in N uptake as date of seeding was delayed. Spring oats had consistantly higher N uptakes compared to spring barley and spring wheat. However, spring barley reached a more mature growth stage with more structural N, which may increase the potential to retain the N over the fall and winter period. Cover crop growth was not limited by N availability since there was no response to manure N application in 1992 or 1993. It was shown that the fruiting cluster dry weight may provide an index of relative berry yield by the strong correlation between fruiting cluster dry weight and estimated berry yield using the method of Daubeny et al. (1986). Using the dry weight of the fruiting cluster to estimate relative yield could save time in future studies because it is fast and treatment differences can be detected. One disadvantage is that it does not provide an actual measurement or estimate of yield. The potential for nitrate leaching for the different treatments and history of manure application should also be considered. If the quantity of water which percolates through the root zone can be approximated as the difference between total precipitation and potential evaporation (Zebarth et al. 1995), the average water flow would be estimated as approximately 1000 mm, based on climatic data for the period 1961-1990 (Section 2.1). To prevent an exceedance of the Canadian Drinking Water Guideline of 112 10 mg nitrate-N L 1, no more than 100 kg nitrate-N ha-1 can be allowed to leach beyond the root zone. Soil nitrate-N content in August may be used to provide an estimate of the potential for nitrate leaching. It can be assumed that nitrogen uptake by the primocanes is minimal after August (Kowalenko 1994a). Soil N mineralization after August may compensate for any plant N uptake. Soil nitrate content in August at the site with a history of manure use was 262 kg N ha 1 in the treatment which received 200 kg N ha-1 as poultry layer manure, approximately 20% greater than the site which received no manure in the same year. The application of 200 kg N ha'1 as poultry layer manure resulted in a 44, 31, 31% increase in soil nitrate content compared to the control treatment at G92, C93 and H92, respectively. The risk of nitrate leaching was high even where no manure or fertilizer N was applied. Soil nitrate content in August was 120 and 99 kg N ha'1 for the control treatment at the site with no history of manure application. Thus, excessive nitrate leaching may occur at this site in some years even where no N is applied. In comparison, soil nitrate content in August for the control treatment at the site with a history of manure use was considerably higher, 181 kg N ha'1, indicating that excessive nitrate leaching occurred that year. That excessive nitrate leaching can occur in the absence of N addition highlights the vulnerability of the aquifer to nitrate contamination. Based on studies at Rothamsted in Britain, Addiscott (1988) concluded that mineralization of soil N as a result of soil breaking is the largest contributor of nitrate to groundwater, outweighting that contributed by fertilizers. Schreier (1983) in a land use study over the Abbotsford Aquifer found the most rapid period of conversion of land use to raspberry production to be between 1969 to 1981. A typical transit time for nitrate to reach the groundwater from the soil surface are in the order of 1 to 1.5 years at the 113 Columbia site (Zebarth pers. comm.). Residence time of groundwater in the aquifer were estimated to be as long as 25 years (Anonymous 1995). Thus, some of