Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental culture of duckweed (Lemnaceae) for treatment of domestic sewage Whitehead, Alan Joseph 1987

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

Item Metadata

Download

Media
831-UBC_1987_A6_7 W45.pdf [ 6.91MB ]
Metadata
JSON: 831-1.0058829.json
JSON-LD: 831-1.0058829-ld.json
RDF/XML (Pretty): 831-1.0058829-rdf.xml
RDF/JSON: 831-1.0058829-rdf.json
Turtle: 831-1.0058829-turtle.txt
N-Triples: 831-1.0058829-rdf-ntriples.txt
Original Record: 831-1.0058829-source.json
Full Text
831-1.0058829-fulltext.txt
Citation
831-1.0058829.ris

Full Text

EXPERIMENTAL CULTURE OF DUCKWEED (LEMNACEAE) FOR TREATMENT OF DOMESTIC SEWAGE by ALAN JOSEPH WHITEHEAD B . S c . , U n i v e r s i t y of V i c t o r i a , 1975 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department o f B i o - R e s o u r c e E n g i n e e r i n g We accept t h i s t h e s i s as conforming to the r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA J u l y , 1987 © Al an Joseph Whitehead In presenting this thesis in partial fulfilment of the requirements for an 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. Department of Bio-Resource E n g i n e e r i n g The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 15 October, 1987 Abstract The culture of the floating aquatic plant, duckweed (Lemna minor), as an agent of domestic sewage treatment was studied in a clarification lagoon at Duncan, British Columbia, during the summer of 1986. Duckweed was grown in plastic fabric tanks (3700 L volume, 1.85 m deep, 2.25 m 2 water surface area) receiving 290 L of sewage per day or 12.8 d hydraulic retention time. Three treatments were tested: cropped duckweed, uncropped duckweed, and no duckweed. Water quality, plant growth and tissue composition were monitored on the basis of weekly sampling. Removals of VSS, COD, total-N and total-P were greater in the presence than in the absence of duckweed. Unmeasured imports of N and P masked the effect of plant uptake on reducing nutrient concentrations in the tank effluents. Sustainable duckweed yields were possible at both cropping rates, despite a severe infestation of aphids. Dry matter yields of 2.0 g/m2.d and 6.4 g/m2.d were obtained at the 15Z/week and 50X/week cropping rates, respectively. Duckweed contained 6.1 - 6.42 N and 1.1 - 1.4X P (dry wt.). Plant harvest removed 0.14 g N/m2.d and 0.03 g P/m2.d at the 157./week and 0.31 g N/m2.d and 0.07 g P/m2.d at the 50X/week cropping rates. Cropping increased the fraction of total-N and total-P loading that could be removed via plant uptake. Performance of the experimental treatments is analyzed in the light of concentration data, mass balances, and mass flux estimations. Possible sources of unmeasured N and P imports are discussed, and recommendations for future research are provided. The results suggest that duckweed may hold promise under certain conditions as a means of polishing sewage lagoon effluent. i i i TABLE OF OONTENTS page ABSTRACT i i LIST OF TABLES i i i LIST OF FIGURES iv ACKNOWLEDGEMENT V 1. INTRODUCTION 1 1.1 Background 1 1.2 Descr i pt i on of Duckweed 2 1.3 Object i ves 4 2. LITERATURE REVIEW 5 2.1 Introduction 5 2.2 Overview of the Lemnaceae 6 2.2.1 Basic description 6 2.2.3 Life history 6 2.3 Ecology of duckweeds 9 2.3.1 Physical ecology 10 2.3.2 Chemical ecology 13 2.3.3 Biological ecology 20 2.4 Productivity 23 2.4.1 Mineral nutrition 23 2.4.2 Growth rate 23 2.4.3 Yield 25 2.5 Composition and uses 27 2.5.1 Composition 27 2.5.2 Use in animal feeds 29 2.5.3 Use as human food 32 2.5.4 Use in energy production 32 2.5.5 Other uses 33 2.6 Duckweed Use in Water Treatment 33 2.6.1 Suspended solids removal 34 2.6.2 Oxygen demand 36 2.6.3 Hydrogen ion concentration (pH) 37 2.6.4 Nitrogen 38 2.6.5 Phosphorus 42 2.6.6 Disinfection 45 2.7 Design and Operation of Treatment Systems Using Duckweed . . . . 46 2.7.1 Previously proposed designs . 48 2.7.2 Harvesting of duckweed . 49 2.8 Research Needs 50 i v 3. MATERIALS AND METHODS 52 3.1 Experimental Set-up 52 3.1.1 Experimental design 52 3.1.2 Apparatus 52 3.1.3 Duckweed cropping 56 3.2 Sampling and Analysis 58 3.2.1 Water quality 58 3.2.2 Plant yield 59 3.2.3 Plant nutrient content 60 3.3 Estimation of Treatment Efficiency 60 3.3.1 Mass balances 60 3.3.2 Mass flux rates 62 4. RESULTS AND DISCUSSION 63 4.1 General Observations 63 4.2 Duckweed Growth 64 4.2.1 Standing crop 64 4.2.2 Duckweed composition 66 4.2.3 Duckweed productivity 68 4.3 Water Quality 71 4.3.1 Temperature 71 4.3.2 Dissolved Oxygen 71 4.3.3 Hydrogen ion concentration (pH) 73 4.3.4 Suspended solids 79 4.3.5 Chemical oxygen demand 82 4.3.6 Nitrogen 85 4.3.7 Phosphorus 94 4.4 Nutrient Mass Balances 102 4.4.1 Nitrogen 102 4.4.2 Phosphorus 107 4.5 Mass Flux 112 4.5.1 Vol at i le sol ids 113 4.5.2 COD mass flux 113 4.5.3 Total nitrogen 114 4.5.4 Anmnonia mass f lux 115 4.5.5 Nitrate mass flux 117 4.5.6 Organic nitrogen mass flux 117 4.5.7 Nitrogen removal by duckweed 118 4.5.8 Total-phosphorus mass flux 120 4.5.9 Dissolved orthophosphate mass flux 121 4.5.10 Phosphorus removal by duckweed 122 4.6 Treatment Efficiencies 122 4.6.1 Suspended solids 122 4.6.2 Chemical oxygen demand 124 4.6.3 Nitrogen 126 4.6.4 Phosophorus 135 4.7 Sources of Unmeasured Nutrient Inputs 139 5. SUMMARY AND CONCLUSIONS 143 6. RECOMMENDATIONS 7. L I T E R A T U R E C I T E D v i LIST OF TABLES Table page 1 Summary of environmental requirements of duckweeds. 11 2 Summary ofavailable growth data for duckweed and two grain crops. 26 3 Summary of reported duckweed standing crop and yield data. 28 4 Composition of selected duckweed genera. 30 5 Amino acid concentrations reported in duckweed protein. 31 6 Removal of suspended sol ids reported for wastewater treatment systems containing duckweed. 35 7 Removal of nitrogen reported for wastewater treatment systems containing duckweed. 39 8 Removal of phosphorus reported for wastewater treatment systems containing duckweed. 44 9 Description of domestic sewage treatment systems reported or proposed to contain duckweed. 47 10 Summary description of the experimental design. 54 11 Composition of duckweed grown in the experimental enclosures during Periods 1 and 2. 67 12 Dry matter production rates of harvested duckweed (Lemna minor) growing on domestic sewage in experimental enclosures, 23 June to 15 September, 1986, Duncan, British Columbia. 70 13 Comparison of dry matter yields of duckweed Lemna minor and other agricultural crops. 72 14 Record of pH in the common influent and in the effluents from the experimental enclosures during Period 2. 78 15 Concentrations of vo1 a t i l e suspended solids (VSS) in the common influent and in the effluents from the experimental enclosures. 80 v i i Table T i t l e page 16 Concentrations of chemical oxygen demand (COD) in the common influent and in the effluents from the experimental enclosures. 83 17 Concentrations of total nitrogen in the common influent and in the effluents from the experimental enclosures. 86 18 Concentrations of total ammonia nitrogen in the common influent and in the effluents from the experimental enclosures. 89 19 Concentrations of n i t r i t e and nitrate nitrogen in the camion influent and in the effluents from the experimental enclosures. 92 20 Concentrations of organic nitrogen in the common influent and effluents from the experimental enclosures. 95 21 Concentrations of total phosphorus in the conmon influent and in the effluents from the experimental enclosures. 98 22 Concentrations of dissolved orthophosphate phosphorus in the common influent and in the effluents from the experimental enclosures. 99 23 Nitrogen mass balance and removal efficiencies in experimental enclosures situated in a municipal sewage lagoon, Period 1: 2 June to 4 August, 1986. 103 24 Nitrogen mass balance and removal efficiencies in experimental enclosures situated in a municipal sewage lagoon, Period 2: August to 15 September, 1986. 106 25 Phosphorus mass balance and removal efficiencies in experimental enclosures situated in a municipal sewage lagoon, Period 1: 2 June to 4 August, 1986. 108 26 Phosphorus mass balance and removal efficiencies in experimental enclosures situated in a municipal sewage lagoon, Period 2: 4 August to 15 September, 1986. 111 27 Mass flux rates of VSS, COD, N and P in the common influent to and effluents from the 2.25 m2 experimental enclosures. 116 28 Mass flux rates of nitrogen and phosphorus in the influent, effluent and duckweed harvest. 119 v i i i LIST OF FIGURES F i gure page 1 External morphology of the duckweeds Lemna minor and Spirodela polyrhhiza. 7 2 The 1ife cycle of duckweeds. 8 3 Overview of the Duncan sewage treatment lagoons. 53 4 Diagram of the experimental enclosure or "Iimno-corral" used in the duckweed culture experiments. 55 5 Schematic diagram of the experimental set-up used in the duckweed culture experiments. 57 6 Changes in the standing crop of duckweed in the cropped treatment (average of tanks A and B) and i n the uncropped treatment (tank C). 65 7 Cumulative duckweed harvest from the cropped treatment. 69 8 Prof i1es of d i sso1ved oxygen concentrat i on measured i n the experimental units containing cropped duckweed, on three occasions during the summer of 1986. 74 9 Prof i1es of d i ssoIved oxygen concentrat i on measured i n the experimental units containing uncropped duckweed, on three occasions during the summer of 1986. 75 10 Profiles of dissolved oxygen concentration measured in the experimental units containing no duckweed (control), on three occasions during the summer of 1986. 76 11 Profiles of dissolved oxygen concentration measured in the host lagoon, (Cell 4) at Duncan, B.C., on three occasions during the summer of 1986. 77 12 Changes in the concentration of volatile suspended solids (VSS) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 81 13 Changes in the concentration of chemical oxygen demand (COD) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 84 i x F i gure T i t l e page 14 Changes in the concentration of total nitrogen (total-N) in the cannon influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 87 15 Changes in the concentration of ammonia nitrogen (NH3-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 90 16 Changes in the concentration of n i t r i t e plus nitrate nitrogen (NO3+NO3-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 93 17 Changes in the concentration of organic nitrogen (organic-N) in the corrmon inf 1 uent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 96 18 Changes in the concentration of total phosphorus (total-P) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 100 19 Changes in the concentration of dissolved orthophosphate (ortho-P) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed, and control treatments. 101 20 Weekly changes in the mass removal efficiency of volatile suspended solids (VSS) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 123 21 Weekly changes in the mass removal efficiency of chemical oxygen demand (COD) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 125 22 Weekly changes in the mass removal efficiency of total nitrogen (total-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 127 23 Weekly changes in the mass removal efficiency of total ammonia nitrogen (NH3-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 129 X Figure T i t l e page 24 Weekly changes in the mass removal efficiency of n i t r i t e plus nitrate nitrogen in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 132 25 Weekly changes in the mass removal efficiency of organic nitrogen in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 134 26 Weekly changes in the mass removal efficiency of total phosphorus in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 136 27 Weekly changes in the mass removal efficiency of dissolved orthophosphate phosphorus in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. 138 xi ACKNOWLEDGEMENTS A special thank-you is in order to D.G. and Lily for their support and patience, as well as to the following persons for their assistance: Alan Patola, Adeline Chen, Al Hudson, Professor Jim Atwater, Tom Tevendale, Bob MacDonald, Jet Blake, Susan Jaspers, Susan Liptak, Paula Parkinson, Paula Wentzell, Jurgen Pehlke, Neil Jackson and Ping Liao. Completion of the work would not have been possible without the help of many others, to whom I have in person expressed my appreciation. The support - academic, monetary and moral - of the members of my Master's Committee, Professors Victor Lo, Ross Bulley and Richard Branion, is also gratefully acknowledged. Funding was provided by the Science Council of British Columbia. 1 1. INTRODUCTION 1.1 B a c k g r o u n d Lagoon technology is one of the most widely used wastewater treatment methods. Whether used for municipal or industrial wastewaters, however, improved lagoon performance is increasingly being required by regulatory agencies, as water quality standards become more stringent. Communities and industries commonly serviced by lagoons are finding that the options available to improve the quality of their effluents are generally very expensive and require major modifications to the existing treatment system infrastructure. This is particularly the case where the removal of dissolved nutrients (i.e. tertiary treatment) is desired. The concept of utilizing aquatic plants and other biological communities as nutrient removal components of a lagoon-based treatment system has been receiving, in this context, considerable attention in recent years (Tourbier and Pierson, 1976; N.A.S., 1976; E.P.A., 1979). Though limited by ambient temperatures, the approach can be effective in cases where the pollution problem occurs during the warmer summer months, or where wastewater can be stored for summertime treatment. This thesis presents the results of an investigation of the floating waterplant Lemna minor, commonly Known as duckweed, as an agent of tertiary treatment in wastewater lagoons. The sewage treatment complex at Duncan, Vancouver Island, British Columbia, is an example of the need to improve lagoon performance. The treatment system treats mainly residential wastewaters from the City of Duncan and the District of North Cowichan, and is operated by a Joint Utilities Board (J.U.B.) of the two municipalities. A study conducted by the Environmental 2 Protection Service (Derksen, 1981) reported that the Cowichan River, which passes through Duncan, is receiving excessive nitrogen and phosphorus from the Duncan sewage lagoons. The problem is related to insufficient dilution of the effluent due to seasonal low flows in the receiving water. Algal growth develops on the riverbed gravel during the summer, to the potential detriment of critical salmon spawning and rearing habitat, and reaches high densities during August and September, just prior to the salmon spawning season. Under the Federal Fisheries Act, the J.U.B. is required to take action to mitigate the environmental impact of its effluent discharge. A preliminary study of conventional nutrient removal methods revealed that the least expensive alternative available, physical-chemical precipitation with liquid alum, would cost $70,000 per annum, (Duncan Associates, 1983; Hudson, pers. comm.). In an effort to find a more cost effective option, and in the light of the aforementioned reports, the potential of taking advantage of the natural duckweed population present on the lagoons was given serious consideration (Tevendale, pers. comm.). 1.2 Description of Duckweed The following paragraphs provide a brief introduction to duckweed culture as a potential wastewater treatment biotechnology. Duckweed is a small floating aquatic plant, or macrophyte, (Fig. 1) that has gained worldwide recognition for its high rates of growth and nutrient uptake, elevated protein content, and ease of harvest due to its free-floating form (N.A.S., 1976; Hillman and Culley, 1978). The plant, often considered to be a minor weed, frequently occurs naturally in sewage ponds. Reports from the southern United 3 States first pointed out that the tolerance of certain duckweed species to cold temperatures, including occasional frost, might make them suitable for use at higher latitudes (Culley and Epps, 1973; O'Brien, 1980). Also, the potential economic value of the harvested plants, for use as a high-protein animal feed (Hillman and Culley, 1978; Culley et al., 1981), soil conditioner and fertilizer (So, 1986) and other uses (Porath et ah, 1985), could reportedly help to defray in part the costs of pollution control There are currently however no guidelines for the design and operation of duckweed culture as a component of a tertiary treatment system. While the need for regular harvesting of the plant biomass is stressed in most of the literature, data on optimum cropping rates and frequencies is lacking. Similarly, though the relationship between nutrient availability and plant growth rate is well documented (Culley and Epps, 1973; Culley et al., 1981), information on the interaction between nutrient loading rate, hydraulic retention time and duckweed cropping rate is necessary in order to optimize nutrient removal efficiency. In addition, as with most "self-designed" biological systems (Odum, 1971), site specificity also appears to play an important role, and it therefore becomes necessary for preliminary research to be carried out in the environs of the eventual application. 1.3 Objectives Based on the information gaps identified above, a pilot-scale study was developed with the general aim of testing the concept of duckweed culture as a method of tertiary treatment under field conditions in British Columbia. Specific objectives were to: 1. compare the removal of suspended solids, oxygen demand, nitrogen and phosphorus from sewage stabilization pond water in the presence and absence of duckweed; 2. quantify the contribution of duckweed to the overall nitrogen and phosphorus removal from the water; and 3. quantify the productivity and potential yield of duckweed biomass and crude protein under experimental field conditions. 5 2. LITERATURE REVIEW 2.1 I n t r o d u c t i o n There is a vast amount of information published on the duckweed family. Scientific accounts date from at least 1868 when Hagelmeier presented Die Lemnaceen - eine monografische Untersuchung ("The Lemnaceae - a monographical study") (Hillman, 1961). The scope of this thesis necessarily limits the coverage primarily to publications dealing with the subject of wastewater treatment. The present literature review therefore collects the available information pertaining to the growth, mineral nutrition and utilization of duckweeds in the context of water pollution control. The following general references, however are cited due to the important background information they provide. A very comprehensive review of the descriptive and experimental literature on the Lemnaceae to 1960 is provided by Hillman (1961). Biosystematics and taxonomy are treated in depth by Landolt (1981). A thorough and most useful survey of the literature to 1979 pertaining to environmental requirements and effects on aquatic plants of physical, chemical and biological parameters is presented by Stephenson et al. (1980). The broad subjects of water physics and chemistry are treated in depth in the treatise by Hutchinson (1957), and in the texts by Wetzel (1975), Ruttner (1969), and Cole (1979). The biology and ecology of aquatic plants is detailed in the volumes by Arber (1963), Sculthorpe (1967) and Hutchinson (1975), among others. Useful coverage of the engineering aspects of wastewater treatment is provided in particular by Metcalf and Eddy (1979) 6 and Sundstrom and Klei (1979). And finally, an important review of the literature to 1968 on the management of aquatic plant populations is presented by Little (1968). 2.2 Overview of the Lemnaceae 2.2.1 Basic d e s c r i p t i o n The information presented in this section is based largely on Hillman (1961) unless otherwise specified. As mentioned in Chapter 1, duckweeds are buoyant vascular plants that inhabit the surface of quiescent bodies of fresh water. Though often monospecific, the floating (or less commonly submerged) mat can contain two or more duckweed species, and may be several layers thick. There are about 20 to 30 species of Lemnaceae (the number varies as the scientific nomenclature is revised [Landolt, 1981]), about half of which are primarily tropical or subtropical, with the others occurring in temperate regions. Lemna minor and Spirodela polyrrhiza (which occur in British Columbia) are cosmopolitan. Figure 1 illustrates the external morphology of these two species. The smaller sized watermeals, Wolff ia spp. and Wolfiella spp., are devoid of roots, and are considered to be the smallest of flowering plants. 2.2.2 L i f e h i s t o r y The duckweed life cycle is summarized in Figure 2. Vegetative budding is the predominant mode of reproduction, while seed and turion formation are more important as means of surviving seasonal temperature extremes or drought (Witzum, 1977, in Stephenson et al., 1980). Turions are specialized fronds 7 Figure 1. External morphology of the duckweeds Lemna minor and Spirodela polyrrhiza. 8 Figure 2. The life cycle of duckweeds. Note: turion formation is found in Spirodela polyrrhiza, but not in Lemna minor. 9 which are adapted to survive cold, dessication and consumption by grazing animals. Flowering, when it does occur, takes place in the late spring or early summer, and is induced by 12-15 hour daylength. Although flowers are microscopic, their appearance may be accompanied, in some species, by visibly increased frond asymmetry. Pollination is brought about by wind, water movement, and small insects or other fauna. Fruits, which are also microscopic, may or may not be buoyant, and the seeds are reported to germinate either on or below the water surface. Genetic uniformity is high within any particular population due to the fact that most, if not all, of the fronds are clones. The lifespan of an individual frond can reach 33 days, but will vary, depending on temperature, illumination, water chemistry and genealogy. Reduced frond lifespan is associated with warmer temperatures and bright light (Hodgson, 1970; Porath and Shaul, 1971, 1973). The first daughter frond is generally longer lived and larger than a certain number of successive generations. This progressive senescence is countered by a rejuvenation process whereby subsequent daughter fronds grow larger and live longer than the parent (for a thorough coverage of this much studied phenomenon, see Hillman, 1961:263). 