the nitrae observed in the groundwater today may reflect nitrate released from the conversion of land to raspberry production. However, Wassenaar (1995), using 1 5N and 1 8 0 isotopes, determined that nitrate in the Abbotsford Aquifer is predominantly derived form poultry manure and to a lesser extent from synthetic fertilizers. No distinction was made for nitrate derived from soil N. Current management practices where high rates of N are applied as manure and fertilizer on crops raspberry fields with low N requirements likely result in significant nitrate contamination of the Abbotsford Aquifer. The high soil organic matter over the aquifer may still reflect the recent breaking of the land. However, in time mineralization rates in non-manured sites will decline. 114 6. REFERENCES CITED Addiscott, T. 1988. Farmers, fertilisers and the nitrate flood. New Scientist. Ocotber:50-54. Adriano, P.C., A.C. Chang, and R. Sharpless. 1974. Nitrogen loss from manure as influenced by moisture and temperature. J. Environ. Qual. 3:258-261. Anonymous. 1995. Map of groundwater residence times in the Abbotsford Aquifer. Armstrong, J.E. 1959. Surficial geology of Sumas map-area, British Columbia. Department of Mines and Technical Surveys Canada, Geological Survey of Canada. Armstrong, J.E. 1984. Environmental and engineering applications of the surficial geology of the Fraser Lowland, British Columbia, Geological Survey of Canada, Paper 83-23. Atmospheric Environment Service. 1985. Principal station data: Abbotsford. A publication of the Canadian Climate program. PSD/DSP-98. Baier, W. and G.W. Robertson. 1965. Estimation of latent evaporation from simple weather observations. Can. J. Plant Sci. 45:276-281. Beauchamp. E.G. 1983. Response of corn to nitrogen in preplant and sidedress applications of liquid dairy cattle manure. Can. J. Soil Sci. 63:377-386. Beauchamp, E.G., G.E. Kidd, and G. Thurtell. 1978. Ammonia volatilization from sewage sludge applied in the field. J. Environ. Qual. 7:141-146. Beauchamp, E.G., G.E. Kidd, and G. Thurtell. 1982. Ammonia volatilization from liquid dairy cattle manure in the field. Can. J. Soil Sci. 62:11 -19. Berendse, F., C. Laurijsen and P. Okkerman. 1988. The acidifying effect of ammonia volatilized from farm-manure on forest soils. Ecological Bulletins 39:136-138. Bitzer, C.C. and J.T. Sims. 1988. Estimating the availability of nitrogen in poultry manure through laboratory and field studies. J. Environ. Qual. 17:47-54. Blake, G.R. and K.H. Hartge. 1982. Bulk density. Pages 363-375 in A. Klute, (ed.), Methods of soil analysis. Part 1. Physical and mineralogical methods. 2nd ed. Agronomy No. 9. ASA, Madison, Wl. Bomke, A.A. and W.D. Temple. 1994. Evaluation of overwinter cover crops in an area with intensive waterfowl usage. Presented at the 1994 Annual Meeting of The American Society of Agronomy, Seattle, Washington. 115 Bowman, R.A. and D.D. Focht. 1974. The influence of glucose and nitrate concentrations upon denitrification rates in sandy soils. Soil Biol. Biochem. 6:297-301. Brady, N.C. 1990. The nature and properties of soils. 10th Ed. MacMillan Publishing Company, New York. Brinsfield, R.B, and K.W. Staver. 1991. Use of cereal grain cover crops for reducing groundwater nitrate contamination in the Chesapeake Bay region. Pg. 79-82. in W.L. Hargrove (ed.), The proceedings of an international conference, Jackson, Tennessee, April 9-11,1991. Soil and Water Conservation Society. Brisbin, P.E. 1996. Agricultural nutrient management in the Lower Fraser Valley. Fraser River Action Plan, Environment Canada DOE FRAP 1995-27. Bristow, P.R. and CA. Brun. 1987. Soil compaction: The effects of soil compaction and cover crop on root distribution in "Meeker" red raspberry. Proc. West. Wash. Hort. Assoc. 77: 151-155. British Columbia Ministry of Agriculture, Fisheries and Food. 1992. Environmental guidelines for poultry producers in British Columbia. Soils and Engineering Branch, BCMAFF, Abbotsford, B.C. British Columbia Ministry of Agriculture, Fisheries and Food. 1994. Berry production guide for commercial growers. BCMAFF. Victoria, B.C. Brunke, R., P. Alvo, P. Schuepp, and R. Gordon. 1988. Effect of meteorological parameters on ammonia loss from manure in the field. J. Environ. Qual. 17:431-436. Cassman, K.G. and D.N. Munns. 1980. Nitrogen mineralization as affected by soil moisture, temperature, and depth. Soil Sci. Soc. Am. J. 44:1233-1237. Castellanos, J.Z. and P.F. Pratt. 1981. Mineralization of manure nitrogen - correlation with laboratory indexes. Soil Sci. Soc. Am. J. 45:354-357. Chae, Y.M. and M. A. Tabatabai. 1986. Mineralization of nitrogen in soils amended with organic wastes. J. Environ. Qual. 15:193-198. Chescheir, G.M., III., P.W. Westerman, and L.M. Safley Jr. 1986. Laboratory methods for estimating available nitrogen in manures and sludges. Agricultural Wastes. 18:175-195. Chipperfield, K. 1992a. Raspberry field soil nitrate survey. Sustainable Poultry Farming Group. Chipperfield, K. 1992b. Nutrient content of poultry manure. Pg. 23-25, in J.W. Paul, G. Hughes-Games and B.J. Zebarth (eds.), Proceedings of a Manure Management Workshop, Jan. 30,1992. Abbotsford, B.C. 116 Christensen, J.R. 1947. Root studies: XI. Raspberry root systems. J. Pomoc. Hortic. Sci. 23:218-226. Crandall, P.C. 1980. Twenty years of red raspberry research in southwestern Washington state. Acta Horticulturae. 112:53-58. Crandall, P.C. 1995. Bramble production: The management and marketing of raspberries and blackberries. Food Products Press, New York. Daubney, H.A., A. Dale, and G. McGregor. 1986. Estimating yields of red raspberries in small research plots. HortScience. 21:1216-1217. Dube, P.A. 1981. Climate and soil requirements for economically important crops in Canada. Agriculture Canada, Ottawa, Ontario. 55 pp. Environment Canada. 1985. Principal Station data: Abbotsford. PSD/DSP-98. Environment Canada. 1990. Groundwater-Nature's hidden treasure. Minister of Supply and Services, Canada. Environment Canada. 1992. State of the environment for the Lower Fraser River Basin. SOE Report No. 92-1. Foweraker, J.C. 1994. Introduction. In Groundwater resources of British Columbia. BC Environment and Environment Canada. Gale, P.M., and J.T. Gilmour. 1986. Carbon and nitrogen mineralization kinetics for poultry litter. J. Environ. Qual. 15:423-426. Gardner, W.H. 1982. Water content. Pg. 493-544 in A. Klute, (ed.), Methods of soil analysis. Part 1. Physical and mineralogical methods. 2nd ed. Agronomy No. 9. ASA, Madison, Wl. Gartner Lee Ltd. 1993. Fraser Valley ground water monitoring program: Phase 1 report. Prepared for B.C. Ministry of Health. Gee, G.W. and J.W. Bauder. 1982. Particle-size analysis. Pg. 383-411 in A. Klute, (ed.), Methods of soil analysis. Part 1. Physical and mineralogical methods. 2nd ed. Agronomy No. 9. ASA, Madison, Wl. Goldin, A. 1987. Reassessing the use of loss-on-ignition for estimating organic matter content in non-calcareous soils. Commun. Soil Sci. Plant Anal. 18:1111 -1116. Goode, J.E. and K.J. Hyrycz. 1968. The response of Mailing Jewel and Mailing Exploit raspberries to different soil moisture conditions and straw mulching. J. Hort. Sci. 43:215-230. 