2.3 Ecology of Duckweeds Floating aquatic plants occupy a habitat at the interface between air and water, and are therefore affected by the nature of both environments. Conversely, such plants can also exert an effect on the overlying air or underlying water. The environmental requirements of duckweeds may vary 10 between genera, species and physiological strains. These plants are reported to flourish at sites that are rich in nutrients or organic matter. In ecological terms, the Lemnaceae tend to be prolific early responders to natural or artificial pulse-loading of nutrients into aquatic ecosystems. Table 1 summarizes the physical environmental requirements of duckweeds. 2.3.1 Physical ecology The physical parameters affecting duckweeds include temperature, light (intensity, spectral distribution and photoperiod), wind, waves and currents Each of these factors is examined in further detail below, based on the review of Stephenson et al. (1980) unless otherwise stated. T e m p e r a t u r e The optimal temperature range for duckweed growth is between 20° and 30° C. Different species are adapted to different optima within this range, with Lemna minor being better adapted to temperatures below, and Spirodela polyrrhiza above, 25° C. Duckweeds have been signalled f o r their cold tolerance and their ability to withstand occasional frosts (O'Brien, 1980; Hillman and Culley, 1978; Landolt, 1957, in Stephenson et ah, 1980). However, their capacity to grow and utilize nutrients at low temperatures has not apparently been investigated. The mats absorb solar radiation directly, such that mat temperatures can be up to 12° C warmer than air temperature and, conversely, and up to 11° C warmer than the underlying water (Dale and Gillespie, 1976). The latter authors report that the presence of a duckweed mat also dampens the diel fluctuations in water temperature below the mat. Table 1. Sunnmary of environmental requirements of duckweeds (adapted from Stephenson, et al., 1980). Parameter Optimum range Minimum Maximum Temperature - °C 17.5-30 -2 42 Illumination - lux 4000 - 8000 <10 >50,000 pH 4.5 - 7.5 3.5 8.5 Salinity - mg/L 0.8 - 3.3 ? a 6.6 Nitrogen - mg/L 5 D ? ? Phosphorus - mg/L 0.031 - 0.31 ? ? Potassium - mg/L 2 D ? ? a no information available. D only one value reported. 12 L i g h t The optimum illuminance for duckweed growth ranges from 400 to at least 8000 lux. Light intensities of 2000 - 3000 lux are detrimental to S.polyrrhiza at chronically high temperatures (Stanley and Madewell, 1976 in Stephenson et al., 1980). Lemna is reported to grow in the absence of light (i.e. heterotrophically) provided that the organic compound Kinetin is present (Hillman, 1961). Frond multiplication rate increases with increasing daylength, and the plants are not injured by continuous illumination or unnatural photoperiodic cycles (Hillman, 1961). Critical daylength for flowering varies with species, but lies between 12 to 15 hours (Hillman, 1961). The presence of a duckweed mat causes considerable shading of the underlying water column. Phytoplankton and submerged macrophytes can therefore be eliminated from water bodies through competitive exclusion (Dewante and Stowell, 1981). This effect can be exploited for removal of algal suspended solids associated with effluents from oxidation lagoons (Ehrlich, 1966; Dinges, 1982). Wind and water movement The susceptibility of duckweed mats to be disrupted by winds has been pointed out as a major drawback to their utilization on full scale sewage lagoons. Duckweed mats do not prevent wind-induced wave action as do larger floating species (Dewante and Stowell, 1981). Likewise, the internal cohesiveness of duckweed mats is minimal due to the structural weakness of the stolons connecting individual plants (Hillman and Culley, 1978). 13 Consequently, in exposed sites wave action will disrupt the mats, and wind will drive the plants onto the shore (DeBusK et al., 1981). Duckweeds are only found in flowing water if the roots become entangled on a solid substrate (Glandon and McNabb, 1978). Culley (pers. comm.) reports that duckweed will generally not grow where current is perceptible to the human eye. 2.3.2 Chemical ecology Chemical parameters of importance include salinity, pH, composition of atmospheric and dissolved gases, and the concentrations of dissolved substances. S a l i n i t y The optimum range of seawater salinity for duckweeds is between 0.8 and 3.3 ppt (parts per thousand). Brackish water of up to 6.6 ppt salinity is tolerated by certain species (Stephenson et al., 1980). Seasonal variations in salinity can coincide with recognizable successions of duckweed species dominance (Hillman, 1961). The effect of macrophyte cover on the concentration of dissolved salts is related to evapotranspiration. Theory would suggest that, since the rate of water loss from water bodies covered with duckweed may exceed that from uncovered waters (Whitehead, unpublished data), there would be an accumulation of salts in the plant covered systems; In this context, there appears to be no 1> information specific to duckweeds. Ratios of evapotranspiration to open water evaporation of 0.9 to 1.1 have been measured for Lemna (DeBusk et al., 1983). Hydrogen ion concentration (pH) The optimum pH range for the Lemnaceae is reported to be between 4.5 and 7.5, with tolerance limits of 3.5 to 8.5 (Hillman, 1961). Serious growth reduction is evident above pH 10 and below pH 4.0 (Stephenson et al., 1980). McLay (1976) reports the following lower-tolerated, optimum and upper-tolerated pH levels, respectively, for three duckweed genera: Lemna: 4 -6.2 -10; Spirodela: 3 - 7 - 1 0 ; Wolffia: 4 - 5.0 - 10. The absorption rate for organic growth substances which are utilized by duckweeds is increased in an acidic medium (Blackman et al., 1959; Blackman and Sargent, 1959). Oxygenation of deeper waters by photosynthesizing microalgae or rooted submergent vegetation is much diminished or prevented, and respiratory release of carbon dioxide (as carbonic acid) is increased in waters covered by duckweed (Lewis and Bender, 1961). Anaerobic conditions, which are usually associated with the formation of organic acids (Mclnerney et al., 1979), have been recorded in sewage lagoons under duckweed mats (Culley and Epps, 1973). Atmospheric and dissolved gases Floating and emergent aquatic plants use atmospheric carbon dioxide as their principal carbon source for photosynthesis. Consequently, and in 15 contrast with submerged species including algae, duckweed is not affected by the chemical or biochemical interactions which influence the availability of carbon in the water. Levels of atmospheric COg, over a range of 100 to 9000 ppm (parts per million) have a direct effect on the growth rate, dry matter per frond and sugar content of Lemna minor (Mueller et al., 1972, in Stephenson et al., 1980); (natural atmospheric C 0 2 concentration is about 3500 ppm). Atmospheric sulfur dioxide, at concentrations as low as 0.15 ppm, can cause (reversible) morphological and physiological changes in the same species (Frankhauser et al., 1976, in Stephenson et al., 1980). Exposure of Lemna minor to ozone (0.25 uL/L) for two hours can decrease RNA and protein content in the plants (Craker, 1972). The rate of oxygen transfer through duckweed mats has been measured in the laboratory to be from 4 to 47 percent of the estimated rate for open water (Morris and Barker, 1977). As mentioned earlier, anaerobic conditions may occur below duckweed mats. Odours associated with gases evolved by anaerobic decomposition of organic matter in lagoons are reportedly controlled by duckweed mats (Wolverton, 1979). N i t r o g e n Duckweeds can utilize ammonium, nitrate and organic nitrogen sources. As is the case with all plants, nitrogen is an essential macro-element. Growth and respiration are known to decrease under conditions of nitrogen starvation (Humphrey et al., 1977). The preferential use of one nitrogen species 16 over another is dependent on the specific environmental conditions, such as pH, temperature and illumination. Thus the reports in the literature are varied. Lemna was found to grow best on a nitrate medium, while Spirodela grew equally well with either ammonium or nitrate, utilizing ammonium first if both were available (Hubald and Angsten, 1977, in Stephenson et al., 1980). Similar results were reported by Joy (1969) for L.minor grown in sterile culture, although ammonia and a balanced mixture of amino acids (from hydrolized casein) also supported growth. Porath and Pollock (1982) reported that L.gibba absorbed ammonium preferentially over nitrate, even at molar ratios of ammonium to nitrate as high as 1:1000. The latter authors suggested that this trait may be due to a specific mechanism linked to the enzyme glutamine synthetase, found only in Lemnaceae, and may give the duckweeds a competitive advantage over bacteria in that utilization of nitrogen as ammonium is an anabolic (i.e. synthesizing) process driven by an external (solar) energy source rather than, as in the case of nitrate, a catabolic (i.e. respiratory) process requiring internal biochemical energy reserves. Stanley and Madewell (1975) found that L.minor grew best on diluted swine manure when less than one third of the total nitrogen was in the form of ammonia. Their work did not address the possibly inhibitory effects of certain organic nitrogen sources (amino acids) as had been reported by Joy (1969). The latter author also reported that nitrate reductase activity was suppressed in the presence of elevated ammonium concentrations. Humphrey et al. (1977, in Stephenson et al., 1980) found that L.minor grew well under laboratory conditions on a medium containing 93 mg/L nitrate. 17 Evidence fo r nitrogen fixation by bacteria living within duckweed mats has been presented by Zuberer et al. (1982), who estimated that up to 20 percent of the nitrogen requirements for duckweed growth could be derived from atmospheric N. Oron et al. (1984) measured nitrogen outputs to be 160 V. of inputs to experimental duckweed cultures grown outdoors on sewage, interpreting the results as possible bacterial N fixation within the mats. (The possibility of unmeasured external inputs via birds or other sources was not addressed in the latter study.) Nitrogen species in the growing medium have been found to affect both the quantity and quality of tissue protein. Ferguson and Bollard (1969) reported that amino acid composition of duckweed tissue varied with nitrogen source. Dicht et al. (1976, in Stephenson et al., 1980) reported that total protein content in L.minor increased by 20-30 percent when grown on a nutrient solution containing ammonium as opposed to nitrate. In general, tissue crude protein content has been found to increase with increasing nitrogen concentration in the growing medium (Allenby, 1978; Culley et al., 1978, in Stephenson et al., 1980; Rejmankova, 1978, in Dykyjova, 1978). Tissue N (as TKN or total Kjeldahl N) contents ranging from 1.5 to 6.4 7. of dry weight have been reported by Culley et al. (1981). Protein synthesis in Lemna minor (Trewavas, 1972) and short-day flowering in L.paucicostata (Tanaka and Takimoto, 1978) have been found to be suppressed by low nitrate concentrations. P h o s p h o r u s Duckweed grows optimally within a range of 0.03 to 0.31 mg/L of phosphate phosphorus (Fekete and Reimer, 1973; Fekete et al., 1976). Phosphorus (P) 18 deficiency causes an immediate decrease in Spirodela growth (ceasing after 14 days), and is accompanied by decreased frond size and root elongation (Bieleski, 1968), increased concentration (10 to 20 fold) and redistribution of phosphatase activity to the roots and frond surface (increasing the recycling efficiency of P released by dying plants) (Bieleski and Johnson, 1972), as well as increased starch content (Reid and Bieleski, 1970). The latter phenomenon was presumed to reflect an energetic limitation to catabolism. Lemna is able to utilize phosphorus adsorbed to clay particles (Heally and McColl, 1974, in Stephenson et ah, 1980). Maximum growth rate in Lemna is reportedly associated with total phosphorus concentrations of 0.031 to 0.1 mg/L (Fekete and Reimer, 1973, in Stephenson et al., 1980). The latter study also reported that root length increased with increasing P in the growth medium up to 0.1 mg/L, decreasing at higher concentrations; above 0.31 mg/L there was an increased weight per plant though not in plant numbers. Tissue P content increases with increasing concentration in the growing medium, up to about 2.1 mg P/L (Sutton and Ornes, 1975; Rejmankova, 1978, in Stephenson et al., 1980) or 4 mg/L (Culley et al., 1981). Tissue P content ranging from 0.56 to 2.6 /. of dry weight have been reported by the latter authors. Duckweed growth removes dissolved P, as well as N and other solutes from the water, and may thus compete with other plants for essential nutrients. No literature specific to nutrient-based competition was available. 19 P o t a s s i u m Scant information was found on the potassium requirements of duckweeds. White (1936-1940, in Hillman, 1961) reported that the potassium requirements increased with light intensity, with optima ranging from 2 mg/L at 646 lux to 200 mg/L at 3230 lux. Deficiency of this element was associated with decreasing root length, as well as reduced frond area, multiplication rate and net assimilation. Tissue potassium contents ranging from 2.0 to 4.4 7. of dry weight were recorded by Culley et al. (1981). Other elements Calcium and sulfate (as well as nitrate) deficiencies can significantly decrease protein synthesis in L.minor (Trewavas, 1972). Exposure of L.perpusilla to a range of growth medium sulfate concentrations increased tissue inorganic sulfate content by 200 to 380 percent, though did not cause increases in growth rate or in tissue organic sulfur (Datko et al., 1978ab). Sulfate, among ten different sulfur compounds tested, induced better growth in S.polyrrhiza under laboratory conditions (Fraser, 1974, in Stephenson et al., 1980). Exposure of L.minor to chlorine was found to affect the uptake rate of certain organic compounds (herbicide constituents), increasing the uptake of phenoxy acetic acid derivatives while decreasing that of benzoic acid derivatives (Kenny-Wallace and Blackman, 1972). (This effect is of potential importance where duckweed mats are grown on dechlorination lagoons and utilization of the plant biomass is contemplated.) It is beyond the scope of the present study to review the copious literature available on heavy metals, non metallic elements and organic 20 compounds, despite their potential importance as micronutrients, growth promoters, or toxins. For further information on these topics the reader is referred to Stephenson et al. (1980), Hutchinson and Czyrska (1975) and Stanley and Madewell (1976, in Stephenson et at., 1980). 2 . 3 . 3 Biological ecology This section reviews information available on the relationship between duckweeds and other living organisms, with an emphasis on temperate zone associations. Growth and productivity are discussed separately in section 2.4. Duckweed mats may be composed of more than one species, the relative dominance being a function of season, water chemistry, direct competition or other factors (Hutchinson, 1975). In mixed cultures, f o r example, L.gibba dominates L.minor and S. polyrrhiza (Clatworthy and Harper, 1962). The increased buoyancy of L.gibba fronds allow this species to literally overlap and shade out L.minor (de Lange, 1975). A similar competitive effect can occur when duckweed shades out the leaves of a larger-leafed water lily (Sculthorpe, 1967). Stowell et al. (in press) reports that in mats containing Lemna and Spirodela, the former may tend to be winter dominant while the latter summer dominant. Wolfiella lingulata can grow under a one centimeter thick mat of other Lemnaceae (Hillman, 1961). Similarly, communities of aquatic plants may include, in addition to duckweed, other species such as the water ferns azolla and riccia (Bristow et al., undated; Hillman, 1961), water lilies, grasses, reeds, rushes and others (Hillman, 1961; Arber, 1963; O'Brien et al., 1979; Ozimek, 1985). Rooted macrophyte species often prevent duckweeds from being swept away, by providing a substrate for anchorage or protection from wind (Ozimek, 1985). 21 Epiphytic bacterial densities of 5 million cells per cm 2 have been measured on duckweed fronds (Hossel and Barker, 1976). The green alga E ndoclonium has been commonly found growing in and on Lemna fronds (Prescott, 1978). There is evidence that dissolved organic carbon secreted by the duckweed is utilized by bacterial species living within the mat (Baker and Farr, 1982). The protozoans Vorticella and Euplotes, as well as the flatworm Stenostomum, have been found to predominate in the rhizosphere of a duckweed mat infested with bluegreen algae (Coler and Gunner, 1971). The latter authors have also described the composition of rootzone microbiota under other environmental conditions. The cladoceran water flea, Daphnia, as well as other microinvertebrates including ostracods, copepods and rotifers have been found to thrive in the rootzone of duckweeds, feeding on or inhabiting the surface of the roots and fronds (Ehrlich, 1966; Dinges, 1974). The macrofauna associated with the Lemnaceae includes a diversity of snails, amphibians, reptiles, arthropods, birds and mammals (Hillman, 1961). Snails are common consumers of duckweed and lay their eggs on the frond undersurfaces. Interestingly, turions of S.polyrrhiza are not consumed by the snail Physa (Jacobs, 1947, in Hillman, 1961). Several species of aphids, spiders and insects that inhabit the mats may be functionally important as pollinators. At least two species of insects, the ephydrid fly Lemnaphila scottandae and the rhyncophorous beetle Tansyphyrus lemnae are obligate or facultative associates of Lemna, whereby eggs are laid on the fronds which then serve as food for both larvae and adults. Herbivorous animals including certain species of ducks and geese, fish, muskrat, nutria, turtles and crayfish are known to consume duckweed (Hillman, 1961; Culley et al., 1981). Waterfowl and mammals play an important role in duckweed 22 dispersal, as the fronds can cling to feet, feathers or fur and be transported from one water body to another (Jacobs, 1947, in Hillman, 1961). Traditional consumption of duckweed {Wolffia) by rural peoples in southeast Asia is reported by Bhantumnavin and McGarry (1971). Lemna species are considered pest weeds in rice paddies in Japan (Yui and Koike, 1955, in Hillman, 1961) and vegetable farms in Hong Kong (So, 1986). Duckweeds, as most organisms, can affect or be affected by, other living species through beneficial, neutral or detrimental relationships. Thus, for example, certain plank tonic species such as the water flea Daphnia are favoured while algae such as Chlorella are shaded out (Dinges, 1974; Ehrlich, 1966; Daborn et ah, 1978). Dense duckweed populations can contribute directly to mosquito control by preventing egg laying and/or larval breathing (Angerelli et al., 1974; Furlow and Hayes, 1972, in Stephenson et al., 1980); the plants appear to produce a compound that interferes with larval development. Bacteriostatic effects of L.minor extracts on the bacterium Sphaerotilus natans are described by Stangenberg (1968). Symbiotic associations of duckweeds and nitrogen fixing bacteria have also been described (Zuberer, 1982). Duckweed growth can be inhibited by the compound cyanobacterin which is secreted by a common freshwater bluegreen alga Scytonema hofmanni; the compound can be lethal at sufficiently high concentrations (Bell, 1984). Aphids and moth larvae can virtually eliminate whole duckweed mats under certain conditions (Rakocy and Allison, 1979). Published information on diseases of duckweeds is not available (Stephenson et al., 1980). 2 3 2 .4 Productivity and Composition 2 .4.1 Mineral n u t r i t i o n It is generally agreed that duckweeds absorb dissolved nutrients primarily through the frond surface rather than the roots (Muhonen et al., 1983; Ice and Couch, 1987). The plants can grow in the absence of water, on damp mud, and frond multiplication will continue even when the fronds are upside down (Meijer and Sutton, 1987). Some researchers ascribe to the root no more than a keel-like function in maintaining the plant orientation on the water surface (Arber 1920, in Hillman, 1961). However, the observation that root length and enzyme activity increases under nutrient-poor conditions (as discussed above) would appear to indicate some role in nutrition as well, though interspecific differences may preclude generalization. 2.4.2 Growth r a t e The simplified anatomy of duckweeds, with the virtual absence of supportive tissue (fiber), give duckweeds an advantage in energetic conversion efficiency over most other plants. In the words of Hillman and Culley (1978): "Most of the cells are thus like those of young or maturing leaves, and little photosynthetic energy is devoted to producing or maintaining any other plant structure." The end result is that duckweeds under favourable conditions can grow much faster than other higher plants (see data for maize and amaranth in Section 2.4.3). Duckweeds grow by frond multiplication, and hence, the growth rate under favourable conditions is exponential. Such growth is different however from the growth of bacteria and other single-celled organisms. The reason for this is that, in contrast with unicellular species, the parental duckweeds do not automatically cease to exist with the appearance of the following generation, but continue to produce successive cohorts (Hillman, 1961). Growth rate is affected by plant density, due to competition for dissolved nutrients and sunlight (Clatworthy and Harper, 1962; de Lange, 1975), such that the production of biomass will decrease once the standing crop exceeds a critical value (Joy, 1969; Hillman and Culley, 1978; DeBusk et al., 1981). Said et al. (1979) have suggested, on the basis of results obtained in outdoor experiments, that the optimum plant density will increase with cropping frequency, from 8.64 g/m2 for weekly to 30.3 g/m2 for daily harvesting. Optimum density of 20 g/m2 (dry weight) has been identified experimentally by DeBusk et a/.