117 Greenwood, D.J., K. Kubo, and I.G. Burns. 1989. Apparent recovery of fertilizer N by vegetable crops. Soil Sci. Plant Nutr. 35:367-381. Hadas, A., B. Bar-Yosef, S. Davidov, and M. Sofer, 1983. Effect of pelleting, temperature, and soil type on mineral nitrogen release from poultry and dairy manures. Soil Sci. Am. J. 47:1129-1133. Halstead, E.C. 1986. Ground water supply-Fraser Lowland, British Columbia. National Hydrology Research Institute Paper No. 26., Environment Canada, Saskatoon, Saskatchewan. Hill, R.G. 1958. Fruit development of the red raspberry and its relation to nitrogen treatment. Ohio Agricultural Experiment Station. Wooster, Ohio. Research Bulletin 803. Hoff, J.D., D.W. Nelson, and A.L. Sutton. 1981. Ammonia volatilization from liquid swine manure applied to cropland. J. Environ. Qual. 10:101-109. Justice, J.K. and R.L. Smith. 1962. Nitrification of ammonium sulfate in a calcareous soil as influenced by combinations of moisture, temperature, and levels of added nitrogen. Soil Sci. Soc. Am. Proc. 26:246-250. Kohut, A.P. 1987. Groundwater supply capability: Abbotsford upland. Province of British Columbia, Ministry of Environment and Parks, Water Management Branch, Victoria, BC. Kohut, A.P., S. Sather, J. Swong, and F. Chwojka. 1989. Nitrate contamination of the Abbotsford Aquifer, British Columbia. Groundwater Section, Water Management Branch, Ministry of Environment, Victoria, British Columbia. Presented at a Symposium on Ground-Water Contamination, June 14-15, Saskatoon, Saskatchewan. Kowalenko, C.G. 1981. Response of raspberries to soil nitrogen and boron applications. Commun. Soil Sci. Plant Anal. 12:1151 -1162. Kowalenko, C.G. 1987. An evaluation of nitrogen use in British Columbia agriculture. Technical Bulletin 1987-3E, Research Branch, Agriculture Canada. Kowalenko, C.G. 1989. The fate of applied nitrogen in a Fraser Valley soil using 1 5N in field microplots. Can. J. Soil Sci. 69:825-833. Kowalenko, C.G. 1993. Nitrogen management in raspberries. Pg. 32-34 in Proc. 35 th Annual LMHIA Short Course, Abbotsford, BC. Kowalenko, C.G. 1994a. Growing season dry matter and macroelement accumulations in Willamette red raspberry and related soil-extractable macroelement measurements. Can. J. Plant Sci. 74:565-571. 118 Kowalenko, C.G. 1994b. Growing season changes in the concentration and distribution of macroelements in Willamette red raspberry plant parts. Can. J. Plant Sci. 74:833-839. Kowalenko, C.G. and D.R. Cameron. 1976. Nitrogen transformations in an incubated soil as affected by combinations of moisture content and temperature and adsorption - fixation of ammonium. J. Soil Sci. 56:63-70. Kowalenko, C.G. and J.W. Hall. 1987. Nitrogen recovery in single- and multiple-harvested direct-seeded broccoli trials. J. Am. Soc. Hortic. Sci. 112:4-8. Kwong, J.C. 1986. Ground water quality monitoring and assessment program. (The occurrence of nitrate-nitrogen in ground water in the Langley-Abbotsford area). B.C. Ministry of Environment, Water Management Branch, Memorandum Report, File 0329563-A. Laurer, D.A., D.R. Bouldin, and S.D. Klaussner. 1976. Ammonia volatilization from dairy manure spread on the soil surface. J. Environ. Qual. 5:134-141. Lawson, H.M. and P. D. Waister. 1972. The response to nitrogen of a raspberry plantation under contrasting systems of management for weed and sucker control. Hort. Res. 12:43-55. Liebhardt, W.C., C. Golt, and J. Tupin. 1979. Nitrate and ammonium concentrations of groundwater resulting from poultry manure applications. J. Environ. Qual. 8:211 -215. Liebscher, H., B. Hii, and D. MacNaughton. 