(1981) and confirmed by Porath et al. (1985) using populations of L.minor and L.gibba, respectively, that were cropped weekly. DeBusk et al. (1981) have reported that yields were not improved by increasing the cropping frequency but, while comparing their results with those of Said et a). (1979), did not attempt to explain the different conclusions. The lack of reports on mathematical modelling of duckweed growth may be due in part to the difficulties of applying Monod or Michaelis-Menten kinetic models (developed for bacterial populations) to duckweed. Nevertheless, growth modelling of Salvinia, a floating plant that exhibits a similar mode of multiplication, has been attempted (Toerien et al., 1983). 25 Growth rates have been reported in terms of doubling time, relative growth rate and crop growth rate (Hillman, 1961; Culley et al., 1980); The doubling time is the time taken for a given mass or frond number to attain twice the starting value. The relative growth rate (RGR) is usually expressed as grams of new production per gram of starting biomass per day (g/g.d), dry weight. Crop growth rate (CGR) expresses the yield of harvested biomass per unit area and time (eg. g/m2.d), on a dry or wet weight basis. Hillman (1961) has warned that growth measurements based on dry weight can be misleading since they may reflect changes in starch content rather than growth, and that (in a laboratory setting) frond number would therefore be a more reliable measure of growth. This advice notwithstanding, the majority of authors have reported growth in terms of dry weight, since mass rather than frond number is the parameter of interest. Table 2 summarizes available duckweed growth information. Evidently, RGRs of up to 0.19 g/g.d have been achieved under unmanaged conditions, while cropping to prevent overcrowding has yielded up to 0.5 g/g.d (Rejmankova, 1975, 1979, in Culley et al., 1981; Said et al., 1979). Also different species growing under identical conditions have exhibited different growth rates (Keddy, 1976; Rejmankova, 1975, in Culley er al., 1981). Endogenous annual cycles in growth have been reported, but refinement of experimental controls have shown these to result from external factors (Hillman, 1961). Table 2. Summary of available growth data for duckweed and two grain crops. Spec i es Growth rate Growing conditions Reference Lemna sp. 0.005 - 0.295 g.g- 1.d _ 1 shallow fishpond Rejmankova, 1978* 0.02 - 0. 19 fishpond, Apr. to Oct. Rejmankova, 1973* 0.061 - 0.098 0.292 unmanaged, outdoor weekly cropping, outdoor Rejmankova, 1975,1979 (in Culley et al., 1981) 0.5 I I manured ponds, daily cropping Said et a l . , 1979 L. mi nor 0.164 fronds.d - 1 ? Keddy, 1976* 0.222 • I Hossell and Baker, undated* L. trisulca 0.024 it Keddy, 1976* L. mi nor 2.2 d doubling time lab. experiment Glandon and McNabb, 1978 2.2 d 4.0 d •i I I it it lab. experiment Wolek, 1974 (in Glandon and McNabb, 1978) Harvey and Fox, 1973 4.5 d • I I I lab. experiment Sutton and Ornes, 1975 Zea sp. (maize) Amaranthus sp. 0.32 0.37 g.g* 1.d - 1 » j 7 Leopold and Kriederman, 1975 (in Hillman and Culley, 1978) * - in Stephenson et al., 1980 ? - conditions not specified. 27 2.4.3 Y i e l d Production data for duckweed have been reported from the United States (Harvey and Fox, 1973; Sutton and Ornes, 1978; Culley et al., 1981), the Soviet Union (Abdulayef, 1969, in Harvey and Fox, 1973), Israel (Porath et al., 1985), Thailand (Bhantumnavin and McGarry, 1971), Czechoslovakia (Rejmankova, 1975, 1978, in Culley et al., 1981), and possibly other countries. Table 3 summarizes the available standing crop and yield data. Standing crops of 70 -1500 kg dry matter per hectare and yields of up to 90,000 kg/ha.yr (dry matter) have been reported. Hillman and Culley (1978) have compared the annual productivity of duckweed and selected crops in the southeastern United States and found that duckweed is capable of annually producing more dry matter and crude protein per unit area than alfalfa, peanuts, soybeans and cottonseed. 2.5 Composition and Uses of Duckweed Interest in the use of duckweed for wastewater treatment and as a potentially usable by-product has resulted in the generation of a considerable amount of information on the mineral and nutritional composition of these plants. Uses in animal feeds and human nutrition, as well as energy (biogas) generation and agriculture, have been investigated. (Section 2.6 examines duckweed use in water treatment in greater detail). 2.5.1 Composition There is general agreement that the composition of duckweed is a function of environmental conditions as much as of genetic differences between and within species. Table 4 summarizes the range of values of duckweed Table 3 . Surrmary of reported duckweed standing crop and yield data (dry matter basis). Species Standing Crop Yield Remarks Reference Letma spp. 3 1 2 . 5 kg.ha-1 154 to 349 kg.ha - 1.d _ 1 laboratory, no mixing Zimpfer, 1982 3 1 2 . 5 3 2 . 5 to 78.1 H surface mixing tf M 3 1 2 . 5 6 6 . 9 to 2 2 6 . 9 sub-surface mixing fl n Lemna sp. 75-1500 M Czechoslovakia, fishpond, May-Aug. Rejmankova, 1978 (in Cul ley et al., 1981) 2 2 . 6 to 2 9 . 3 M Florida, domestic sewage Jan. - Mar. Sutton and Omes, 1975 L. mi nor L.gibba 580 1000 It ft Czechoslovakia, maximum in wiId Rejmankova, 1975 (in Culley et al., 1981) L. mi nor 2 5 2 . 5 M laboratory, domestic wastewater Harvey and Fox, 1973 L. mi nor 200 50 If Florida, outdoor, early average DeBusk et al., 1981 L.gibba 1000 75 ft 1srae1, outdoors Porath et al., 1985 350 " 150 M Israel, outdoors oron et al., 1984 Spirodela polyrrhiza 12 fl Florida, outdoors, Nov. - Dec. Sutton and Ornes, 1977 Mixed duckweed species 17,757 Kg.ha - 1.yr - 1 Louisiana, 1i vestock waste 1agoon Culley et al., 1978 (in Stephenson et al., 1980) (unidentified) 6 7 , 5 0 0 to 9 0 , 0 0 0 M U.S.S.R., reservoi r Abdulayef, 1969 (in Harvey and Fox, 1973) C O 29 constituents reported in the literature (for greater detail see Stephenson et al., 1980). Concentrations of amino acids reported from duckweed protein are presented in Table 5. Fresh duckweed contains between 86 and 97 percent water. The major mineral constituents include carbon (30-44Z), nitrogen (1-7/.), phosphorus (0.2-2.8X), calcium (0.1-10X), potassium (0.03-5.5X) and magnesium (0.2-2.8X), all on a dry matter basis. Many minor elements and metals can be accumulated in duckweed tissue to concentrations many times greater than those in the growing medium (bio-accumulation) or than physiologically required (luxury uptake) (Rodgers et al., 1978; Glandon and McNabb, 1978). Tissue samples that have been washed before drying for subsequent analysis have been found to show lower concentrations of ash, silica, Ca, P, Mg, and Fe than unwashed samples (Burton et al., 1977). 2.5.2 Use in animal feeds Duckweed has been extensively documented as a potential ingredient in animal feeds. Culley et al. (1981) provide a very comprehensive review of the literature to 1980. Positive results have been reported when feeding f resh duckweed to fish, chickens, laying hens, ducklings, swine, sheep and dairy heifers, in amounts ranging from 2.5 to 75 percent of the diet dry weight. Ensilage with agricultural residues has also been studied (Eversull et al., 1981). Feed pellets made from dried duckweed have been produced on an experimental basis (Ngo, pers. comm.). The true digestibility of 30 Table 4. Composition of selected duckweed genera. (After Stephenson et al. , 1980) Component units3 Lemna Spirodela Wolffia HpO X 86 - 97 91 - 95 95 - 96 C X - 30.5 - 44.0 -N X 2.3 - 6.8 1.8 - 7.2 0.9 P X 0.2 - 1.9 0.3 - 2.8 1 K X 1.9 - 4.2 0.03 - 4.2 3.75 - 5.5 Ca X 1.0 - 2.8 0.1 - 9.8 0.5 - 0.8 Mg X 0.3 - 0.4 0.2 - 2.8 0.21 Ash X 12.3 - 24.8 5.8 - 40.0 14.5 - 18.3 Prote i n X 10.6 - 42.6 11.1 - 44.7 13 - 26 Fat X 1.8 - 7.5 2.2 - 9.2 5 - 5.5 Carbohydrate X 22.5 - 35 15 - 75 11-44 Fiber X 9.4 - 11.2 5.7 - 13.5 10.6 - 13.3 Energy MJ.kg-1 10.3 - 13.5 15.4 - 17 -a (dry matter basis, except for HJJO) 31 Table 5. Amino acid concentrations reported in duckweed protein, (grams per 100 grams of protein, dry matter basis) AMINO ACID REFERENCE Culley et al., Burton et al., FAO reference 1981 1977 standard* Aspartic acid 6.92 - -Threon i ne 3.12 4.4 2.8 Ser i ne 2.63 - -Glutamic acid 7.26 - -Proline 2.89 - -Glycine 3.68 5.8 -Alanine 4.40 - -Valine 4.39 5.8 -Methionine 0.90 1.5 2.2 Isoleucine 3.61 4.8 4.2 Leuc i ne 6.68 8.5 4.8 Tyros i ne 2.82 2.9 -Phenylalanine 4.16 4.4 2.8 H i st i d i ne 1.78 - -Lys i ne 4.01 5.7 4.2 Arg i n i ne 4.54 5.3 -Tryptophan - 1.2 1.4 Cystine - 0.6 -* source: Culley et al., 1981. - no value available. 32 duckweed has been found to be about 65 percent (Van Dyke and Sutton, 1977, in Culley et al., 1981). The value of the Lemnaceae as natural food for wildlife has also been pointed out (Culley et al., 1981). 2.5.3 Use as human food Direct references to historical consumption of duckweeds by humans are apparently limited to one report by Bhantumnavin and McGarry (1971), who cited cultivation of Wolffia arrhiza in rain-fed ponds in northern Thailand, Burma and Laos. Extraction of protein concentrate from L.gibba, S.oligorrhiza and S.punctata was assessed by Rusoff et al. (1980). The content of true protein and distribution of amino acids were found to reflect a high nutritional quality. The latter authors suggested that duckweed protein concentrate could effectively complement a human and livestock diet based on grain. 2.5.4 Use in energy production Many authors have suggested that methane generation from aquatic plants might be technically and economically feasible. Laboratory tests using a mixture of Spirode/a and Lemna produced 0.016 m3 CH4/kg dry matter (53.3X biogas methane content) (Wolverton and McDonald, 1979). Methane production by other aquatic plants, including water hyacinth (Eichornia crassipes) and water pennywort (Hydrocotyle umbellatta) were higher. Mixing duckweed at a 1:2 ratio with kudzu (Pueraria lobata) yielded 0.282 m3 CH4/kg dry matter (62.\7. biogas methane content), the highest value among all species combinations investigated. 33 2.5.5 Other uses The value of aquatic plants as soil conditioners and fertilizers has been pointed out by Lumpkin and Plucknett (1981), who have studied the water fern Azolla in particular. Soil water retention and fertility has been significantly improved in soils amended with fresh duckweed (So, 1986). The nutrient content of duckweeds, among other macrophytes, has been assessed with regard to its use in compost (Vavruska, 1966, in Culley et al., 1981). 2.6 Duckweed Use in Water Treatment Reports on the experimental use of duckweeds in water treatment are available from Europe (Switzerland, Czechoslovakia, Poland, Soviet Union), North America (Louisiana, Mississippi, Florida, California, British Columbia), Israel and Thailand. Several excellent reviews are available, namely those of Hillman and Culley (1978), Culley et al. (1981), Stephenson et al. (1980), and Stowell et al. (in press). The water treatment effects of this aquatic plant are due to direct sorption of dissolved substances and to equally significant indirect biotic and abiotic effects, some of which have been mentioned in preceding sections. This section describes in greater detail the information available on suspended solids, oxygen demand, pH, dissolved N and P, metals, organic compounds and pathogens, and discusses the operational aspects of water treatment using duckweed. 3 4 2.6.1 Suspended solids removal Removal of suspended solids, where these are composed primarily of microalgae, is effectively achieved by a duckweed cover. The role of duckweed in controlling algae through light attenuation is well recognized (N.A.S., 1976; Wolverton, 1979; Dinges, 1976, 1978). A large fraction of solids removal in duckweed systems is undoubtedly due to physical sedimentation alone. Table 6 summarizes some of the available suspended solids removal data. (No reports relating to other than domestic sewage were located.) Ehrlich (1966) in Israel found that a duckweed mat growing on secondary sewage effluent provided a protective habitat for Daphnia, an effective consumer of suspended particulate organic matter (including coliform bacteria). The removals (as calculated from the data reported) of algal cells in aquaria stocked with either Daphnia only, Lemna only, a combination of both, or neither species, were 80.7X, 80.6Z, 96.8X and 74.1X, respectively. Concentrations of coliform bacteria were reduced by 89-100/. by Daphnia under a partial cover of Lemna, at hydraulic retention times (HRT) between 8 hours and 10 days. Dinges (1974) corroborated Ehlich's findings and further defined the role of zooplankton in sewage lagoons, eventually proposing a multi-species treatment system design (Dinges, 1976). Wolverton and McDonald (1980) provided data from which removals of total suspended solids (TSS) were calculated. Between May and November, TSS removal efficiencies in a duckweed covered lagoon in Mississippi receiving aerated lagoon effluent (61 to 397 mg TSS/L), ranged from 80.2/. in October to 973V. in May. The mean HRT in the 0.7 ha, 1.5m deep lagoon was 22 days. Dewante and Stowell (1981) compared five macrophyte species (duckweed, water hyacinth, water primrose, cattail, and bulrush), and rated the light Table 6. Removal of suspended solids reported for wastewater treatment systems containing duckweed. Wastewater Location Season Effluent TSS Removal Efficiency Remarks Reference mean (range) mean (range) mg. r' a Y. Domestic sewage secondary effluent Mississippi May - Oct. 13.3 (8-22) 88.8 (95.6-83.3) full scale, Nov.- Apr. 23.0 ( 11 - 33) 49.5 (-20.8-82.5) no cropping Wolverton and McDonald, 1979. Domestic sewage secondary effluent Cal ifomia Oct.- Jan. 3.3 (1 - 6) 71.2 (33.3-88.9) pi lot scale, no cropping Dewante and Stowed, 1981 Domestic sewage, secondary effluent Israel December 5.2° (3.2-7.9) 96.8 (94.8-97.9) 31.5 (6.6-52.6) 81.3 (65.9-91.3) 70.6 (12.6-76.0) 75.0 (51.1-91.3) Outdoor lab., Lemna * Daphnia Lemna only No Lemna Ehrlich, 1966 a (unless otherwise indicated) D (eel Is.rmr3, hundreds) 36 attenuating properties of the former as excellent. TSS (total suspended solids) removal efficiency of 71.2'/ was measured in experimental, duckweed covered trenches (7.5m by 0.75m by 0.60m deep ) receiving unfiltered secondary municipal effluent and operated at HRTs of 1 and 3 days. 2.6.2 Oxygen demand It is thought that, while removal of biochemical or chemical oxygen demand (BOD or COD) does take place in systems containing aquatic plants, the plants do not play more than a physical role by providing a substrate for microbiota and for coalescence of suspended solids (Stowell et al., 1981). There is no information available on the relative contribution of duckweed to removal of soluble or particulate BOD. It is likely that duckweed contribute more to the removal of BOD associated with particulate than dissolved organic matter, for the reasons mentioned in the preceding section. In species having large root systems, the additional surface area afforded by the roots has the effect of increasing the carrying capacity of decomposer organisms and their associated food web (Serfling and Alsten, 1979). In this context, however, the importance of the duckweed root system is probably insignificant. BOD removal efficiencies calculated from data provided by Wolverton and McDonald (1980) for effluent from an aerated sewage lagoon (13-67 mg BOD5 /L) passing through a duckweed covered lagoon between May and November ranged from 69X (May) to 157. (September). The slight difference between average BOD removal efficiencies during the growing season (45.1X - May to October) and the dormant season (43.1X -November to April) would appear to confirm the largely abiological role of duckweed mentioned above. The reported 22d HRT in 37 the duckweed lagoon was based only on the estimated sewage flows. No mention was made of the probable contribution of precipitation to the HRT-related variability in final effluent quality. Dewante and Stowell (1981) in California have provided data showing that B O D 5 removal during autumn was about 33.3X, yielding a range of 5 to 11 mg BOD/L in the effluent from outdoor unharvested duckweed channels treating dilute, unfiltered secondary municipal effluent. The biochemistry of oxidative stabilization (mineralization) of organic matter in water depends on whether the environment is aerobic or anaerobic (Metcalf and Eddy, 1979). Dense mats of duckweed, as discussed previously, inhibit the physical and photosynthetic transfer of oxygen into the water. Wolverton and McDonald (1980) recorded a decrease in dissolved oxygen (DO) concentrations from 3.7-7.3 mg/L in the influent to 0.4-1.7 mg/L in the effluent from a duckweed-covered lagoon (2ld HRT) receiving aerated raw sewage. Aeration of the the duckweed lagoon effluent by a 0.9 m drop at the outlet effectively raised the DO in the final discharge to between 5.0 and 6.0 mg/L. While the potentially adverse environmental effect of hypoxic effluents from duckweed covered lagoons appears to be readily mitigated, the reported hypoxia may be important as well with regard to nitrogen removal through denitrification (see section 2.6.4). 2.6 .3 Hydrogen ion concentration (pH) The general effect of duckweed on water pH has been reviewed in section 2.3.2. The buffering of pH in aquatic plant systems has been reported by Seidel (1976). Wolverton and McDonald (1980) have measured changes in the pH of raw sewage flowing through a system consisting of an aeration cell (36d 38 HRT) followed by a duckweed covered lagoon (21d HRT). Raw sewage with a mean influent pH of 7.4 (range 7.2-7.8) was raised to 7.8 (range 7.5-8.1) in passing through the aeration lagoon and subsequently decreased to 7.2 (7.0-7.5 range) in the effluent from the duckweed lagoon. Additional references specific to wastewater pH control by duckweeds have not been located. 2.6.4 N i t r o g e n Removal of nitrogen from wastewaters has been reported in terms of ammonia or ammonium (no distinction being made between the ionic and molecular forms), nitrite and nitrate (usually expressed as a sum), and TKN. (Reports on uptake of dissolved organic N from wastewaters were unavailable). Table 7 summarizes the data reported by various authors on N removal from wastewater by duckweed. Direct comparison between studies is not possible due to fundamental differences in experimental conditions. Of particular importance is whether or not the experiments were carried out under conditions of static or dynamic water flow, and whether or not the plants were routinely harvested. Outdoor laboratory-scale trials using raw municipal wastewater were conducted in Israel by Oron et al. (1984), during the summer. Influent ammonia-N concentrations in 0.2 m2 mini-pond the summer. Influent ammonia-N concentrations of 50 mg/L were reduced by 59X to 92X in 0.2 m3 mini-ponds operated at an HRT of 20d. Duckweed was cropped three times weekly, removing 50Z of the biomass on each occasion. Dewante and Stowell (1980) in California reported negligible NH4-N removal in unharvested duckweed channels during winter. The wastewater was unfiltered secondary municipal effluent containing 4-16 mg NH4-N/L, and the HRT was either one Table 7. Removal of nitrogen reported for wastewater treatment systems containing duckweed, (mg.l_1) Wastewater and Nitrogen concentration Removal Hydraulic Cropped Observations References location influent effluent V. regime Domestic sewage, secondary effluent, Mississippi secondary effluent, Iaboratory 10.7 - 15.5 3.0 - 5.4 3.5 - 5.0s 7.9 - 9.2° 0.5 - 1.0 3.6 - 6.3 67 - 72 Flowing (54 d HRT) 75 - 89 Static, 21-60 (10 d) No Outdoor, winter No 24° C, indoor Wolverton and McDonald, 1979 Harvey and Fox, 1973 secondary effluent, California Raw sewage, Israel Fishtank water, Israel Coal ash sluicewater, South Carolina Unspecified, Czechoslovakia 4 - 1 6 5 - 1 5 50c 3.8 - 20.5 20 4.2a 25" 1.6b negligible flowing No (1,3 d HRT) 59 - 92 flowing Yes (20 d HRT) 95 74 stat i c NO f I ow i ng No 50 flowing Yes (31 d HRT) Outdoor, Oct.-Jan. pi lot scale Outdoor lab., summer Outdoor lab., partial shade Natural swamp, annual mean values Dewante and StowelI, 1981 Oron et al., 1985 Porath and Pollock, 1982 Rodgers et aI., 1978 Full scale, summer Rejmankova, 1979 (in Culley et al., 1981) a as TKN D as nitrate-N c as arrmonia-N (all other values as total nitrogen) L O ^0 or three days. The different results between the two studies, though due in part to season, were also probably due to the absence of cropping and the relatively high ammonium mass loading to the latter system. Wintertime reductions in TKN concentrations of 67-72/. have been reported by Wolverton and McDonald (1979) in southern Mississippi. Although no cropping was practiced, the long HRT of 56 days, relative to the aforementioned California study, probably explains the high TKN removals. Harvey and Fox (1973), using undiluted secondary municipal sewage under static conditions, measured TKN reductions of 75-89Z (mean: 86.5X) after 10 days. Nitrate removals were 21-60/ (mean: 39.0'/). The experiment was conducted at 24° C, indoors, in 8 L aquaria exposed to 12 hour daylength provided by artificial illumination. The relatively low nitrate removal efficiencies compared to TKN were due in part to nitrification taking place in the systems, as evidenced by increases in nitrite-N during the experimental period. Rodgers et ah (1978) measured nitrate concentrations at the inlet and outlet of a partially shaded, duckweed covered lagoon receiving coal ash sluice water in South Carolina. Average annual nitrate removal efficiency was 74X, and the final effluent contained 1.2 mg NO3-N/L. Although the water was not static, no HRT estimate was provided. Water temperatures were high, averaging 29.6° C (maximum: 44.5° C). Porath and Pollock (1982) studied the feasibility of using duckweed for ammonia stripping in circulating aquaculture. Using 200 L tanks situated outdoors, circulation of fish culture water under a duckweed mat yielded 50X and 90X removal of ammonia (to <0.2 mg/L NH3==NH4) in 24 and 48 hours, respectively, while ammonia levels in a duckweed free control remained constant. The water also contained nitrate (3 to 5 mg/L), but ammonia was 41 taken up preferentially. Ammonia removal was greater at a pH level of 8.6 than at 7.1, while the reverse was true for nitrate removal. Under stagnant conditions ammonia levels increased regardless of treatment, but decreased rapidly in the duckweed covered tanks once mixing was instituted. Water temperature was 18-27° C. Whitehead et al. (1986) in British Columbia measured removals by duckweed of total ammonia, nitrate (including nitrite) and total-N from dilute dairy manure during the summer. Using 0.09 m2 channels under greenhouse conditions, HRTs of 7 to 40 days and daily duckweed cropping rates of 5-30X of the channel area were tested. The optimum treatment combination, 10d HRT and lOX/d duckweed cropping rate, yielded removal efficiencies of 97.07. for ammonia-N, 8.8X for nitrate-N and 45.6X for total-N; the corresponding effluent concentrations were 2.1, 1.2, and 159.3 mg N/L, respectively. The duckweed harvest accounted for a maximum of 58.3X of the influent N mass. The data obtained suggested that additional N entered the duckweed channels, a phenomenon attributed to import via insects or microbial N fixation (though the actual source was not established). Denitrification, accomplished by obligately anaerobic bacteria (Middlebrooks and Pano, 1983), would appear to be favoured in duckweed covered lagoons. While only one study (Culley and Epps, 1973) was found which addresses the issue, it would appear that (carbon not being the limiting nutrient) a duckweed covered lagoon receiving a highly oxidized (i.e. nitrified) wastewater might provide a suitable environment for N removal by d e n i t r i f i c a t i o n . k2 Nitrogen removals have also been reported on the basis of unit area (eg. g/m2.d or kg/ha.d). Culley et al. (1978, in Stephenson et al., 1980) have reported N removals of 1019 kg/ha.yr or 1.0 mg/L.d from swine manure in Louisiana; estimating an 240 day (8 month) growing season, this would represent approximately 4.3 kg/ha.d. Culley and Myers (1980) reported that daily harvests of mixed Lemnaceae species from dairy barn wastewater lagoons extracted 1374 kg TKN/ha.yr (an estimated 5.7 kg TKN/ha.d based on a 240 day season), or reduced the concentration by 1.0-1.25 mg TKN/L.d. Said et a/.(1979) measured duckweed TKN contents of 5.14 to 6.96 V. dry matter and productions of 14.9 g/m2.d by daily harvesting (33X/d) in ponds containing static dairy barn wastewater; from these data it is estimated that nitrogen removal via duckweed was between 7.7 and 10.4 kg TKN/ha.d. Kvet et al. (1979, in Culley et al., 1981) has reported 2 kg/ha.d N removal rate by duckweed growing on ponds in Czechoslovakia. Stowell et al., (in press) have estimated from published literature that duckweed systems are capable of a maximum total-N uptake rate of 4.8 kg/ha.d from sewage. Evidently, higher N removal rates are possible from wastewaters containing higher N concentrations than sewage. 2.6.5 Phosphorus The removal of this nutrient from wastewater by duckweed is considerably less than that of nitrogen, due to the lower metabolic requirement for P than for N. The N:P ratio in duckweed tissue, based on published reports, ranges from 2 to 9, with typical values of about 4 (Culley et al., 1981; Stowell et al., [in press]; Hillman and Culley, 1978). Other factors also affecting P (and N) removal include season, phosphorus loading and duckweed cropping rate. Table 8 summarizes some of the data available on P removal from wastewater by duckweed systems. Wolverton and McDonald (1979) in Mississippi reported wintertime P removal efficiencies of 35 to 54Z in outdoor ponds (54 d HRT, unharvested) receiving domestic wastewater with 1.8-4.0 mg P/L. Harvey and Fox (1973) reported P removals of 56-81Z after 10 days, using recirculated, undiluted secondary wastewater (initial concentrations of 4.6-15.4 mg P/L) under unharvested laboratory conditions. Sutton and Ornes (1977) in Florida measured P removal efficiencies of 80X, 92X and 97Z after 14, 28 and 84 days, respectively, from static secondary municipal effluent initially containing 3.3-3.5 mg P/L. The experiment was conducted in an outdoor laboratory under partial shade and 50Z of the duckweed was cropped weekly. Whitehead er al. (1986) in British Columbia reported summertime P removal efficiencies of 13V. to 43X from diluted dairy manure in cropped, intermittently loaded channels in a greenhouse. At the optimum treatment combination of 20d HRT and 10X/d duckweed cropping rate, an average overall P removal efficiency of 21.0X was achieved. Duckweed harvest was found to account for all of the P removed. Rodgers et al. (1978) in South Carolina reported annual mean orthophosphate removals of 80Z from coal ash sluice water passing through a natural lagoon containing duckweed. Phosphorus removal rates per unit area have also been reported. Culley and Epps (1973) projected from laboratory results, using septic tank effluent and diluted swine manure, that P removal rates of 192-720 kg/ha.yr (equivalent to 0.82-3.00 kg/ha.d over 240 days) would be possible under favourable conditions. Said et al. (1979) reported data from which removals of 0.017 Table 8. Removal of phosphorus reported for wastewater treatment systems containing duckweed. Wastewater and location Phosohorus influent concentration effluent Removal V. Hydrau1i c regime Cropped Observat i ons References Domestic sewage, secondary effluent Mississippi 1.8 - 4.0 1.0 - 2.6 35-54 Flowing No Outdoor, winter wolverton and McDonald, 1' Domestic sewage, secondary effluent F1 or i da 3.3 3.3 0.65 0.25 80 92 Static (14 d) (28 d) Yes Yes Outdoor lab., <50Z ful1 sun Sutton and Omes, 1975 m M « it 3.5 0.09 97 (84 d) Yes Sutton and Ornes, 1977 secondary effluent 1aboratory 4.6 - 15.4 1.4 - 2.6 58-61 Static (10 d) No 24°c, indoor Harvey and Fox, 1973 secondary effluent 7.3 7.1 1.4 flowing (1,3 d HRT) No Outdoor, autumn, pi lot scale Dewante and Stowell, 1981 Coal ash sluicewater, 0.5a 0.1a 80 flowing No Natural swamp, Rodgers et al., 1978 South Carolina annual mean values a as ortho-phosphate, all other values as total phosphorus 4^ ^5 kg P/ha.d and 0.003 kg P/ha.d were estimated for summer and winter, respectively. The experiments used static dairy barn wastewater, and the duckweed was cropped at 33%/d during the summer and lOZ/d during the winter. The large difference between the laboratory and outdoor results in the latter two studies was not addressed by the latter authors despite the fact that the report was co-authored by members of the earlier study team. Harvey and Fox (1973) estimated a P removal rate of 1.7 kg P/ha.d from indoor duckweed cultures growing on secondary municipal effluent. Sutton and Ornes (1977), in outdoor experiments also using municipal effluent, measured removals of 0.015-0.149 kg P/ha.d. The above results indicate that extrapolations from laboratory to field scale may over estimate P removal. 2.6.6 D i s i n f e c t i o n Pathogen removal in the presence of duckweeds reportedly does take place to a certain extent. Possible mechanisms include physical sedimentation, secretion of bacteriostatic compounds, or as a result of direct consumption of bacterial cells by zooplankton (Ehrlich, 1966; Dinges, 1974). Amborski and Larkin (1981) in Louisiana have reported that there is a potential for bacterial and viral disease transmission through harvested duckweed, though they were not able to quantify the risk. Their study found that chlorination or other disinfection methods would probably be necessary to remove bacterial and viral pathogens associated with duckweed harvested from dairy waste lagoons. The latter authors concluded that there was a need to further document the potential problem of disease transmission associated with the use of duckweed systems, and proposed a series of specific research areas. 46 Additional information on wastewater disinfection in association with duckweed was not available. 2.7 Design and Operation of Treatment Systems using Duckweed It is important to state at the outset that there are no known full-scale, operating wastewater treatment systems which purposefully use duckweed as a functional system component. Despite the abundance of literature on the potential usefulness of duckweed in wastewater treatment, there are actually few reports on how to design and operate a duckweed system. A greater amount of information is available for the design of aquatic processing units (APU, Stowell et al., 1981) based on water hyacinth (Wolverton, 1979; Gee and Jenson, 1980) and emergent macrophytes such as reeds and rushes (Seidel, 1976; Wile et al., 1982; Reed et al., 1984; de Jong, 1976). Some of the concepts applied in the latter systems may be applicable to duckweed systems. A useful overview of the concepts involved is provided by Stowell et al. (1981), and a review of the performance of existing duckweed systems by Stowell et al. (in press). Table 9 summarizes available information on wastewater treatment systems that include duckweed. The important design criteria for duckweed systems include basin geometry, hydraulics and duckweed management regime. Basin geometry includes such specific criteria as lagoon length:width ratio, orientation relative to prevailing winds, and water depth. The hydraulics include such considerations as water inlet and outlet structures, influent loading rate, additional flow control structures, and hydraulic retention Table 9. Descr ipt ion of domestic sewage treatment systems reported or proposed to contain ducKweed. Wastewater Location Dimensions (m) HRT 8 Remarks References length width water depth days Cropped Other features secondary e f f luent as above as above aerat ion lagoon e f f luent raw sewage secondary e f f luent raw sewage primary e f f luent Card i f f , Cat i fo rn ia Rosevi1le, C a l i f o r n i a Bay St . Louis Miss iss ippi B i 1 ox i , Miss iss ippi Israel 4.0 7.5 Texas 0.5 49 Miss iss ippi ? Miss iss ippi 152 2.4 1.5 0.75 0.5 ? 1.5 0.4 0.2 9 2.8 ? 3 9 0.9 4.5-5 1,3 54 21 >5.3 15 7 NO NO No No manifold i n l e t , in greenhouse outdoor, autumn Stewart and Ser f I ing . 1979 Dewante and StowelI, 1981 f u l l s c a l e , Dec. -Feb. Wolverton and McDonald, 1979 volunteer duckweed f u l l s c a l e , volunteer duckweed Wolverton and McDonald, 1980 10,20 50Z, outdoor l a b . , summer, Oron et al., 1984 3/week intermittent f low. No proposed design, mul t i -spec ies system Dinges, 1976 5Z/day proposed design, odour Wolverton, 1979 c o n t r o l , primary and 5Z/day secondary treatment 1* a hydraul ic retent ion time D design proposed f o r a small corrmunity of 250-500 population equivalents. 4 8 time. DucKweed management regime refers to aspects such as the mat density (standing crop), cropping frequency and intensity, micronutrient manipulation, pest control, harvesting methods and machinery, as well as wind control structures. 2.7.1 Previously Proposed Designs Wolverton (1979) proposed two designs to treat domestic wastewater from a population of 250-500 people. One was an anaerobic duckweed APU of approximately 0.09ha, 3m deep. The other was an aerobic duckweed APU of approximately 0.14ha (depth not specified) and a length to width ratio of approximately 16.7:1. Hillman and Culley (1978) proposed a system for a 100 cow dairy farm. The system would consist of six lagoons having a depth of 3 feet (1 m), covering a total of 10 acres (4 ha); individual lagoon dimensions were not specified. In this particular system, the harvested duckweed would be fed to cows, and the manure used for energy production by means of an anaerobic digester. The lagoons would receive the digester effluent. Stowell et al. (in press) suggested that, in the absence of information on the maximum surface area that can be covered reliably by a duckweed mat, individual duckweed APUs should not exceed 0.2ha; groupings of duckweed APUs (including berms and roadways) should not exceed I.Oha; and the pond clusters should be surrounded by a windbreak. A system of floating wind barriers that subdivide any pond surface into small cells of <3m diameter has been designed and patented in the United States (Ngo, pers. comm.). The system, which does not interfere with the proposed harvesting method, is reputed to eliminate the lagoon size constraint. 4-9 2.7.2 Harvesting of DucKweed Given the importance that appears to have been placed on harvesting as a means of optimizing the performance of water treatment systems utilizing ducKweed, it is surprising that larger scale experiments aimed at assessing this plant have not practiced ducKweed cropping (Dewante and Stowell, 1981; Serfling and Alsten, 1979). Nevertheless, harvesting of ducKweed has been investigated to a limited extent. Authors of laboratory scale studies or reviews on ducKweed generally suggest that harvesting could be achieved quite simply by means of sKimmer/conveyor devices similar to those used in the petroleum industry (Harvey and Fox, 1973; Culley and Epps, 1973; Hillman and Culley, 1978). Culley et al. (1978) described a system wherein ducKweed overflowed a weir into a gutter pipe which conducted the ducKweed/water slurry to a pumping well. The slurry was pumped to a screening device to remove the water, and the fresh biomass was then available for use or further processing. A problem was encountered however in maintaining the ducKweed well mixed with the water in the pumping well (Culley, pers. comm.). A mobile sKimmer/conveyor has recently become commercially available (Lemna Corporation, 1986). The system is being marKeted for use in conjunction with the aforementioned floating windbreaK apparatus. Potential problems associated with excessive consumption of the plants by waterfowl (Blair, pers. comm.) do not appear to have been addressed in the l i t e r a t u r e . 50 2.8 Research Needs It is evident from the literature that considerable information is available on environmental requirements and tolerances of Lemnaceae and on their capacity to contribute to water treatment. Similarly, much is known about the mineral composition and nutritional quality. There is a distinct lack of information however regarding the actual deployment of wastewater treatment systems wherein the qualities of duckweed are put to use in a purposeful rather than incidental manner. The following specific data voids have been identified as a result of this literature survey: - the relationship between temperature or season, macro and micronutrients, hydraulic loading, and cropping rate, as these affect the overall nutrient removal by duckweed from a wastewater stream; - the relationship between windspeed, pond configuration and surface area, and the disruption of duckweed mats, as applicable to windscreen design; - the method and frequency of harvesting or cropping, as it affects wastewater treatment performance, with reference to suspended solids removal and the release of nutrients from the plants; - the actual design and performance of equipment for full scale harvesting of duckweed; 51 - the options available for post-harvest processing, such as washing, drying, disinfection and storage; - the potential or actual non-feed uses of duckweed, such as land application or extraction of special substances; - the technical and commercial feasibility of employing duckweed as a feed ingredient, soil additive, or other product, and the economics of duckweed use in water treatment; - the use of controlled environments for optimizing the treatment potential and productivity of duckweeds, as a means of extending both the geographic range and the growing season; - the inter and intra-specific differences in duckweed physiology, with respect to temperature tolerances, growth rates, nutrient accumulation rates, composition and nutritional quality, and flowering requirements, as applicable to selective breeding of superior genotypes. - the diseases and pests of duckweeds, as these might occur in water treatment systems. 52 3. MATERIALS AND METHODS 3.1 Experimental Set-up The study reported here was carried out at the Duncan sewage treatment lagoon complex, from June 2 through September 15, 1986. The entire experimental set-up was situated within the outlet end of Cell 4, the final clarification lagoon (Fig. 3). Placed in the lagoon, the experimental enclosures were exposed to the same environmental conditions (weather, water temperature, biota) as the lagoon itself. 3.1.1 Experimental design The experimental design consisted of a comparison of three treatments, representing cropped duckweed, uncropped duckweed, and duckweed-free (control) conditions. Table 10 describes each treatment in detail. 3.1.2 A p p a r a t u s The basic experimental unit consisted of a large plastic fabric bag or tank (Fig. 4) attached to a flotation collar constructed of wood and plastic foam blocks. Each bag had a volume of approximately 3700 L, water depth of approximately 1.85 m, and a water surface area of 2.25 m2. The tanks were provided with an inlet at the bottom, and an outlet at 20 cm below water level. The inlet and outlet were placed diagonally opposite one another to prevent short circuiting of flows. A wooden frame covered with two-inch (5 cm) wire mesh was placed over each tank, in an attempt to exclude waterfowl. 53 a * 3 A E R A T I O N C E L L No. I To Cowlchon River 7 C E L L No. 5 S 0 2 Addit ion Chlorine Addition Plant lChlorine-502 Bldg. and Lab C L A R I F I E R C E L L Na 2 No- 4 A E R A T I O N C E L L No. 3 Figure 3. Overview of the Duncan sewage treatment lagoons. Raw sewage from the City of Duncan enters Cell No. 1 and that from the District of North Cowichan enters Cell No. 3. 54 Table 10. Summary description of the experimental design (see text). Experimental unit: Plastic fabric tank ("1imno-corraP') Volume = 3700 1 Depth = 1.85 m Enclosed surface area = 2.25 m2 Treatment description: Cropped ducKweed Tanks A and B (n=2)a two cropping rates in series. Rate during Period 1 D = 15/ of enclosed area Rate during Period 2 C = 507. of enclosed area Cropping frequency = once every 7 days Uncropped duckweed (n=1) Tank C No duckweed (n=1) Tank D (control) Hydraulic regime: Loading rate = 290 1.d~1 or 129 l.m~2.d~1 Hydraulic retention time = 12.8 d. a n = number of replicates b 4 June to 4 August, 1986 c 4 August to 15 September, 1986 5 5 PVC pipe from distribution box1" plywood float plastic hose lacing plastic float outlet white plastic fabric Figure 4. Diagram of the experimental enclosure or "limno-corral" used in the duckweed culture experiments. (redrawn a f t e r Power and W e n t z e l l , 1 9 8 5 ) 56 Wastewater was taKen directly from the host lagoon using a submersible pump (Model 3-E12N, Little Giant Pump Co., Oklahoma City, Oklahoma) placed at approximately 0.6 m depth. Pumped water was delivered to a four-chambered, wooden distribution box which was mounted on a work raft. Figure 5 illustrates the experimental set-up. The distribution box was fitted with a common inlet and separate, electrically valved outlets, one to each experimental tank. The transparent plastic delivery hoselines were wrapped in opaque foil to prevent algal growth. At 90-minute intervals, the distribution box was automatically filled with lagoon water and the contents of each chamber drained into the four tanks. The activation of the pump and outlet valves was controlled by two electric timers ( Time Command 48, AMF/Paragon, Two Rivers, Wisconsin), and was set to coincide with the daily pattern of flow to the head of the lagoon system (12:00 to 06:00 hours). The sizing of the distribution box was based on the volume requirements to simulate the average summertime hydraulic retention time in the host lagoon, of 12.8 days. Thus each tank received 290 L of influent per day. 3.1.3 Duckweed cropping Once per week a measured surface area of the duckweed mat within tanks A and B was harvested. Tank C, which also contained duckweed, was sampled weekly but not harvested. During the first nine weeks of the experiment, 15X of the enclosed surface area was harvested, and during the final six weeks the fraction was increased to 50X. For discussion purposes, the intervals corresponding to the two cropping regimes are referred to as Periods 1 and 2, respectively. During Period 1, harvesting entailed removing with a dipnet all the plants within a 0.338 m2 wooden frame that was floated carefully on Figure 5. Schematic diagram of the experimental set-up used in the duckweed culture experiments, a: 115v A.C. electrical outlet; b; electrical timer; c: submersible pump (activated by timer); d: four-chambered distribution box; e: solenoid valves (activated by timer); f : flexible plastic hose, covered with aluminum foil; i : screened pump intake (situated at 0.6 m). ^ -58 the undisturbed duckweed mat. During Period 2, the cropping procedure involved dividing the duckweed mat in half with a specially constructed, removable wooden partition, and then harvesting all the duckweed from one side of the divider. The duckweed biomass in the uncropped tank was also measured by the same procedure described above for the cropped tanks during Period 1, although most of the fresh duckweed was immediately returned to the tank. Samples of approximately 200 g of fresh duckweed were taken weekly, placed in sealable plastic bags, and stored in an ice chest for transport to the laboratory. Duckweed fresh weights were measured in the field using a dietetic scale. Externally adsorbed water was removed prior to weighing by centrifugally dewatering the cropped biomass in a simple, fine-meshed (0.5 mm) nylon basket. Dry weights were determined in the laboratory after drying the duckweed samples to constant weight at 100° C. 3.2 Sampling and Analysis 3.2.1 Water quality The overall sampling regime was based on weekly grab samples of the common influent and individual tank effluents. All water samples were taken at 0.4 m depth during mid morning, using an extension-pole sampler equipped with a one-litre plastic bottle that was filled by removing a remotely controlled stopper. (Initially, shallower sampling was practiced, but was discontinued due to contamination with duckweed solids due to turbulence during filling. 