1992. Nitrates and pesticides in the Abbotsford Aquifer, southwestern British Columbia. Inland Waters Directorate, Environment Canada, North Vancouver, BC. Ljones, B. 1965. Fertilizer effects on raspberry yield. Meld. Norg. Landbrttogsk. 44:1-13. Lockyer, D.R., and B.F. Pain. 1989. Ammonia emissions from cattle, pig and poultry wastes applied to pasture. Environmental Pollution. 56:19-30. Luttmerding, H.A. 1980. Soils of the Langley-Vancouver map area. Vol. 1. Soil map mosaics and legend. B.C. Soil Survey Report No. 15. Luttmerding, H.A. 1981. Soils of the Langley-Vancouver map area. Vol. 3. Description of the soils. B.C. Soil Survey Report No. 15. Maidl, F.X. and G. Fischbeck. 1989. Effects of long-term application of slurry on soil nitrogen mineralization. J. Agron. and Crop Sci. 162:310-319. Martin, L.W., E. Nelson and M.H. Chaplin. 1980. Plant and fruit measurements of 'Meeker' red raspberry related to pruning height and nitrogen application. Acta Hort. 112:157-161. 119 McLean, E.0.1982. Soil pH and lime requirement. Pg. 199-224 in A.L. Page, R.H. Miller and D.R. Keeney (eds.), Methods of soil analysis. Part 2. Chemical and microbiological properties. 2nd ed. Agronomy No. 9. ASA, Madison, Wl. Meisinger, J.J., W.L. Hargrove, R.L. Mikkelsen, J.R. Williams, and V.W. Benson. 1991. Groundwater impacts: Effects of cover crops on groundwater quality. Pg. 57-68 in Cover crops for clean water. W.L. Hargrove (ed.). The proceedings of an international conference West Tennessee Experiment Station April 9-11, 1991, Jackson, Tennessee, Published by the Soil and Water Conservation Society. Molloy, S.P. and H. Tunney. 1983. A laboratory study of ammonia volatilization from cattle and pig slurry. Ir. J. Agric. Res. 2:37-45. Papp, J., I. Kobzos Ppai, and J. Nagy. 1984. Effect of nitrogen application on yield, leaf nutrient status and fruit chemical composition of raspberry and red current varieties. Acta Agronomica Academiae Scieniarum Hungaricae. 33:337-343. Patrick, Z.A., R.M. Sayre, and H.J. Thorpe. 1965. Nematocidal substances selective for plant-parasitic nematodes in extracts of decomposing rye. Phytopathology. 55:702-704. Paul, J.W. and E.G. Beauchamp. 1989. Relationship between volatile fatty acids, total ammonia, and pH in manure slurries. Biological Wastes. 29:313-318. Pehoto, L.T. 1968. Response of raspberries to site and soil. (Russian). Vestn. sel'-hoz. Nauki. 13:44-47 (English abstract). Pratt, P.F., S. Davis, and R.G. Sharpless. 1976. A four-year field trial with animal manures II. Mineralization of nitrogen. Hilgardia. 44:113-125. Pratt, P.F., F.E. Broadbent, and J.P. Martin. 1973. Using organic wastes as nitrogen fertilizers. Calif. Agric. 27:10-13. Province of British Columbia. 1930. Air photo mosaic of Lower Mainland, B.C. Province of British Columbia. 1969. Air photo # BC5317-043 (1:12000). Province of British Columbia 1974. Air photo # BC C93 0078 (1:7000). Province of British Columbia. 1977. Air photo # BC 5765 130 (1:12000). Reddy, K.R. 1982. Mineralization of nitrogen in organic soils. Soil Sci. Soc. Am. J. 46:561-566. Robertson, R.N. 1979. Evaluation of nitrate in ground water in the Delaware coastal plain. Ground Water. 17:328-337. Rolston, D.E., D.L. Hoffman, and D.W. Toy. 1978. Field measurements of denitrification: I. Flux of N 2 and N 20. Soil Sci. Soc. Am. J. 42:863-869. 120 Sanderson, K.A. and J.A. Cutcliffe. 