59 The hydrogen ion concentration in each water sample was measured immediately with a standard electrical pH meter. Each sample was then subdivided into unfiltered and filtered (Whatman Inc., grade 202 filter paper) sub-samples, stored in pre-labelled plastic bottles, and refrigerated at 2° to 4° C for up to 24 hours until delivered to the laboratory. The water quality parameters measured in the field included pH, dissolved oxygen (DO) and temperature, the latter two using an oxygen probe equipped with a thermister (YSI Model 54, Yellow Springs Instrument Co., Yellow Springs, Ohio). Parameters determined in the laboratory included total ammonia (NH3-N), nitrate plus nitrite (referred to henceforth as nitrate, NO3-N), total Kjeldahl nitrogen (TKN), dissolved orthophosphate (ortho-P), total phosphorus, chemical oxygen demand (COD), total suspended solids (TSS), and volatile suspended solids (VSS). All laboratory analyses were performed according to standard methods (A.P.H.A., 1985). Nitrogen and phosphorus were determined colorimetrically with an Auto-Analyzer II (Technicon Corp., Tarrytown, N.J.) in accordance with procedures recommended by the manufacturer, and concentration values were calculated as molecular N or P. 3.2.2 Plant yield Duckweed relative growth rate (RGR) and crop growth rate (CGR) were estimated by linear regression analysis. Cumulative dry matter yields were plotted versus time, for the 15X and 50X weekly cropping regimes (Periods 1 and 2, respectively). The slopes of the lines of best fit were computed utilizing a computer spreadsheet program (1-2-3™, Lotus Development 60 Corporation, Cambridge, Mass.). The slope was then used as an estimate of the RGR. Analysis of covariance and the Newman-Keuls multiple range test were then used to test for statistical significance (Zar, 1974). 3.2.3 Plant nutrient content Total Kjeldahl nitrogen and total phosphorus content of dried duckweed tissue were determined for each weekly sample from the cropped tanks. The uncropped treatment was sampled less frequently. Tissue sample preparation involved digestion of 0.2 g of dried plant material according to the method of Schumann et al. (1973). The extracts were then diluted prior to colorimetric determination of TKN and total-P using the Auto-Analyzer II previously described. Weekly mass removals of N and P via plant uptake were then calculated by multiplying the dry matter yield data by tissue N or P content. 3.3 Estimation of Treatment Efficiency The analysis of nutrient removal was based on two different approaches: mass balance and mass flux. 3.3.1 Mass balances Overall N and P mass loading and discharge (via influent and effluents, respectively), as well as nutrient mass removed by duckweed cropping, were calculated for each cropping period. Nutrient removal efficiencies were calculated from the mass data by expressing, as a percentage of the total (measured) mass loading, the difference between total nutrient mass added and the total discharged. The relationship is described in Equation (1), below: 61 E 0 = 100X . [(c-e)/c] ( 1 ) where E Q = overall nutrient removal efficiency; c = nutrient mass loaded via influent; e = nutrient mass discharged via effluent. The contribution of duckweed to the overall removal of N and P was calculated according to equations (2) and (3), below: E d = 100Z . [(f + h) - b] / (b + c) ( 2) E c = (E d / E Q) . 100/. ( 3 ) where E d = percent removal, via duckweed, of total mass loaded; b = nutrient mass content of initial duckweed stock; c = as in equation (1); f = nutrient mass removed via cropped duckweed; h = nutrient mass remaining in unharvested duckweed; E c = contribution of duckweed to overall efficiency. Nutrient removal efficiency due to non-duckweed processes (E n) was estimated by subtracting E d from E 0. 6 2 3.3.2 Mass flux r a t e s Estimates of N and P mass flux were also used to compare the performance of the three treatments. This approach was selected in addition to the mass balance method to provide data which might be useful for design purposes, particularly if expressed on the basis of unit surface area or unit volume, in the present work, mass flux rates were expressed per unit area on the rationale that duckweed applications would be more likely to be retro-fitted to existing lagoon systems of a fixed available surface area. The rates were estimated by linear regression and statistically compared, as described above for plant yield. 6 3 4 RESULTS AND DISCUSSION 4 . 1 General Observations Dense duckweed populations were obtained in the experimental enclosures, in the dechlorination lagoon (Cell 5) and, to a lesser extent, in Cell 4 (Fig. 3). It was clearly evident that wind was the major factor limiting the establishment of a complete duckweed cover on the larger and more exposed Cell 4. On occasions when wind was absent, a nearly full duckweed cover did occur on the latter lagoon. All the duckweed mats growing on the more quiescent waters became infested by aphids within two weeks after start-up. Aphid population density was measured on one occasion to be in excess of 10 individuals per cm2. By the end of the experiment, predatory ladybird beetles had become established on the duckweed mat; up to 5 beetles were counted on one occasion within a 2.25 m2 enclosure. It is interesting to note that, despite the intensity of the beetle infestation (which from a distance took on the appearance of a gray-brown "mold") the duckweed mats did not show signs of diminishing productivity. Aphids did appear to be less prevalent on the more impermanent duckweed mats which formed and dispersed on the open waters of Cell 4. The major duckweed consumers observed were aphids, snails and ducks. The density of the aphid population has already been mentioned above. The snail population was also very dense, estimated to be in excess of 100 individuals /m2, even in the cropped duckweed mats. A resident population of 10-15 64 ducks was commonly sighted on Cell 4. Other fauna included abundant cladoceran and, to a lesser extent, copepod zooplankton (particularly in the influent), mosquito larvae (control tank only), water boatmen, amphipods, bloodworms (particularly in the uncropped mats), as well as damselflies and dragonflies (and their larvae). Seagulls, red-winged blackbirds, kingfishers, a great blue heron, a muskrat and a mink were also seen in or about the lagoons. 4.2 DucKweed Growth 4.2.1 Standing c r o p Figure 6 illustrates the changes in duckweed biomass per unit area over time. Aside from an accidental spill of the whole duckweed mat from tanks A and B during the fourth week (due to submergence of the tank rims), a trend of increasing standing crop was evident in all duckweed-covered tanks during the first nine weeks. Variability between tanks A and B was l.&V. during Period 1 and 10.6X during Period 2. During Period 1, the 15X/week cropping rate in the harvested treatment was evidently exceeded by the rate of duckweed production, and dry matter densities reached approximately 130 g/m2. The unharvested duckweed mat (tank C) exhibited greater increase in biomass density, reaching approximately 230 g (dry wt.)/m2 by the end of the period. There was a marked drop in standing crop of the uncropped treatment (to approximately 130 g/m2) during the final week of Period 1. The reason for the decline is not well understood, but may have been due to sampling error, consumption by aphids or waterfowl, massive mortality due to crowding, or other unexplained factors. 250 X J C S I E ch CL O a CJ) x» c o - » — CO 15%/week cropping 200-150-50%/week cropping 100 re-start after accidental spill Legend O CROPPED UNCROPPED 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Figure 6. Changes in standing crop of duckweed in the cropped treatment (average of tanks A and B) and in the uncropped treatment (tank C). Day 0 was 2 June and day 105 was 15 September, 1986. 66 During the final six weeks of the experiment (Period 2), when the duckweed cropping rate was increased to 50X/week, the biomass areal density decreased in the cropped units, to a relatively constant level of 60 to 80 g (dry wt.)/m2. The uncropped treatment underwent a marked decrease in duckweed density (to approximately 130 g/m2) during the first week of Period 2, increasing thereafter to a final level of about 210 g/m2. Dry matter densities measured in the lagoons in late July (day 77) were 23.4 g/m2 in Cell 5, and 31.2 g/m2 in Cell 4. The density in the uncropped enclosure on the same day was approximately 170 g/m2. 4.2.2 Duckweed composition Table 11 summarizes the nutrient and water content of the duckweed grown in the experimental enclosures. Water content averaged 93.7 percent and was not significantly different between the cropped and uncropped treatments. There was likewise no difference in water content due to the change in cropping rate. Dry matter nitrogen content ranged from 6.8X in the cropped to 6.5X in the uncropped treatment during Period 1. During Period 2, tissue N content was 5.8X in the cropped duckweed and 6.2X in the uncropped duckweed. The corresponding crude protein content ranged from 36.4X to 42.5X. Phosphorus content during Period 1 was 1.2X in both the uncropped and cropped treatments, and during Period 2, ranged from 1.2X in the cropped, to 1.4X in the uncropped treatment. 67 Table 11. Composition of duckweed grown in the experimental enclosures during Periods 1 and 2. (see text) Component (*) Treatment Cropped3 Period 1 Period 2 Uncropped Period 1 Period 2 Mo i sture D Dry matter Nitrogen 0 1 Phosphorus Crude protein e mean S.D.C minimum max i mun sample no. mean S.D. minimum max i mun sample no. mean S.D. min imum maximun sample no. mean S.D. minimum max i mun sample no. mean S.D. mini mum maximun sample no. 94.10 0.63 93.19 95.17 12 5.90 0.63 4.83 6.81 12 6.80 0.59 5.99 8.13 12 1.31 0.09 1.18 1.45 12 42.48 3.71 37.45 50.81 12 93.76 0.24 93.33 94.23 12 6.24 0.24 5.77 6.67 12 5.80 0.50 4.90 6.70 12 1 .24 0.07 1.08 1.31 12 36.37 2.84 30.37 41.87 12 93.00 0.77 92.50 94.00 6 7.00 0.77 6.00 7.50 6 6.54 0.77 5.99 7.08 2 1.16 0. 03 1. 14 1.19 2 40.85 4.81 37.45 44.25 2 93.85 0. 16 93.70 94.00 6 6. 15 0.16 6.00 6.30 6 6.22 0.32 5.99 6.45 2 1.41 0. 39 1. 14 1.69 2 38.88 2.02 37.45 40.31 2 a cropping was 15X/week during Period 1, 50X/week during Period 2. D fresh weight basis, a l l others on dry matter basis. c standard deviation d as total Kjeldahl N or TKN. e TKN X 6.25 68 The changes in nutrient content do not appear to be related to cropping as much as to other potential factors such as nutrient availability and weather effects. Higher influent nutrient concentrations and less cloudy weather were recorded during Period 2 than Period 1. 4.2.3 Duckweed productivity The cropped duckweed mats yielded an average of 784 grams of dry matter (range = 754 - 815 g) per 2.25 m2 enclosure over the 105-day experiment. Figure 7 illustrates the cumulative harvest (dry weight) over time. Regression lines are drawn through the data for Period 1 (15X/week cropping) and Period 2 (50X/week cropping). The slopes of the lines represent the relative growth rates (RGR) which, when expressed per unit area represent the crop growth rate (CGR) or yield (Table 12). The average CGR obtained at the 15X/week cropping rate during Period 1 was 2.02 g/m2.d (dry wt.). During Period 2, the 50%/week cropping rate yielded 6.37 g/m2.d (dry wt.), or 3.2 times more duckweed than the earlier cropping rate. The difference in RGR between periods was statistically significant (P<0.001). The increase in duckweed productivity between Periods 1 and 2 can be attributed largely to the 3.33-fold increase in cropping rate. However, other factors such as decreased crowding, increased concentrations of nutrients in the influent water, and weather may also have had an effect. The coefficients of determination ( r 2 values) presented in Table 12 indicate that the high CGRs recorded during the 50Z weekly cropping regime were sustainable. Such a result implies that, over a 120-day growing season that can be expected for duckweed at the Duncan location (Hudson, pers. comm.), duckweed yields of 7.6 3 5 0 - r T J E ty) d) > O > 3 3 0 0 2 5 0 -2 0 0 -1 5 0 -_g 100 3 5 0 -CGR = 6.37 g.rrf ' .d" (rJ = 0.987) CGR = 2.02 g.m~\d (r ? =0.977) Legend O 15%/WEEK • 50%/WEEK - l 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Figure 7. Cumulative ducKweed harvest from the cropped treatment; CGR or Crop growth rate is the slope of the regression line drawn through the data f o r each period; r 2 is the coefficient of determination (Zar, 1974). Table 12. Dry matter production rates of harvested duckweed {Lemna minor) growing on domestic sewage in experimental enclosures, 23 June to 15 September, 1986, Duncan, British Columbia. Interval Cropping rate RGRa CGRD (day no.) {'/. area) (g.d~1) (g.rrr 2.d~ 1) 21 - 63 15X/week 4.5 2.0 (r2=0.967) 63 - 105 50X/week 14.3 6.4 (r2=0.987) a Relative Growth Rate, calculated from slope of linear regression; enclosed area was 2.25 m2. D Crop Growth Rate, (RGR expressed per unit area). 71 tonnes (dry matter) per hectare are achievable. This yield compares favourably with other agricultural crops (Table 13). The total mass of N and P that was removed via duckweed uptake over the 42-day, 50X cropping interval corresponds to 277 mg N /m2.d and 56 mg P /m2.d. The average crude protein yield over the same interval is equivalent to 1.7 g/m2.d. The above nutrient uptake rates represent yields of 332 kg N/ha, 67 kg P/ha and 2.1 tonnes of crude protein per hectare, over a 120-day growing season. 4.3 Water Quality 4.3.1 T e m p e r a t u r e Water temperatures ranged between approximately 17° and 25° C, the warmest period being during the middle of August. There was no marked difference in temperature between the experimental units and the host lagoon, and similar water temperatures were recorded in the presence or absence of duckweed. Deeper waters were generally cooler than surface waters. The maximum temperature difference between surface (5 cm) and bottom (>150 cm) water within the experimental tanks was <2.5 C°, and a similar condition generally existed in the lagoon. However, a maximum surface-to-bottom differential of 6.5 C° was recorded in the host lagoon on August 11. 4.3.2 Dissolved oxygen Figures 8 through 11 summarize the dissolved oxygen (DO) profiles recorded during mid morning. Lower DO levels were generally found in the duckweed-covered tanks than in the control or in the lagoon. The hypoxic 72 Table 13. Comparison of dry matter yields of duckweed (Lemna minor) and other agricultural crops. Crop Yield (t.ha - 1.yr - 1) Crude Protein (V.) Duckweed3 2.4 - 7.6 42.5 - 36.4 Soybeans'3 1.6 41.7 Barley 0 1.9 -Oatsc«d 1.9 -Spring wheat0 2.1 -A l f a l f a hayc 4.4 - 15.7 17.0 Tame hayc 6.2 -3 present study; range represents 15/ and 50X weekly cropping. b Hi 1lman and Culley, 1978. c B r i t i s h Columbia Ministry of Agriculture and Food, 1982. d as grain 73 nature of the duckweed units was established early during the experiment, with DO levels of 0.5 mg/L being common at 5 cm depth during the day. The cropped units tended to have higher DO in the shallower waters than the uncropped tank. The lagoon and the duckweed-free control exhibited DO profiles that are typical of a plankton-rich community, with oxygen concentrations of 2-15 mg/L being common in the illuminated shallower zone, and levels approaching zero mg/L DO at the bottom. Cell 4 itself exhibited DO profiles which are typical of a facultative wastewater lagoon (Caldwell et al., 1977). The changes in DO over time largely reflected the changes in hydraulic and organic loading, as well as duckweed cover. The development of a dense duckweed cover on Cell 4 coincided with the summertime minimum hydraulic loading (and consequently with the maximum hydraulic retention time), both factors contributing to the creation of hypoxic to anaerobic conditions throughout the water column. 4.3.3 Hydrogen ion concentration (pH) Measurements of pH were made only during Period 2, and only in surface water samples (Table 14). The pH in the duckweed-covered tanks ranged from 6.1 to 6.7, compared to 6.5 to 7.8 in the control and 6.0 to 7.8 in the influent (Cell 4). The cropped treatments exhibited relatively stable pH levels over time (range: 6.2 - 6.6), while the uncropped unit exhibited increasing acidification. The influent and the effluent from the control exhibited similar trends in pH, levels in the latter being slightly higher than in the influent. Legend A 21 June O 28 July • 8 Sept. Dissolved Oxygen (mg.l" 1) Figure 8. Profiles of dissolved oxygen concentration measured in the experimental units containing cropped ducKweed, on three occasions during the summer of 1986. E CL Q 100 150 200 Legend A 21 J u n e O 2 8 Ju l y • 15 S e p t . Dissolved Oxygen ( m g . f ) Figure 9. Profiles of dissolved oxygen concentration measured in the experimental units containing uncropped duckweed, on three occasions during the summer of 1986. Legend A 21 June ^ O 28 July • 15 Sept. - i i i I i i I i i 0 2 4 6 8 10 12 14 16 18 Dissolved Oxygen (mg.l" 1) gure 10. Profiles of dissolved oxygen concentration measured in the experimental units containing no duckweed (control), on three occasions during the summer of 1986. O N Legend A 21 June O 28 July • 15 Sept. Dissolved Oxygen ( m g . f 1 ) gure 11. Profiles of dissolved oxygen concentration measured in the host lagoon (Cell No. 4) at Duncan, B.C. on three occasions during the summer of 1986. Table 14. Record of pH in the camion influent and in the effluents from the experimental enclosures during Period 2 (see text). Day No. Treatment Influent Duckweed No Duckweed (host lagoon) cropped" uncropped 63 7.4 6.6 6.7 7.8 70 7.8 6.5 6.5 7.7 77 6.1 6.4 6.4 6.9 83 6.0 6.6 6.4 6.5 90 6.2 6.4 6.4 6.6 98 6.4 6.2 6.3 6.5 105 6.5 6.5 6.1 6.7 Duckweed cropped at 50X/week. 79 Interpretation of the pH results is hampered somewhat by the lack of data from different depths. Nevertheless some causal relationships can be inferred from the available results. The acidification trend in the uncropped duckweed system may be associated with the accumulation of organic acids from decomposing plant biomass. However, the influent pH was lower than might have been anticipated in a system containing a dense phytoplankton population. Levels of pH>8 have been commonly reported from nutrient-rich lagoons during summer (Caldwell et al., 1973; Cole, 1979). The low values recorded at Duncan may have been due in part to the time of sampling (usually before 11:00 a.m.), as the nocturnal acidification of the water would not yet have been offset by the alkalization due to algal photosynthesis (Cole, 1979). 4.3.4 Suspended solids Volatile suspended solids (VSS) comprised in all samples more than 90/ of the total suspended solids (TSS), indicating that the majority of the suspended solids consisted of organic matter. For this reason all subsequent data analysis was based on VSS. Table 15 and Figure 12 summarize the VSS concentrations. Mean influent VSS concentration was higher during Period 1 than Period 2. Possible reasons for this include the greater duckweed cover on Cell 4 during the latter period, and the increased retention time in the lagoon resulting from the virtual absence of rainfall during the same period. From a mean influent VSS concentration of 44 mg/L during Period 1, the cropped duckweed units yielded an average effluent concentration of 23 mg/L, while the uncropped treatment averaged 35 mg/L and the control 53 mg/L. During Period 2, 80 Table 15. Concentrations of volatile suspended solids (VSS) in the common influent and effluents from the experimental enclosures. Treatment Period 3 mean (range) s.d. b C . v . c n' (mg/L) (X) i nf1uent 1 43.7 (18.0-60.0) 19.9 45.5 9 2 36.0 (24.0-66.0) 15.2 42.2 7 cropped 1 23.3 (12.0-42.0) 7.1 30.5 20 ducKweed 2 23.0 (15.0-44.0) 8.4 36.5 14 uncropped 1 34.8 (14.0-37.0) 7.3 21.0 9 ducKweed 2 20.7 (15.0-38.0) 8.0 38.6 7 control 1 52.9 (28.0- 109.0) 31.6 59.7 10 2 56.3 (18.0- 106.0) 40.7 72.3 7 a Period 1 from 2 June to 4 August, weeKly cropping of 15/ of enclosed area; Period 2 from 4 August to 15 September, weeKly cropping of 50/ of enclosed area. D standard deviation c coefficient of variability d number of samples Figure 12. Changes in the concentration of volatile suspended solids (VSS) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. ^ Cropping rate was 15Z/week during Period 1 and 50Z/week during Period 2. 82 the average influent VSS level was 36 mg/L, compared to 23, 21 and 56 mg/L in the effluents from the cropped, uncropped and plant-free treatments, respectively. A dense algal population developed in the control unit during mid summer, which was reflected in increased VSS concentrations (Fig. 12). The duckweed cover was evidently able to suppress algal growth. The higher average solids content in the uncropped versus cropped treatment during Period 1 may be attributed to the decomposition of unharvested duckweed. During Period 2 there was no difference in effluent VSS concentrations between cropped and uncropped treatments. The decomposition of unharvested duckweed in the uncropped tank was evidently not reflected by a measurable increase in VSS during the latter period. The reason for this unexpected result is not known, but may include consumption of the duckweed by ducks or macro-invertebrates, absence of settling of detrital duckweed due to entrapment within the living duckweed mat, and in situ decomposition leading to the release of soluble rather than particulate decomposition products. 4.3.5 Chemical oxygen demand Table 16 and Figure 13 summarize the influent and effluent COD concentrations. Average influent COD concentrations were higher during Period 1 (121 mg/L) than Period 2 (110 mg/L), highly variable concentrations being evident during the first period. Effluent COD concentrations were lower than influent levels during both Periods in all duckweed units, though not in the control. During Period 1, the cropped duckweed units had an average effluent COD concentration of 94 mg/L, while the uncropped and control units averaged 91 mg/L and 123 mg/L, respectively. During Period 2, the average effluent COD levels were 101 mg/L in the cropped, 106 mg/L in the uncropped and 136 mg/L in Table 16. Concentrations of chemical oxygen demand (COD) in the conmon influent and effluents from the experimental enclosures. Treatment Period 3 mean (range) (mg/L) i nf1uent 1 2 cropped 1 duckweed 2 uncropped 1 duckweed 2 control 1 2 121.3 (92.0-162.0) 100.5 (101.1-123.0) 93.8 (62.0-124.0) 100.5 (85.0-122.0) 91.0 (69.0-140.0) 105.7 (97.0-126.0) 123.2 (91.0-192.0) 136.0 (99.0-194.0) s.d. b C.V.C n d (/.) 27.6 22.8 9 8.6 8.6 7 12.2 13.0 20 10.9 10.8 14 12.1 13.3 10 10.4 9.8 7 30.7 24.9 10 33.7 24.8 7 3 Period 1 from 2 June to 4 August, weekly cropping of 15/ of enclosed area; Period 2 from 4 August to 15 September, weekly cropping of 50/ of enclosed area. D standard deviation c coeff i c i ent of var i ab i1i ty d number of samples 220 0 Period 1 Period 2 I ! -1 1 1 1 1 - T 1 1 1 I I 1 r 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend INFLUENT CROPPED  UNCROPPED CONTROL Figure 13. Changes in the concentration of chemical oxygen demand (COD) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15X/week during Period 1 and 50X/week during Period 2. 00 85 the control units. Coefficients of variability were considerably higher in the control than in the other treatments. Cropping appeared to improve COD reductions slightly during the latter half of Period 1 and during most of Period 2. Effluent COD levels in the control were higher than influent levels, probably due to high concentrations of organic particulates (see VSS concentrations, Fig. 12). Although there was considerable variability, the presence of a cropped ducKweed population was associated during Period 2 with a measurable decrease in COD compared to both the uncropped population and the plant-free control. COD concentrations of <100 mg/L were achieved only in the presence of ducKweed. 4.3.6 Nitrogen Total nitrogen Table 17 and Figure 14 summarize the influent and effluent total-N concentrations. Influent concentrations were on the whole higher during Period 1 (average: 20.4 mg/L) than during Period 2 (average 16.1 mg/L). Effluent total-N concentrations were generally lower than influent levels during Period 1 and during the final three weeKs of Period 2. Mean effluent concentrations during Period 1 were 16.6 mg/L in the cropped, 16.5 mg/L in the uncropped, and 17.2 mg/L in the control treatments. During Period 2, the corresponding values were 15.7 mg/L (cropped), 15.7 mg/L (uncropped), and 14.3 mg/L (control). Lower total-N concentrations were evident in the ducKweed covered units than in the control during Period 1. Such a difference was not clearly 8 6 Table 17. Concentrations of total nitrogen in the common influent and effluents from the experimental enclosures. Treatment Per i od a mean (range) (mg/L) s.d. b C.V.C n d influent 1 20.4 (15.9-27.0) 2 16.1 (13.5-18.2) 3.5 17.2 9 1.7 10.6 7 cropped duckweed uncropped duckweed 1 16.6 (13.8-21.1) 2.2 13.3 18 2 15.7 (12.9-20.0) 2.4 15.3 14 1 16.5 (14.7-22.2) 2.3 13.9 9 2 15.6 (14.1-18.5) 1.7 10.9 7 control 1 17.2 (15.0-17.6) 2 14.3 (12.6-17.1) 3.2 18.6 8 1.7 11.9 7 a Period 1 from 2 June to 4 August, weekly cropping of 15/ of enclosed area; Period 2 from 4 August to 15 September, weekly cropping of 50'/ of enclosed area. b standard deviation c coeff i c i ent of var i ab i1i ty d number of samples 0 Period 1 Period 2 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend INFLUENT CROPPED  UNCROPPED CONTROL gure 14. Changes in the concentration of total nitrogen (total-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was l5'/./week during Period 1 and 50/./week during Period 2. oo 88 discernible during Period 2 and, in fact, the lowest effluent total-N level was obtained in the absence of duckweed. These results indicate that, despite major differences in the types of organisms involved in nitrogen cycling in the duckweed and algal systems, the net result, in terms of effluent N concentrations, was approximately the same. The implications of these results are examined further in Section 4.7. Inorganic nitrogen - ammonia The ammonia-N concentrations are summarized in Table 18 and Figure 15. Ammonia was the predominant N species, comprising 51 to 68 percent of the total-N. During Period 1, ammonia concentrations in the effluents were generally lower than in the influent, the greatest reductions being measured in the control. The influent mean ammonia-N content during this period was 13.5 mg/L, compared to 12.4 mg/L in the cropped, 11.9 mg/L in the uncropped, and 9.6 mg/L in the control treatments. During the first half of Period 2, influent ammonia levels decreased markedly (Fig. 15); effluent levels also decreased, but remained higher than influent levels until the final three weeks. During this period, the average ammonia-N concentrations were 10.0 mg/L (influent), 10.4 mg/L (cropped), 11.6 mg/L (uncropped) and 7.1 mg/L (control). Effluent ammonia levels in the control were always lower than in the influent, and followed a parallel trend during the period. Inorganic nitrogen - nitrate Nitrate (as NOg + N O 3 ) was the least abundant nitrogen species in both the influent and effluents, comprising between 1.2 and 6.6 percent of the Table 18. Concentrations of total ammonia nitrogen in the common influent and effluents from the experimental enclosures. Treatment Period 3 mean (range) s.d. D C.V.C n' (mg/L) (/.) influent 1 2 cropped 1 ducKweed 2 uncropped 1 ducKweed 2 control 1 2 13.6 (11.2-15.5) 10.0 (8.2-11.7) 12.4 (9.7-16.0) 10.4 (8.6-13.9) 11.9 (6.9-15.5) 11.6 (10.1-14.2) 9.6 (6.3-12.4) 7.1 (6.0-8.3) 1.5 11.0 10 1.3 13.0 7 1.8 14.5 20 1.9 18.3 13 2.4 20.2 10 1.6 13.8 7 2.2 22.9 10 0.9 12.7 7 3 Period 1 from 2 June to 4 August, weeKly cropping of 15/ of enclosed area; Period 2 from 4 August to 15 September, weeKly cropping of 50/ of enclosed area. D standard deviation c coefficient of variability d number of samples 18-17-16-15-14-7 13-12-i 11-i 10-I E 9 -8 -7 -6 -5 -Period 1 Period 2 - i 1 1 1 r — i 1 1 1 1 1 1 1 1 1 — 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Legend INFLUENT CROPPED  UNCROPPED CONTROL Time (days) Figure 15. Changes in the concentration of ammonia nitrogen (NH4-N) in the common influent and in the effluents from the cropped ducKweed, uncropped ducKweed and control treatments. Cropping rate was 15X/weeK during Period 1 and 50X/weeK during Period 2. O 91 total-N. Concentrations are summarized in Table 19 and Figure 16. The elevated coefficients of variability during both periods are due in part to the fact that the nitrate concentrations were near the analytical detection limits. Effluent nitrate concentrations during Period 1 were generally higher than influent levels, indicating that nitrification was taking place in the tanks. During this period, mean nitrate concentrations were 0.30 mg/L in the influent, compared to 0.33 mg/L in the cropped, 0.40 mg/L in the uncropped, and 1.72 mg/L in the control treatments. The highest nitrate levels were recorded in the control during a three-week interval of this period. Corresponding decreases in ammonia-N were evident at the time (Fig. 15). During Period 2, influent nitrate concentrations averaged 0.14 mg/L, while effluents averaged 0.09 mg/L in the cropped and uncropped duckweed treatments, and 0.12 mg/L in the control. Organic nitrogen Organic-N was the second most abundant N species in all water samples, comprising 30 to 42 percent of the total-N. Concentrations are summarized in Table 20 and Figure 17. Effluent organic-N levels in the duckweed-covered units were generally lower than influent levels, though not in the control. Organic-N concentrations in the influent exhibited an increasing trend during Period 1, remaining relatively constant during Period 2. The mean influent concentrations, however were similar for both periods (5.8 mg/L and 5.9 mg/L, respectively). Effluent concentrations during Period 1 were 3.7 mg/L in the cropped, 4.1 mg/L in the uncropped, and 5.1 mg/L in the control treatments. 92 Table 19. Concentrations of n i t r i t e and nitrate nitrogen in the corrmon influent and effluents from the experimental enclosures. Treatment Period 3 mean (range) s.d. b C.V.C n' (mg/L) (X) influent 1 0.30 (0.10--0.68) 0. 16 53.3 10 2 0.14 (0. 10--0. 35) 0. 10 71.4 7 cropped 1 0.33 (0. OS--1.00) 0.35 106.1 20 ducKweed 2 0.09 CO. 06--0. 10) 0.02 22.2 14 uncropped 1 0.40 (0.05--1.50) 0.52 130.0 10 ducKweed 2 0.09 (0.06--0. 10) 0.02 22.2 7 control 1 1.72 (0. 10-4.60) 1.84 107.0 10 2 0. 12 (0.07--0.20) 0.05 41.7 6 3 Period 1 from 2 June to 4 August, weeKly cropping of 15X of enclosed area; Period 2 from 4 August to 15 September, weeKly cropping of 50X of enclosed area. b standard deviation c coefficient of variability d number of samples (to 4.9) Period 1 Period 2 "l4 21 28 35 42 49 56 63 70 77 8 4 91 98 105 112 Time (days) Legend INFLUENT CROPPED  UNCROPPED CONTROL Figure 16. Changes in the concentration of nitrite plus nitrate nitrogen (NO2+NO3-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15/./week during Period 1 and 50//week during Period 2. 94 During much of Period 2, suspended algal solids in the control tank caused effluent organic-N to exceed influent levels. The peaks evident in Figure 17 on day 28 for all samples are possibly associated with an algal bloom that occurred in the host lagoon (Cell 4) during the previous week (see VSS concentrations, Fig. 12). During Period 2, effluent organic-N concentrations averaged 5.2 mg/L (cropped), 4.0 mg/L (uncropped), and 7.1 mg/L (control). The effluent from the duckweed-free tank exhibited an increase in organic-N of 1.2 mg/L over the influent, reflecting the increased algal density. The lower levels of organic-N exiting the duckweed-covered units than the control during both periods are probably related to uptake by the plants and, or, the lower algal densities in the shade of the duckweed mats. The latter explanation is also suggested by the VSS data (Fig. 12). 4.3.7 Phosphorus Total phosphorus Influent and effluent total-P concentrations are summarized in Table 21 and Figure 18. The average influent total-P concentration during Period 1 was 4.1 mg/L, of which 3.5 mg/L (85X) occurred as orthophosphate-P. Effluent total-P averaged 3.4 mg/L in the cropped, 3.4 mg/L in the uncropped, and 3.8 mg/L in the control treatments. Orthophosphate-P fraction of total-P averaged 88X in the duckweed units (no difference between cropped and uncropped treatments) and 83X in the control. Presence of duckweed was associated with a reduction in total-P of 0.72 to 0.74 mg/L, compared to 0.27 mg/L in the control. The total-P concentrations during Period 1 largely paralleled total-95 Table 20. Concentrations of organic nitrogen in the common influent and effluents from the experimental enclosures. Treatment Per i od a mean (range) S. d. b C.V.C n d (mg/L) (X) influent 1 5.8 (3.0-7.8) 1. 7 29.3 8 2 5.9 (5.2-8.5) 1. 2 20.3 7 cropped 1 3.7 (2.2-6.1) 1. 1 29.7 18 duckweed 2 5.2 (4.2-7.5) 1. 1 21.2 13 uncropped 1 4.1 (2.7-7.5) 1 . 8 43.9 9 duckweed 2 4.0 (2.9-5.1) 0. 7 17.5 7 control 1 5. 1 (3.0-12.8) 1. 6 31.4 9 2 7.1 (4.3-10.9) 2. 1 29.6 7 a Period 1 from 2 June to 4 August, weekly cropping of 15/ of enclosed area; Period 2 from 4 August to 15 September, weekly cropping of 50/ of enclosed area. b standard deviation c coefficient of variability d number of samples (26.6) Period 1 Period 2 i i i i i i i r 1 r 21 28 35 42 49 56 63 70 77 84 98 105 112 Time (days) Legend jNFLUENT CROPPED  UNCROPPED CONTROL Figure 17. Changes in the concentration of organic nitrogen (organic-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15X/week during Period 1 and 50X/week during Period 2. O N 97 N, although the difference between the control and ducKweed treatments was more pronounced. During Period 2, the influent total-P concentration averaged 4.5 mg/L, while effluent concentrations averaged 4.2 mg/L in the cropped, 4.2 mg/L in the uncropped, and 4.4 mg/L in the control treatments. The corresponding fraction of dissolved orthophosphate-P was 93X in the influent, 95X in the cropped, 99X in the uncropped, and 93X in the control treatments. Reductions of total-P concentrations in the presence of ducKweed were 3.7 to 4.8 fold higher than in the control. During this Period, the presence of ducKweed was associated with a 0.22 mg/L (uncropped) to 0.29 mg/L (cropped) decrease in total-P concentration relative to the influent, compared to a 0.06 mg/L reduction in the control. Cropping the ducKweed resulted in a 0.03 mg/L reduction during Period 1, and a 0.07 mg/L reduction in effluent total-P during Period 2, over the uncropped condition. Dissolved orthophosphate Orthophosphate concentrations increased throughout the experiment (Fig. 19, Table 22). The increases in phosphorus over time may reflect the diminishing dilution effect from groundwater infiltration to the sewerage system (DerKsen, 1981). Another possible cause would be the seasonal release of P from the sediments due to oxygen depletion (Cole, 1979). There was a consistent pattern of higher dissolved P in the presence of ducKweed than in the influent or control, suggesting a release of soluble P from the plants. The higher orthophosphate concentrations measured in the uncropped versus cropped treatment suggest that removal of ducKweed biomass through cropping diminished or prevented the release of P from decaying plants to the water. Table 21. Concentrations of total phosphorus in the camion influent and effluents from the experimental enclosures. Treatment Per i od a mean (range) s.d. D C.V.C n d (mg/L) (/) influent 1 4.1 (3.1-4.8) 0.7 17.1 9 2 4.5 (3.7-4.8) 0.4 8.9 7 cropped 1 3.4 (2.8-4.4) 0.5 14.7 18 duckweed 2 4.2 (3.7-4.5) 0.2 4.8 14 uncropped 1 3.4 (2.9-4.2) 0.4 11.8 9 duckweed 2 4.2 (3.7-4.6) 0.3 7.1 7 control 1 3.8 (3.1-5.3) 0.7 18.4 9 2 4.4 (4.0-4.7) 0.3 6.8 7 a Per i od 1 from 2 June to 4 August, weekly cropping of 15/ of enc1osed area; Period 2 from 4 August to 15 September, weekly cropping of 50/ of enclosed area. D standard deviation c coefficient of variability d number of samples 99 Table 22. Concentrations of dissolved orthophosphate phosphorus in the common influent and effluents from the experimental enclosures. Treatment Period 3 mean (range) s.d. b C.V.C n' (mg/L) (X) i nf1uent 1 2 cropped 1 ducKweed 2 uncropped 1 ducKweed 2 control 1 2 3.5 (3.0-4.1) 4.2 (3.7-4.6) 3.0 (2.0-3.9) 4.0 (3.5-4.1) 3.0 (2.0-3.7) 4.2 (3.6-4.5) 3.2 (2.0-3.9) 4.1 (3.5-4.4) 0.4 11.4 9 0.4 9.5 7 0.5 16.7 20 0.2 5.0 14 0.5 16.7 10 0.3 7.1 7 0.6 18.8 10 0.3 7.3 7 3 Period 1 from 2 June to 4 August, weeKly cropping of 15X of enclosed area; Period 2 from 4 August to 15 September, weeKly cropping of 50X of enclosed area. b standard deviation c coefficient of variability d number of samples CD J , Q_ _l_ 10-9.5-9 8.5-8-7.5 7 6.5 6-5.5-5-4 .5-4 3.5-3-2.5 2H 1.5 H i 7 Period 1 Period 2 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend INFLUENT CROPPED  UNCROPPED CONTROL Figure 18. Changes in the concentration of total phosphorus (total-P) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15X/week during Period 1 and 50X/week during Period 2. o o 4.5-1.5 Per iod 1 Per iod 21 1 I i 1 1 1 i i 1 i 1 1 1 r 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend INFLUENT C R O P P E D  U N C R O P P E D C O N T R O L Figure 19. Changes in the concentration of dissolved orthophosphate (ortho-P) in the common influent £ and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. ^ Cropping rate was 15X/week during Period 1 and 50//week during Period 2. 102. 4 . 4 Nutrient Mass Balances 4 . 4 . 1 N i t r o g e n P e r i o d 1 Table 23 presents the mass balance for N during Period 1. Total measured N outputs were 90.6X, 85.5X, and 83.8X of the measured inputs, in the cropped, uncropped and control treatments, respectively. Unaccounted N was likely lost primarily through denitrification and possibly via some ammonia v o l a t i l i z a t i o n . The highest overall N removal efficiency (E 0) of 21.5X was obtained in the uncropped treatment, followed by 20.6X in the cropped treatment and 18.4X in the control. Duckweed N uptake efficiency (EgO was 6.9X (cropped) and 3.7X (uncropped), accounting for 33.5X of E Q under cropped and 17.2X under uncropped conditions. Non-duckweed removal processes (E Q - E,-)) accounted f o r 13.7X and 17.8X, respectively, of the N removal under cropped and uncropped conditions. The latter value is not very different from the 18.4X N removal efficiency obtained in the duckweed-free control. The results indicate that the presence of duckweed slightly increased N removal, but that cropping did not improve N removal efficiency. The reason for this appears to be related to the fact that the 15X/week cropping rate was associated with a decrease in the relative contribution of non-duckweed processes to overall N removal. The non-duckweed N removal mechanisms that can be identified include settling and adsorption of particulate organic matter, uptake by microbial autotrophs, volatilization of molecular ammonia ( N H 3 ) and denitrification. Table 23. Nitrogen mass balance and removal efficiencies in experimental enclosures in a municipal sewage lagoon, Per i od 1: 2 June to 4 August, 1986. COMPARTMENT TREATMENT Cropped Uncropped Control duckweed* duckweed (no duckweed) NITROGEN INPUT (g) (a) i n i t i a l water (b) i n i t i a l duckweed (c) pumped influent (d) Total N input 58. 1 1.5 423. 1 482.7 58.1 1.5 423.1 482.7 58. 1 0 423.1 481.2 NITROGEN OUTPUT (g) (e) discharged water 336.1 (f) duckweed harvest 14.1 (g) remaining water 70.7 (h) remaining duckweed 16.5 (i) Total N output 437.4 (INPUT - OUTPUT) (g) 45.3 REMOVAL EFFICIENCIES (X) 332.2 0 63.6 17.1 412.9 69.8 345. 1 0 58.2 0 403.3 77.9 Overa11: E 0 = 100X.[(c-e)/c] Duckweed related: E d = 100X.[(f+h)-b]/(b+c) Non-duckweed related: -n [Eo " E d] Duckweed fraction: E,-| as X of E Q 20.6X 6.97. 13.77. 33.57. 21. 5X 3.7X 17.8X 17.2X 18.4X 18.4X * - Duckweed cropping rate was 15X of the enclosed 2.25 m2 area per week. 10^ As examined below on theoretical grounds, cropping of ducKweed does not appear to affect any of these mechanisms in a manner that would explain why non-ducKweed N removal processes would be diminished. It can probably be safely assumed that settling and adsorption would not differ between ducKweed-covered and uncovered tanKs; temperature profiles (Figs. 8 and 9) suggest that there would be no marKed difference in potential thermal convection currents due to the presence of ducKweed. The slightly higher E 0 values obtained in the uncropped ducKweed treatments than in the control suggest that reduced algal N uptaKe due to shading by ducKweed was more than compensated for by ducKweed N uptaKe. The impact of denitrification on the mass balance was probably minimal since nitrate was the least abundant N species, accounting for less than 7/ of the influent total-N. Limited volatilization of ammonia could be expected to have taKen place given the preponderance of total ammonia in the influent; however, NH3 volatilization would have been greater from the control than from the ducKweed covered treatments, given the more basic pH recorded in the absence of ducKweed (Table 14). If the ducKweed cover in fact reduced the water-to-air diffusion of both NH3 and Ng, the aforementioned reduction in non-ducKweed N removal would be associated more with the undisturbed uncropped mat than with the cropped mat. Further study is therefore required to confirm that ducKweed cropping in fact diminishes non-ducKweed N removal and, if confirmed, to explain the mechanisms involved. Period 2 Table 24 presents the mass balance for N during Period 2. Total measured N outputs were 101/, 101/, and 88/ times the respective measured inputs to 105 the cropped, uncropped and control treatments, respectively. The highest overall N removal efficiency (E 0) f o r the period was 10.7X, obtained in the control, followed by 3.9X in the cropped treatment, and 1.7X in the uncropped treatment. DucKweed N uptake efficiency (E^) under cropped conditions was 12.6X, accounting for 323.1X of E 0. Under uncropped conditions E d was 6.4X and accounted f o r 376.5X of E 0. The above results imply that additional, unmeasured N was imported to the duckweed systems (see Section 4.7). Non-duckweed removal values (E n) were negative under both cropped and uncropped conditions, accounting respectively for -8.7X and -4.7X of the N removal. Such negative values again imply a net N import. These results indicate that, under the existing experimental conditions, overall N removal efficiency was greater in the absence of duckweed. The indication that N uptake via duckweed harvest exceeded N removal from the water (i.e. E d > E Q) is evidence that additional, unmeasured N was entering the duckweed tanks (and not entering the control unit). This subject is discussed in detail in section 4.7. The contribution of duckweed to overall N removal was greater during Period 2 than during Period 1, due to the lower N influent concentrations (Table 17) and, possibly, due to more favourable temperature and sunlight regimes. Non-duckweed N removal could not be compared between periods due to the negative efficiency values mentioned above. Decreased efficiency would have been predicted in the absence f cropping, due to release of N from a higher proportion of detrital material in the unharvested mat. Overall N removal efficiency in the control was also less during Period 2, reflecting the lower N loading rate, and/or unmeasured N addition. 