1988. Effect of inter-row soil management on growth and yield of red raspberry. Can. J. Plant Sci. 68:283-285. Schreier, H. 1983. Soil survey data for land use planning: A case study of raspberry cultivation in British Columbia. J. Soil Water Conserv. 38:499-503. Schuman, G.E., T.M. McCalla, K.E. Sazton, and H.T. Knox. 1975. Nitrate movement and its distribution in the soil profile of differentially fertilized corn watersheds. Soil Sci. Soc. Amer. Proc. 39:1192-1197. Shuval, H.I. and N. Gruener. 1977. Infant methemoglobinemia and other health effects of nitrates in drinking water. Prog. Wat. Tech. 8:183-193. Seipp, D. 1986. Effect of nitrogen supply on growth and yield of raspberry. (German). Einfluss der Stickstoffversorgung auf Wachstum und Ertrag der Himbeere. Erwerbsobstbau. 28:98-101 (English abstract). Sexstone, A.J., T.B. Parkin, and J.M. Tiedje. 1985. Temporal response of soil denitrification rates to rainfall and irrigation. Soil Sci. Soc. Am. J. 49:99-103. Sims, J.T. 1986. Nitrogen transformations in a poultry manure amended soil: Temperature and moisture effects. J. Environ. Qual. 15:59-63. Sims, J.T. 1987. Agronomic evaluation of poultry manure as a nitrogen source for conventional and no-tillage corn. Agron. J. 79:563-570. Spallacci, P. 1981. Nitrogen losses by leaching on different soils manured with pig slurry. Pg. 284-288 in J.C. Brogan (ed.), Nitrogen Losses and Surface Run-off. ECSC, EED, EAEC, Brussels-Luxembourg. Stanford, G. and S.J. Smith. 1972. Nitrogen mineralization potentials of soils. Soil Sci. Soc. Amer. Proc. 36:465-472. Stanford, G., M.H. Frere, and D.H. Schwaninger. 1973. Temperature coefficient of soil nitrogen mineralization. Science. 115:321-323. Stanford, G., J.N. Carter, D.T. Westermann, and J.J. Meisinger. 1977. Residual nitrate and mineralizable soil nitrogen in relation to nitrogen uptake by irrigated sugarbeets. Agronomy J. 69:303-308. Stevenson, F.J. 1982. Origin and distribution of nitrogen in soil. Pg. 1-39 in Nitrogen in agricultural soils. Agronomy No. 22. Szeto, S.Y., G. Grove, H. Liebscher, B. Hii, and B.J. Zebarth. 1994. Nonpoint-source groundwater contamination by 1,2,2-trichloropropane, a trace impurity in soil fumigant formulations. J. Environ. Qual. 23:1367-1370. 121 Tottman, D.R., R.J. Makepeace, and H. Broad. 1979. An explanation of the decimal code for the growth stages of cereals, with illustrations. Annals of Applied Biology. 93:221-234. Van der Molen, J., H.G. Van Faassen, M.Y. Le Clerc, R. Vriesema, and W.J. Chardon. 1990. Ammonia volatilization from arable land after application of cattle slurry. I Field estimates. Netherlands J. of Agric. Sci. 38:145-158. Vet, R.J., N.W. Reid, J.E. Hunt. 1983. Assessment of the potential incremental acid deposition in the Greater Vancouver area from the Burrard Thermal Generating Plant. CSC 110.313-1. Prepared for British Columbia Hydro and Power Authority by Concord Scientific Corporation. Vetter, H., and G. Steffens. 1981. Leaching of nitrogen after spreading of slurry. Pg. 251-269 in J.C. Brogan (ed.), Nitrogen losses and surface run-off. ECSC, EEC, EAEC, Burssels-Luxembourg. Voroncihina, Z.N. 1967. The root system of raspberries. (Russian). Dolkl. mosk. sel'.-hoz. Akad. K.A. Timirjazeva. 132:55-59 (English abstract). Wassenaar, L.I. 1995. Evaluation of the origin and fate of nitrate in the Abbotsford Aquifer using the isotopes of 1 5N and 1 8 0 in N0 3 \ Applied Geochemistry. 10:391-405. Welte, E. and F. Timmerman. 1987. Effects and sources of nitrate pollution and possibilities for emission reduction. Pg. 27-41 in E. Welte and I. Szabolcs (eds.), Protection of water quality from harmful emissions with special regard to nitrate and heavy metals. 5 t h International CIEC Symposium, Balatonfured, Hungary. Westerman, P.W., L.M. Safley, J.C. Barker, Jr. 1988. Available nitrogen in broiler and turkey litter. Trans, of the ASAE. 31:1070-1075. Wright, C.J. and P.D. Waister. 1980. Seasonal changes in the mineral nutrient content of the raspberry. Acta Hortic. 112:295-304. Wood, C.A., M.M. Anderson, and A.M. Smith. 1962. Quantities and composition of crop materials removed from an established raspberry plantation. Hortic. Res. 1:85-94. Zadoks, J .C, T.T. Chang, and C F . Konzak. 1974. A decimal code for growth stages of cereals. Weed Research. 14:415-421. Zbeetnoff Consulting. 1991. B.C. Raspberry industry profile. Report prepared for the B.C. Raspberry Industry Advisory Committee under the B.C/Canada Agi-food Regional Development Subsidiary Agreement. Zebarth, B.J. and J.W. Paul. 1995. Impact of changes in agricultural land use on nitrogen loading to the Abbotsford Aquifer. Pacific Agricultural Research Centre (Agassiz), Technical Report No. 119. 122 Zebarth, B.J. and J.W. Paul. 1996. Growing season nitrogen dynamics in manured soils in South Coastal British Columbia: Implications for a soil nitrate test for silage corn. Can. J. Soil Sci. (in press) Zebarth, B.J., S. Freyman, and C.G. Kowalenko. 1993. Effect of ground covers and tillage between raspberry rows on selected soil physical and chemical parameters and crop response. Can. J. Soil Sci. 73: 481-488. Zebarth, B.J., P.A. Bowen, and P.M.A. Toivonen. 1995. Influence of nitrogen fertilization on broccoli yield, nitrogen accumulation and apparent fertilizer-nitrogen recovery. Can. J. Plant Sci. 75: 717-725. Zebarth, B.J., J.W. Paul, O. Schmidt, and R. McDougal. 1996. Influence of the time and rate of liquid-manure application on yield and nitrogen utilization of silage corn in south coastal British Columbia. Can. J. Soil Sci. 76:153-164. Zebarth, B.J., D.M. Dean, C.G. Kowalenko, J.W. Paul and K. Chipperfield. 1994. Improved manure nitrogen management in raspberry production. Agassiz Research Station Technical Report No. 96. 123 APPENDIX A: SOIL PROFILES FOR THE COLUMBIA AND HUNTINGDON SITES List of Tables Table Page A.1 Soil profile at the Columbia site 125 A.2 Soil profile at the Huntingdon site 126 Symbols used: SBK -Subangular blocky AB -Angular blocky 124 CD •*—• 'co CO 1Q E ZZ O O CD sz +—> -*—' CO c o o co 0 "O 0 o I— C L O CO < J D -Cl CO CD CO CO o O o E zz o o O CO co ca O CD 4—» o 4—• CO CD CL CO +-* c CD E CD CO o 4S 'o § - 1 - CD co co CD c o co C L CD Q c o N 1 o x CO •o DC > c o CO CO co co JQ co 0 > CO 1 CD CO ZZ CO CD CO CO i CD LO CNJ T3 CD C O 0 CO r t CO DC > LO 0 c CQ CO T 3 0 T 3 • § § 3 O co iz CO > CO 1 cn CO 0 co c tz "6 5 o | •9 .8 £ 8 3£ r f >-LO CM c $ o CO 0 o LO CO I LO CM CQ 0 c CQ < o o E co CO 0 o CO I LO CO Si CQ co -o DC >-o ?2 T - rj-0 C CQ CO ZZ • co 0 CO I— CO o o E CO CL ZZ JD CO O C O I CO O T 3 0 •o C ZZ o 0 co zz o o 0 T 3 c 13 O co 0 c o CO T 3 c CO T 3 C CO co 0 cS CO + o CO CM O 125 CD CO c 0 TD c n c 4—» c 3 1 CD CO c O O CO CD TD 0 o i CD-'S CO cJ < 0 J D CO 0 CO CO o O o o I I I x _o o O co co <S O CD O 3 L. 4—• CO CD Q . CO 4 - » c 0 E c n co LL CO TD C 3 O CO co co 0 c 4 - « o c co Q . 