106 Table 24. Nitrogen mass balance and removal efficiencies in experimental enclosures in a municipal sewage lagoon, Period 2: 4 August to 15 September, 1986. COMPARTMENT TREATMENT Cropped Uncropped Control duckweed* duckweed (no duckweed) NITROGEN INPUT (g) (a) i n i t i a l water 70.7 63.6 58.2 (b) i n i t i a l duckweed 16.5 17.1 0 (c) pumped influent 191.2 191.2 191.2 (d) Total N input 278.4 271.9 249.4 NITROGEN OUTPUT (g) (e) discharged water 183.7 187.9 170.7 (f) duckweed harvest 32.1 0 0 (g) remaining water 54.8 55.5 47.4 (h) remaining duckweed 10.6 30.4 0 (i) Total N output 281.2 273.8 218.1 (INPUT - OUTPUT) (g) -2.8 -2.9 31.1 REMOVAL EFFICIENCIES {'/.) Overal1: E Q = 100X.[(c-e)/C] 3.9/ 1.7/ 10.7/ Duckweed related: E d = 100/.[(f+h)-b]/(b+c) 12.6/ 6.4/ Non-duckweed related: E n = [E Q - E d] -8.7/ -4.7/. 10.7/. Duckweed fraction: E d as / Of E 0 323.1/ 376.5/. - Duckweed cropping rate was 50 of the enclosed 2.25 m2 area per week. 107 In summary, higher overall N removal efficiencies were obtained in all treatments during Period 1 than during Period 2. The higher cropping rate was associated with a higher contribution by duckweed to N removal. During Period 2, duckweed accounted for a greater fraction of the overall N removal than during Period 1, regardless of whether or not cropping was present. This was probably due to the warmer and less cloudy weather over the latter half of the summer. Also, the addition of N from sources other than the influent was evident in the tanks containing duckweed. 4.4.2 Phosphorus Period 1 Table 25 presents the mass balance for P during Period 1. Total measured P outputs accounted for 93.4X, 89.2/, and 97.9/ of the measured inputs, in the cropped, uncropped and control treatments, respectively. The highest overall P removal efficiencies (E 0) of 17.9/ and 17.8/ were obtained in the presence of duckweed, followed by 6.3/ in the control. Duckweed P uptake efficiency (E^) under cropped conditions was 7.2/, accounting f o r 40.4/ of E 0. Under uncropped conditions E^ was 3.5/ and accounted for 19.6/ of E 0. Non-duckweed removal processes ( E 0 -E^) accounted f o r 10.6/ and 14.4/, respectively, of the P removal under cropped and uncropped conditions. The latter E j values are higher and lower, respectively, than the 6.3/ removal obtained in the absence of duckweed. Table 25. Phosphorus mass balance and removal efficiencies in experimental enclosures in a municipal sewage lagoon, Period 1: 2 June to 4 August, 1986. COMPARTMENT TREATMENT Cropped Uncropped Control duckweed* duckweed (no duckweed) PHOSPHORUS INPUT (g) (a) i n i t i a l water 11.5 11.5 11.5 (b) i n i t i a l duckweed 0.3 0.3 0 (c) pumped influent 83.6 83.6 83.6 (d) Total P input 95.4 95.4 95.1 PHOSPHORUS OUTPUT (g) (e) discharged water 68.7 68.6 78.3 (f) duckweed harvest 2.7 0 0 (g) remaining water 14.1 13.3 14.8 (h) remaining duckweed 3.6 3.2 O (i) Total P output 89.1 65.1 93.1 (INPUT - OUTPUT) (g) 6.3 10.3 2.0 REMOVAL EFFICIENCIES (/) Overal1: E0 = 100/. [(c-e)/c] 17.8/ 17.97. 6.37. Duckweed related: E d = 100/.[(f+h)-b]/(b+c) 7.2/. 3.5/ Non-duckweed related: E n = [E c - E d] 10.6/ 14.4/ 6.3/ Duckweed fraction: E d as / of E Q 40.4/ 19.6/ * - Duckweed cropping rate was 15/ of the enclosed 2.25 m2 area per week. 109 The results indicate that the presence of duckweed was associated with a marked (almost three-fold) increase in overall P removal efficiency, though cropping did not further improve performance. Nevertheless, cropping increased the duckweed-related P removal by over two times. Non-duckweed P removal increased in the uncropped treatment but decreased in the cropped treatment, relative to that in the control. An explanation for why non-duckweed P removal was diminished in the cropped treatment may be found by examining the non-duckweed P removal mechanisms operative in the tanks. The P removal mechanisms that can be identified in a duckweed-free system include settling of particulates, adsorption to tank walls, uptake by autotrophs and consumption by fauna. Settling and adsorption of P would not be expected to differ between planted and unplanted treatments. Also, it is clear that P uptake by duckweed exceeded the uptake by phytoplankton. The predominance of dissolved P (compare Tables 21 and 22) precludes consumption by fauna as a significant direct P removal mechanism. The results suggest that in the absence of cropping there was not a release of dissolved P from decomposing duckweed while, in the cropped treatment, release of P was somehow augmented. Possibly, harvesting interfered with the establishment of a significant decomposer/detritivore community within the cropped enclosures. Mention was made earlier to the fact that bloodworms, which are important consumers of detritus, were observed only in the uncropped duckweed mat. Period 2 Table 26 presents the mass balance for P during Period 2. Total measured P outputs were 104.6/, 106.9/, and 101/ of the respective measured inputs to the cropped, uncropped and control treatments. The highest overall P removal 110 efficiency (E 0) was 7.7X in the cropped treatment, followed by 5.6X in the uncropped treatment and 2.5X in the control. DucKweed P uptaKe efficiency (Ed) under cropped conditions was 8.9X, accounting for 115.6Z of E 0. Under uncropped conditions, E^ was 8.0Z and accounted for 142.9/ of E0. Non-ducKweed removal processes (E n) accounted for -1.2/ and -2.4/ of the P removal under cropped and uncropped conditions, respectively. The results for this period indicate that substantially higher P removals were obtained in the presence of ducKweed, and that cropping at the 50//weeK rate further increased P removal performance. Phosphorus outputs exceeded inputs in all treatments, reflecting, as in the case of N, additional unmeasured P inputs (this subject is discussed in detail in section 4.7). The uncropped treatment exhibited a two-fold increase in P removal efficiency (E 0), relative to the control. Non-ducKweed P removal in the planted tanKs was not quantifiable due to the import of unmeasured P. The data suggest, however, that non-ducKweed P removal was favoured by cropping at the 50//weeK rate. The influence of greater zooplanKton densities under the cropped ducKweed mat on removal of particulate P is a possible explanation for this phenomenon (Ehrlich, 1966), although confirmation would require further research. A similar pattern was evident for P as for N in comparing treatment efficiencies between Periods 1 and 2. Higher overall P removal efficiencies were obtained in all treatments during Period 1, reflecting the higher influent P concentrations and the net absence of unmeasured P imports during this period. Uncropped ducKweed accounted for a greater fraction of the overall P removal during Period 2 than during Period 1, probably as a result of more favourable weather during the latter half of the summer. The I l l Table 26. Phosphorus mass balance and removal efficiencies in experimental enclosures in a municipal sewage lagoon, Period 2: 4 August to 15 September, 1986. COMPARTMENT TREATMENT Cropped Uncropped Control duckweed* duckweed (no duckweed) PHOSPHORUS INPUT (g) (a) i n i t i a l water 14.1 13.3 14.8 (b) i n i t i a l duckweed 3.6 3.2 0 (c) pumped influent 55.8 55.8 55.8 (d) Total P input 73.5 72.3 70.6 PHOSPHORUS OUTPUT (g) (e) discharged water 51.5 52.7 54.4 (f) duckweed harvest 6.7 O 0 (g) remaining water 16.5 16.7 17.0 (h) remaining duckweed 2.2 7.9 0 (i) Total P output 76.9 77.3 71.4 (INPUT - OUTPUT) (g) -3.4 -5.0 -0.8 REMOVAL EFFICIENCIES (/) Overa11: E 0 = 100/.[(c-e)/c] 7.T/. 5.6/. 2.5/ Duckweed related: E d = 100/.[(f+h)-b]/(b+c) 8.9/ 8.0/ Non-duckweed related: E n = [E 0 - E d] -1.2/ -2.4/ 2.5/ Duckweed fraction: E d as / of E Q 115.6/ 142.9/ - Duckweed cropping rate was 0/ of the enclosed 2.25 m^  area per week. 112 establishment of a more significant detritivorous community in the root zone of the uncropped system does not appear to be indicated by the data. The higher cropping rate during Period 2 was associated with a slightly higher contribution by duckweed to P removal than during Period 1. 4.5 Mass Flux 4.5.1 Volatile solids mass flux Average influent VSS flux rates were higher during Period 1 than during Period 2 (Table 27). Discharge from the control was unchanged from the loading rate during Period 1. Statistically significant decreases were obtained however in the presence of duckweed, lower VSS discharge rates being recorded in the cropped than in the uncropped treatment. Evidently some factor in the cropped unit, such as less detritus from the harvested mat, enhanced VSS removal. During Period 2, the control exhibited a significant increase in VSS mass discharge rate relative to the loading rate. The lowest VSS discharge rate was obtained from the uncropped duckweed treatment, although the rate was not significantly different from the slightly higher value obtained from the cropped treatment. This result indicates that there was no measurable release of particulates from the unharvested duckweed mat despite the larger plant density and the larger fraction of detrital (dead and decomposing) biomass. Possible explanations include the consumption of detritus by macro-invertebrates which would not be represented in the water samples, entrapment of detritus within the living mat, and/or a climatological effect. 113 Comparison of VSS discharge rates between periods is difficult due to the difference in the corresponding loading rates. Common to both periods is the greater reduction in VSS flux through the ducKweed-covered systems, probably due to shading of phytoplanKton. The DO and ammonia concentrations suggest that consumtion of organic particulates by zooplankton would not necessarily have been enhanced in the duckweed units, although zooplankton was commonly observed in the influent and control. In the duckweed-free unit, there was no change in VSS flux during Period 1, and a net production of VSS during Period 2 despite the lower influx. This change may be the result of an increase in both temperature and available sunlight (less cloudiness) during the second half of the summer. 4.5.2 COD mass flux The COD loading rates during Period 1 were higher than during Period 2 Table 27), as was the case with VSS. During the first period, the lowest COD discharge rate was measured in the cropped duckweed treatment, although there was no significant difference between the latter rate and that in the uncropped treatment. The discharge rate from the control was significantly higher than that from the duckweed units, though still significantly lower than the loading rate. During Period 2, all duckweed systems exhibited significant decreases in COD mass flux, while the control exhibited a significant increase. The decrease in the cropped duckweed treatment was only slightly higher than that recorded in the uncropped treatment. Cropping had no significant effect on the COD discharge rate of the duckweed treatments, although replicate B (cropped) did exhibit a significantly lower discharge rate than replicate A and the uncropped unit. 114 The decrease in COD mass flux through all experimental units during Period 1 indicates that oxidative conditions existed throughout, and/or, that organic particulates were being removed from the water. The VSS data suggest that removal of particulates was the dominant process in the presence of ducKweed, while oxidation (see also nitrogen mass flux) predominated in the control. Conditions appeared to differ in the control (algae-dominated system) during Period 2, a net production of chemically oxidizable matter (i.e greater primary productivity via phytoplankton) being recorded, possibly due to more favourable weather. In the duckweed systems during the latter period, the result that there was usually no significant difference in COD discharge rates due to presence or degree of cropping, coupled with the significant differences recorded for VSS flux rates (as discussed above), suggest that the duckweed mat contributed more dissolved than suspended organic matter to the water column. 4.5.3 T o t a l nitrogen The total-N loading rate was significantly greater during Period 1 than during Period 2 (Table 27). Discharge rates via the effluents were in all cases significantly lower than the loading rates. During Period 1, the total-is discharge rates were not significantly different between treatments. The lowest total-N discharge rate was obtained in the uncropped, followed by the cropped and control units. During Period 2, duckweed presence yielded both an increase and a decrease in total-N mass flux, while that from the control was significantly lower, than the loading rate. The lowest total-N discharge rate during this period was obtained in the control, followed by tanks A (cropped), C 115 (uncropped) and B (cropped). There was no significant difference in discharge rates between duckweed units. The uncropped treatment and one cropped replicate (tank B) exhibited significantly higher total-N discharge flux than the control, due to elevated ammonia; the difference between the cropped replicates is also related to the ammonia fraction. 4.5.4 Ammonia mass flux During Period 1, ammonia discharge rates decreased relative to the loading rate in all units, and most so in the control, followed by the uncropped and cropped duckweed treatments (Table 27). These data indicate that ammonia removal (via plant uptake, nitrification, and possibly volatilization) was taking place more rapidly in the absence than in the presence of duckweed. The ammonia discharge rates in the duckweed units were significantly higher under cropped than under uncropped conditions, suggesting that cropping reduced net ammonia removal capacity in some (unidentified) manner. During Period 2, ammonia discharge rates from one cropped unit (tank A) and from the control were lower than the loading rate. The difference between cropped and uncropped treatments was statistically significant, the uncropped treatment exhibiting the largest increase over the loading rate. The data imply that decomposition of organic matter in the uncropped mat was complete, since ammonia (more so than organic-N) was released into the water below the mat. Cropping (by preventing in situ decomposition) would be expected to yield a decrease in ammonia flux, although such a result was only recorded in tank A. Tank B exhibited a significant increase, possibly related to allochthonous imports. In the control unit, algal uptake and nitrification 116 Table 27. Mass flux rates of VSS, COD, N and P in the conmon influent to and effluents from the 2.25 m2 experimental enclosures (g/d). [Values in any row not sharing the same letter are significantly different by the Neuman-Keuls Multiple Range test (Zar, 1974).] Treatment* Parameter Period** Influent A B A&B C D VSS 1 15.12 a 8.18 b 7.52 b 7.85 b 9.40 C 15.00 a 2 12.41 a 7.70 b 7.35 b 7.53 b 6.60 b 17.08 C COD 1 52.19 a 29.98 b 30.02 b 30.00 b 30.54 b 38.74 c 2 32.29 a 31.10 b 27.76 c 29.43 be 30.61 b 41.84 d Total 1 7.63 a 6.01 b 5.93 b 5.97 b 5.94 b 6.18 b N i trogen 2 4.53 a 4.38 ac 4.59 a 4.49 a 4.57 a 4.16 be Ammonia 1 4.57 a 3.94 b 4.02 b 3.98 b 3.83 b 3.26 c N i trogen 2 2.74 a 2.79 b 3.02 c 2.90 be 3.30 d 1.97 e Nitrate* 1 0.09 a 0.16 b 0.11 b 0. 13 b 0.16 b 0.61 c N i trogen 0.03 a 0.03 a 0.03 a 0.03 a 0.03 a 0.03 a Organ i c 1 2.96 a 1.91 b 1.80 b 1.86 b 1.95 b 2.31 c N i trogen 2 1.76 a 1.57 b 1.55 b 1.56 b 1.24 c 2.16 d Total 1 1.49 a 1.21 b 1.23 b 1.22 b 1.22 b 1.36 c Phosphorus 2 1.31 a 1 .21 b 1.24 c 1.22 be 1.24 c 1.28 d Ortho-P04 A 1 1.07 a 0.93 b 0.95 b 0.94 b 0.94 b 1.01 c Phosphorus 2 1.21 a 1.16 b 1. 15 b 1.16 b 1.23 a 1.20 a * Treatment descriptions are as follows: A - cropped duckweed; B - cropped duckweed (replicate); A&B - cropped duckweed average; C - uncropped duckweed; D - no duckweed (control). ** Period 1: 2 June to 4 August, 1986; duckweed cropping @ 15X/week. Period 2: 4 Aug. to 15 Sept., 1986; duckweed cropping <s> 50X/week. # n i t r i t e plus nitrate. A dissolved 0-PO4. 117 yielded a reduction in ammonia flux (although accompanied, as might be expected, by an increase in organic-N). 4 .5 .5 Nitrate mass flux There was no significant difference in the nitrate discharge rates between the duckweed-covered units during Period 1. The nitrate discharge rate from the control increased however during this period relative to the loading rate (Table 27). These results suggest that, in the control, the rate of nitrification was greater than the rate of nitrate removal via phytoplankton uptake and, or, denitrification. The duckweed-covered systems exhibited a situation in which what little nitrate was loaded was being either absorbed by the duckweed or microbially denitrified. Both mechanisms were probably operative, as the low DO and elevated ammonia concentrations suggest that conditions in the presence of duckweed were not as favourable for nitrification. There was no significant effect on N O 3 flux attributable to the presence or absence of duckweed cropping. During Period 2, there was no significant difference between the N O 3 loading and discharge rates within any treatment (Table 27). The loading rate was however approximately 67X lower than during Period 1, reflecting a decrease in net nitrification in Cell 4 (from which the influent was drawn) during late summer. 4 .5 .6 Organic nitrogen mass flux The duckweed-covered and control units exhibited statistically significant decreases in the flux of organic-N during Period 1 relative to the loading rate (Table 27). The control exhibited the lowest organic-N discharge 118 rate, followed by the uncropped and cropped duckweed treatments; however the differences between discharge rates were not significant. During Period 2, the control exhibited a significant increase, while the duckweed treatments exhibited significant decreases in organic-N discharge rate. The lowest discharge rate was obtained from the uncropped unit during this period, there being a significant difference between the cropped and uncropped duckweed treatments (Table 27). The decreases in organic-N mass flux in the presence of duckweed indicate that either settling of organic particulates, mineralization (ammonification) or uptake of organic-N were taking place. The VSS and COD data suggest that settling of influent organic particulates was an important component of organic-N removal. The lower organic-N discharge rate and lower biomass density in the uncropped than cropped duckweed treatments during Period 2 suggest that the higher cropping rate may have diminished the algal shading effect of the duckweed mats. This explanation is also supported by the VSS and COD data. The reduction in the duckweed-free control during Period 1 is indicative of either mineralization or uptake of organic-N by planktonic algal and bacterial populations. The net increase in organic-N flux through the control during Period 2 was largely due to algal photosynthetic conversion of inorganic-N (ammonia and nitrate) to particulate organic-N (i.e. plankton). 4.5.7 Nitrogen removal by duckweed The average rate of N removal via duckweed harvesting was 0.14 g/m2.d at the 15X/week cropping rate and 0.31 g/m2.d at the 50X/week cropping rate (Table 28). These N harvest rates correspond to 4.1X and 15.4Z, respectively, of the loading rates during each period. It is interesting to 119 Table 28. Mass flux rates of nitrogen and phosphorus in the influent, effluent, and duckweed harvest. (g.m~2.d~1) Treatment Period Nitrogen Phosphorus Flux ( r 2 ) a Flux (r 2) Influent 1 b 3.13 (0.984) 0.62 (0.998) 2 C 1.98 (0.999) 0.59 (0.913) Cropped 1 2.44 (0.992) 0.53 (0.988) duckweed effluent 2 1.81 (0.998) 0.56 (0.999) Duckweed 1 d 0.14 (0.931) 0.03 (0.919) harvest 2 e 0.31 (0.982) 0.07 (0.984) Uncropped 1 2.39 (0.992) 0.50 (0.990) duckweed effluent 2 1.88 (0.999) 0.56 (0.999) Control 1 2.48 (0.991) 0.57 (0.994) effluent 2 1.74 (0.998) 0.57 (0.999) a coefficient of determination (Zar, 1974); see text. b 23 June to 4 August, 1986 c 4 August to 15 September, 1986 d duckweed cropped at 15X of enclosure surface area per week, e duckweed cropped at 50X of enclosure surface area per week. 120 note that the factor by which the duckweed N removal rate increased between periods (2.2-fold) was less than that by which the cropping rate was increased (3.3-fold). The reason for this is may be related to a decrease in standing crop (Fig. 6). 4.5.8 Total-phosphorus mass flux The total-P loading rates were higher during Period 1 than Period 2 (Table 27). During this period, all total-P discharge rates were significantly lower than the loading rate. The lowest discharge rate was obtained in the presence of duckweed, no difference being recorded between cropped and uncropped units. The total-P discharge rate from the control was significantly higher than in the duckweed treatments. During Period 2 as well, all total-P discharge rates were significantly lower than the loading rate. The lowest discharge rate was recorded in replicate A of the cropped treatment; replicate B and the uncropped unit both exhibited a slightly higher discharge rate. The discharge rate in the absence of ducKweed was significantly higher than that in the duckweed-covered treatments. The data for both periods indicate that the presence of duckweed was associated with significant removal of P, whereas in the control incorporation of P into plankton did not remove the mineral from the water column. Between periods there was no difference in total-P discharge rate from the cropped treatment, and only a slight increase in the uncropped duckweed treatment, despite the difference in loading rates. The latter increase was likely due to P release from detritus accumulating in the unharvested duckweed mat. The control exhibited a decrease in total-P discharge rate between periods, reflecting the lower loading rate during Period 2. 