0 Q c o .N i_ O X c 5 o . Q CO CO CO cc >-o CQ < TD 0 T3 3 O co i i 0 > CO cn o. 3 co CM c 5 o i_ . Q co 7 3 5; CC >-LO fc E 3 VJ 0 E CQ < TD 0 3 O CO tl 0 > CO cn jo 3 cn 0 CO 3 T3 CO O) CO I CM CQ CO CC > o E 3 TD 0 E CQ < TD 0 3 O co t; 0 > CO _^ c n _co 3 cn 0 co 1 I 1 1 CO 3 TD CO i _ c n •xt CO as CQ T3 0 ® c n •= t= CO co c n T3 0 TD 3 O co fc. CO 0 o o + •x-f O 126 APPENDIX B: CLIMATE NORMALS FOR THE ABBOTSFORD AIRPORT 127 o < CD J O 03 E 0 c CD CO 1 b CQ J D . Q CO o > o Q . • ° I to O ) |3 3 1-3 C 3 l^> CO CO CD C CO 1-3 co fc -xf CM CO CM d o CM o C D 3 "S PEI E EM mu X CO < 2 LO d o CM CD CO •xf LO CO T - CO CD co d cb CM T - i -•xt LO o cb CM d fc T — CO fc d CM d •xf CO CO o fc d CM CO LO fc CO i — cb 00 q CM CO d T — i b o fc •xt cb d -xf LO LO CO •xf • Y — E 3 E c c CO CD o O 00 CD CO CM CO I -•xt CO LO i — •1— CM fc- o 00 CO •xf o CM CM CM fc CM CM CO •r- LO 00 i b CD 00 LO a> -xT •xt d cb LO d CO LO LO CO o CO -xf d 00 d d 00 CD ib LO LO •xf CO CM cb fc c o 'is '5. U C D E E co c 'co CC o d CM O d o d o d E o « = co o c to CD LO LO CM LO •xf CO CM cb fc a> co -xt T— d CM o o T - -xt CO CO r O i CM i - CO I — CD LO cb d •xt T — LO T — CO fc "xt d CM d I-— CO o T — CM E E "co o 5: • UJ T3 C UJ CO CO CM yd 2 co d CO CD fc CO £ 00 co 5 : • > oo co r : co ^ LU LU UJ g o CD CD CO > o CM CM O cb 73 CD 0 _ CO od «> T- O 00 "xt CM O •xf T-" LO CM CO CM CM CD h- CM fc 1^  CD d CO -xt CD fc cb ib fc • * CO o ib LO •xt LO CM C D c !E tn c 3 C O CD d CM CD cb o CM C35 CO CD LO d LO O CD fc -xf T - "Xt CM -xt •xt O d i - o T- CO CD 00 cb CD fc CM fc O fc ib CD CM CO CD C 1c co c 3 CO CQ .Q CO CO O Q . 128 APPENDIX C: ZADOKS CODES 129 Table C.1. The ten principal growth stages and secondary growth stages of Zadoks Codes. Code Growth stage 0 Germination 00 Dry seed 01 Start of imbibition (water absorption) 02 — ' 03 Imbibition complete 04 — 05 Radicle (root) emerged from caryopsis (seed) 06 — 07 Coleoptile (shoot) emerged from cayopsis 08 — 09 Leaf just at coleoptile tip 1 Seedling growth 10 First leaf through coleoptile 11 First leak unfolded 12 2 leaves unfolded 13 3 leaves unfolded 14 4 leaves unfolded 15 5 leaves unfolded 16 6 leaves unfolded 17 7 leaves unfolded 18 8 leaves unfolded 19 9 or more leaves unfolded 2 Tillering 20 Main shoot only 21 Main shoot and 1 tiller 22 Main shoot and 2 tillers 23 Main shoot and 3 tillers 24 Main shoot and 4 tillers 25 Main shoot and 5 tillers 26 Main shoot and 6 tillers 27 Main shoot and 7 tillers 28 Main shoot and 8 tillers 29 Main shoot and 9 or more tillers cont... 130 Table C.1 (continued) Code Growth stage 3 Stem elongation 30 Pseudostem (leaf sheath) erection 31 First node detectable 32 2nd node detectable 33 3rd node detectable 34 4th node detectable 35 5th node detectable 36 6th node detectable 37 Flag leaf just visible 38 — 39 Flag leaf ligule just visible 4 Booting 40 — 41 Flag leaf sheath extending 42 — 43 Boots just visibly swollen 46 — 45 Boots swollen 46 — 47 Flag leaf sheath opening 48 — 49 First awns visible 5 Inflorescence (ear/panicle) emergence 50 — 51 First spikelet of inflorescence just visible 52 — 53 1/4 of inflorescence emerged 54 — 55 1/2 of inflorescence emerged 56 — 57 3/4 of inflorescence emerged 58 — 59 Emergence of inflorescence completed cont... 131 Table C.1 (continued) Code Growth stage 6 Anthesis (flowering) 60 — 61 Beginning of anthesis 62 — 63 — 64 — 65 Anthesis half-way 66 • — 67 — 68 — 69 Anthesis complete 79 — 80 Dough development 80 — 81 — 82 — 83 Early dough 84 — 85 Soft dough 86 — 87 Hard dough 88 — 89 — 9 Ripening 90 — 91 Caryopsis hard (difficult to divide) 92 Caryopsis hard (not dented by thumbnail) 93 Caryopsis loosening in daytime 94 Over-ripe, straw dead and collapsing 95 Seed dormant 96 Viable seed giving 50% germination 97 Seed not dormant 98 Secondary dormancy induced 99 Secondary dormancy lost 132 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099144/manifest

Comment

Related Items