121 4.5.9 Dissolved orthophosphate The loading rate of dissolved ortho-P was lower during Period 1 than Period 2 (Table 27). During Period 1, ortho-P represented 71.8X of the total-P loading rate. The ortho-P discharge rates from all treatments were significantly lower than the loading rate. The lowest discharge rates were obtained in the duckweed treatments, there being no effect due to 15X weekly cropping, followed by the control. During Period 2, there was no significant difference between the ortho-P loading and discharge rates in the control and uncropped duckweed treatments, although the latter was higher than the loading rate. During this period ortho-P represented 92.4Z of the total-P loading rate. The lowest ortho-P discharge rate was obtained in the cropped duckweed treatment, followed by the control and the uncropped units. Discharge rates of ortho-P during Period 2 were all higher than during Period 1, reflecting the increased loading rate. The higher dissolved ortho-P fraction of total-P in all fluxes during Period 2 suggests that soluble P was being added to the influent (lagoon 4). Annual P loading from the sediments is known to occur in lagoons during summer once reducing conditions become established due to oxygen depletion (Cole, 1979; Hutchinson, 1957). The contribution of soluble P from decaying duckweed in the uncropped system was also evident. The greater difference in soluble P discharge rates between cropped and uncropped duckweed treatments during the latter period suggests that the 50/ weekly cropping rate was associated with a significantly lower release of P from the duckweed mat. 122 4.5.10 Phosphorus removal by duckweed The phosphorus removal rate by duckweed harvesting was 0.03 g/m2.d at the 15X/week, and 0.07 g/m2.d at the 50X/week cropping rates (Table 28). These rates represented 4.5/ and 12.0X, respectively, of the P loading rate during each period. Increasing the cropping rate by 3.33 times (from 15/ to 50/ per week) was associated with only a 2.67 times increase in the P removal rate by duckweed. The reason for this is would appear to be the lower standing crop that existed during the second period, as mentioned above for N. 4.6 Treatment E f f i c i e n c i e s The foregoing two sections have dealt with the mass balances and mass flux for each experimental period as a whole. At this juncture it is useful to examine the changes in treatment performance within each period in order to gain a better understanding of the treatment processes involved. This section examines the changes in treatment efficiency based on weekly mass removals of VSS, COD, IM and P. 4.6.1 Suspended solids Figure 20 illustrates the changes in volatile suspended solids (VSS) removal efficiency over time. The low initial efficiencies are associated with water sampling depth, as the samples were taken at <20 cm below the water surface and were contaminated by surface solids. During Period 1, after the first month (or start-up period) the cropped duckweed treatment was associated - 3 5 H 1 1 1 1 l | 1 1 1 1 1 1 1 I 1 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) F i g u r e 20. WeeKly changes in the mass removal efficiency of volatile suspended solids (VSS) in the common influent and in the effluents fromi the cropped ducKweed, uncropped ducKweed and control treatments. Cropping rate was 15X/weeK during Period 1 and 50X/weeK during Period 2. ro 1 2 ^ with an increase in VSS removal efficiency from 34/ to 50X, while the uncropped duckweed and control treatments exhibited a net decrease from 38/ to 29/, and from 9.17. to -21X, respectively, despite a temporary increase. The change from 15/ (Period 1) to 50/ (Period 2) weekly cropping was associated with a slight decrease in VSS removal efficiency. During Period 2, the cropped treatment exhibited a slight but constant decrease in VSS removal efficiency, from 48/ to 43/. The uncropped treatment exhibited during the same period a slight but constant increase in VSS removal efficiency from 2.97. to 33/. The control underwent a net decrease, from -21/ to -23/, although a minimum of -31/ was recorded on day 77. The difference between the cropped and uncropped treatments was likely due in part to contribution of solids to the water from an uncropped duckweed mat that was becoming more dense and containing an increasing proportion of decaying biomass. The development of phytoplankton in the control tank (see Fig. 12, days 49 to 77) contributed to negative VSS removal efficiencies. 4.6.2 Chemical oxygen demand Figure 21 illustrates the changes in COD removal efficiency over time. The cropped duckweed treatment exhibited the highest efficiencies, followed by the uncropped and the control units. There was a generalized trend, among treatments, of of a gradual decline in COD reduction efficiency throughout the experiment. During Period 1, the decline was least evident in the cropped units, from 42/ to 37/ (replicate A exhibiting a nearly constant 44/ COD removal efficiency). In the uncropped and control treatments, weekly COD removal efficiencies decreased from 50/ to 31/, and from 41/ to 18/, respectively, during this period. Per iod 1 P e r i o d 2 Time (days) Figure 21. Weekly changes in the mass removal efficiency of chemical oxygen demand (COD) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15/./week during Period 1 and 50'/./week during Period 2. 126 The change from 15/ to 50X duckweed cropping was associated with a decrease in COD removal efficiency. However, a causal relationship cannot be established, as the uncropped and duckweed-free units also exhibited lower efficiencies during Period 2. As with VSS, the COD average influent concentration was lower during Period 2 than during Period 1. This condition may have been reflected in the lower COD reduction efficiencies during Period 2 (except in the control). Such a result also suggests that the plants were contributing to the effluent COD load, and that a "background" level (or minimum possible concentration value) may have been reached , due to the release of matter from the plants into the water, (Dewante and Stowell, 1981). 4.6.3 N i t r o g e n Total nitrogen Figure 22 depicts the changes in total-N removal efficiency over time. Peak efficiency of 19X to 23X N mass removal was achieved in all treatments within four weeks after start-up, and sustained during the subsequent three weeks (most of Period 1). Highest efficiencies were obtained in the control, followed by the cropped and uncropped treatments. Over the following five to six weeks N removal in the control decreased to a new steady state of about 15X efficiency, which was sustained for the remainder of Period 2. The decline in N removal efficiency coincides, in the duckweed units with an increase in biomass density (Fig. 6) and, in the control, with an increase in the discharge of nitrate (Fig. 16). It thus appears that plant crowding was associated with a decrease in N uptake by duckweed and, in the absence of 2 5 > o Q) to o 2 0 -1 5 -1 0 -Perlod 1 Period 2 Time (days) Legend O CROPPED • UNCROPPED A CONTROL Figure 22. WeeKly changes in the mass removal efficiency of total nitrogen (total-N) in the common influent and in the effluents from the cropped ducKweed, uncropped ducKweed and control treatments. Cropping rate was 15X/weeK during Period 1 and 50X/weeK during Period 2. ro -v3 128 ducKweed (control), that the nitrate consumption rate (through algal uptaKe or denitrification) was much lower than the corresponding loading rate. The decline in N0 3 removal rate may be related to the planKton species composition associated with the high ammonia-to-nitrate ratio (Leonardson and Rip!, 1979). Another very real possibility is the contribution of N to all tanKs by aphids, snails and ducKs, and, or, by leaKage through the tanK walls (see Section 4.7). A m m o n i a Figure 23 depicts the changes in ammonia removal efficiency over time. Removal efficiencies increased rapidly in all treatments during the first five to six weeKs of Period 1. The control maintained a steady state efficiency level of approximately 31/ for the remainder of the experiment. During Period 1, the ducKweed-covered units exhibited peaK ammonia removal efficiencies approximately 17/, although there was no significant difference between cropped and uncropped treatments. Steady-state conditions were not achieved in the ducKweed units during this period, possibly due to the contribution of ammonia from the biomass to the water, and efficiencies by the end of the period were approximately 5/ for both treatments. During the first half of Period 2, all ducKweed units exhibited a decline in ammonia removal efficiency, while the control maintained a similar performance as during Period 1. Ammonia removal efficiency in the cropped treatment declined to almost zero by day 84, but recovered somewhat thereafter, reaching 2/ by the end of the period. In the absence of cropping, ammonia removal efficiency declined throughout the period; net ammonia export 40 > o 35 H 30 25 o5 20 cn w 15 10 5 0 - 5 - i 0 Period 1 6 X p Period 2 V • 1 1 1 1 1 1 I I •—D 1 1 1 1 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend O CROPPED • UNCROPPED A CONTROL Figure 23. WeeKly changes in the mass removal efficiency of total ammonia nitrogen (NH4-N) in the common influent and in the effluents from the cropped ducKweed, uncropped ducKweed and control treatments Cropping rate was 15X/weeK during Period 1 and 50X/weeK during Period 2. 130 (i.e. negative removal) was evident after day 70 and continued for the duration of the experiment. It is evident that, despite the variation in influent NH4-N concentrations (Fig. 15), the algal (control) system was more effective at removing ammonia from the water. As discussed in the following section, nitrification appears to have been an important process in this regard. Increasing NH4-N concentrations were evident in the ducKweed covered tanks during the latter half of Period 1; this suggests that release of ammonia from the ducKweed mat (including the associated macro- and micro-fauna) was responsible for the decline in ammonia removal efficiency. The change from 15X to 50/ weeKly cropping did appear to have a positive effect on ammonia removal efficiency; the mechanisms involved may have included increased uptaKe, increased nitrification, and, or, decreased deamination from decomposing biomass. The onset of the decline in ammonia removal efficiency in the cropped units was not associated with the change from 15Z to 50X cropping. It did however coincide with a slight increase in nitrate removal efficiency (Fig. 24), possibly suggesting conversion of nitrate to ammonia (via algal or bacterial uptaKe and subsequent decay). The latter explanation cannot however account for all of the decline in ammonia removal efficiency due to the large differences in N mass involved. These data, coupled with the results of the mass balance calculations (presented above) indicate that there was an additional source of ammonia influx to the tanks. Possible sources of additional N are discussed in Section 4.7. 131 Nitrate plus nitrite Figure 24 illustrates the changes in N O 3 removal efficiency over time. Positive removal efficiencies were obtained in all units only during the first two weeks after start-up. During Period 1, all treatments exhibited a marked decline in N O 3 removal, to negative levels in all duckweed units and, in the control, to -750/ efficiency. These results provide evidence of considerable nitrification, an explanation that is also supported by the corresponding increase in ammonia removal efficiencies in all tanks during the same interval (Fig. 23, days 7 to 35). During Period 2, N 0 3 removal efficiencies in all units steadily increased, though remaining at negative levels. No measurable differences in N O 3 removal efficiency could be attributed to the change in cropping rate although nitrate production was less in the cropped than in the uncropped treatments. Though there was a marked difference due to presence or absence of duckweed, parallel trends were observed in all treatments, suggesting that N O 3 removal performance was also influenced by factors that were common to all experimental units, such as influent water quality and seasonal effects. Organic nitrogen Figure 25 depicts the changes in organic-N removal efficiency over time. Semi steady-state conditions were established after about 5 weeks, though only in the duckweed units. The control peaked after a similar interval, but declined thereafter. Organic-N removal efficiency of the cropped duckweed units increased during the remainder of Period 1, from the low to the high 30 percentile level, while the uncropped treatment exhibited a relatively 100 50 0 - 5 0 - 1 0 0 -- 1 5 0 -- 2 0 0 -- 2 5 0 -- 3 0 0 -- 3 5 0 -- 4 0 0 -- 4 5 0 -- 5 0 0 -- 5 5 0 - 6 0 0 H - 6 5 0 - 7 0 0 H - 7 5 0 Period 2 o — o — o — o — o - a — o - a — • \ A - A - — A -A — A ' - - A A \ / i 7 n i r \ r T 1 r 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) Legend O CROPPED • UNCROPPED A CONTROL uj Figure 24. Weekly changes in the mass removal efficiency of nitrite plus nitrate nitrogen (NO2+NO3-N) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15Z/week during Period 1 and 50Z/week during Period 2. 133 constant efficiency of approximately 28X. The control underwent a marked reduction in efficiency during the start-up interval (exhibiting negative organic-N removal efficiencies); subsequently, the efficiency increased to a peak of about 31/ during week 6, declining thereafter to approximately 15/ by the end of the period. The decrease in the control is very likely a reflection of the organic-N content of the phytoplankton bloom that developed in the absence of duckweed. During Period 2, the efficiency of the cropped units decreased gradually, from approximately 34% to 28X, attaining similar levels to the uncropped treatment (which showed no change between periods) by the end of the experiment. The control exhibited a decreasing trend during the first half of the period, but appeared to achieve a steady state condition at approximately 6/ organic-N removal efficiency during the final three weeks of the experiment, once the plankton bloom had subsided. Considering the aforementioned excess of N in the duckweed units, it is interesting that the organic-N removal efficiencies remained quite stable during most of the experiment. This suggests that the duckweed was able to absorb the excess N input which, if consisting of animal excretions, would largely consist of organic-N. it is possible that the decline in organic-N removal in the cropped treatment after day 49 (Fig. 25) reflects the increasing contribution of organic-N from ducks or other fauna. > o E cn V) V) O I o 'c D CJ) o 40 35 30 25 20 15 10 5 0 - 5 -10 H -15 A~. \ T 7 •A A - . A — A 1 1 1 1 1 1 1 1 1 I 1 1 1 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) L e g e n d O C R O P P E D • U N C R O P P E D A C O N T R O L Figure 25. Weekly changes in the mass removal efficiency of organic nitrogen (organic-N) in the common influent and in the effluents from the cropped ducKweed, uncropped duckweed and control treatments. Cropping rate was 15/./week during Period 1 and 50X/week during Period 2. 135 4.6.4 Phosphorus Total phosphorus Figure 26 depicts the changes in weeKly total-P removal efficiency over time. Higher removal efficiencies were achieved in the presence than in the absence of ducKweed. During Period 1, peaK efficiencies were obtained after six to seven weeKs, levels of 12/. to 16/ being attained in the ducKweed units and 12*/ in the control. Total-P removal efficiencies decreased in all units during the remainder of the period. The change from 15/ to 50/ weeKly cropping was associated with a decrease in the rate at which the total-P removal efficiency declined in the ducKweed units. The curve for the uncropped unit suggests that the absence of croppin was associated with a continuous decrease in P removal efficiency compared to the relatively constant levels in the cropped and control units. By the end of Period 2, a semi steady-state total-P removal efficiency level of about 9/ was achieved in the cropped treatment, and of <5Z in the uncropped and control units. It is evident that, although the presence of ducKweed was associated throughout the experiment with a marKedly higher total-P removal performance, both the ducKweed-covered and ducKweed-free units were affected similarly by common external factors, resulting in parallel trends in total-P removal efficiency. The most liKely common factor was the increase in influent dissolved ortho-P throughout most of the experiment (Fig. 19). The trends of ammonia (Fig. 23) and total-P removal efficiency are quite similar for the ducKweed-covered systems, suggesting that similar mechanisms may have been operative in the removal of both nutrients. 18 Legend O CROPPED • UNCROPPED A C O N T R O L Time (days) Figure 26. Weekly changes in the mass removal efficiency of total phosphorus (total-P) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15Z/week during Period 1 and SOX/week during Period 2. 137 Dissolved orthophosphate Figure 27 illustrates the changes in weeKly ortho-P removal efficiency over time. MarKed increases in influent ortho-P concentrations during the fi r s t weeK (Fig. 19) resulted in net ortho-P export (i.e. negative removal efficiencies) in all treatments. Ortho-P removal efficiencies during Period 1 peaKed in all units after six weeKs, slightly higher levels being reached in the cropped (10X) than in the uncropped (6X) ducKweed treatment. The control achieved a slightly higher maximum efficiency level VT'/.) than the uncropped ducKweed (&'/.), though the difference between these treatments was not marKed during the remainder of the period. During Period 2, there was a marKed decline in dissolved ortho-P removal efficiency in both ducKweed treatments. The cropped treatment attained a relatively constant ortho-P removal efficiency of approximately 4.5% during the latter half of the period. The uncropped treatment exhibited a steady decline throughout the period, net production of ortho-P being in evidence during the final four weeKs of the experiment. The control also declined initially, but achieved a steady state of approximately AY. ortho-P removal efficiency during the latter half of the period. The above results suggest that more dissolved P was released from the uncropped and 15/ cropped treatment than from the 50% cropped ducKweed and control treatments. > o E <D Cxi (A V) O Q_ I o Q_ 12-10-8 -6 4 2 0 - 2 - 4 - 6 - 8 -10 -12 Period 1 r J Period 2 / / / / Legend O CROPPED • UNCROPPED A CONTROL 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Time (days) 00 Figure £7. Weekly changes in the mass removal efficiency of dissolved orthophosphate (ortho-P) in the common influent and in the effluents from the cropped duckweed, uncropped duckweed and control treatments. Cropping rate was 15X/week during Period 1 and 50X/week during Period 2. 139 4.7 Sources of Unmeasured Nutrient Inputs Theoretically, the sum of the nutrient discharge via the effluent plus the nutrient harvest should be equal to the loading rate. Any discrepancies would indicate unmeasured sources or sinks. During Period 2, surpluses of N and P were measured in all but one instance, the latter being N in the control treatment. The N surplus in both duckweed treatments was similar (2.8 g and 2.9 g; Table 24). The P surplus (Table 26) was more variable, ranging from 0.8 g in the control to 5.0 g in the uncropped duckweed treatment. These results strongly imply that the amounts of N and P entering the cropped duckweed units were not quantified in their entirety. Undoubtedly, a certain fraction of the surplus mass can be accounted for by random or experimental error. While that fraction can only be quantified through the use of many replicates per treatment (more than were used in the present experiment), it is entirely possible that all of the recorded discrepancy could have been the result of random error. Nevertheless, the duckweed-covered units consistently exhibited N and P surpluses, while the control exhibited only a slight P surplus and a net N loss. Unmeasured N inputs were contributed to all experimental tanks in the form of animal droppings, invertebrate immigration via the influent or over the tank rims. As mentioned in the literature review, biological fixation of atmospheric Ng is also possible in duckweed systems (Zuberer, 1982). Unmeasured P imports could also have come from animal excretion and immigration. The possibility of leakage through the tank walls was considered 140 as well, since there was evidence that the fabric was permeable to gases evolved from the pond sediments. These potential sources of additional N and P are examined in the following paragraphs. Bird droppings are Known to contain considerable nitrogen in the form of uric acid (Schmidt-Nielsen, 1979). It must be assumed that P as well would be present in the feces in significant amounts. The experimental site was used by ducKs as a resting area (Hudson, pers. comm.), and ducK droppings were observed on the flotation collars of all experimental tanKs. All tanKs were equipped with a wire screen cover designed to prevent access by ducKs to the water or ducKweed. However, the screens did not prevent the entrance of droppings from birds perched on the screen or on the tanK flotation collars. Immigration of snails, aphids and other biota could theoretically contribute N and P to the experimental enclosures. The potential contribution of N and P by macro-invertebrate immigration is not considered to have had as much of an impact on nutrient budgets due to the fact that most of the nutrients contained in, and excreted by, such animals as snails and aphids would have originated from within the enclosures. The presence of snail egg masses, as well as various aphid instars (stages of metamorphic development) confirm that reproduction was taKing place within the ducKweed tanKs. Nevertheless, further studies would be needed to determine whether or not immigration of snails, aphids and other invertebrates could have contributed significantly to N and P loading. Biological fixation of atmospheric nitrogen is another possible source of unmeasured N inputs, though not of P. The cycling of atmospheric Ng to organic matter has been measured in ducKweed mats by Zuberer (1982). Excess 14-1 nitrogen (i.e. higher input than output) has been reported from experimental ducKweed cultures growing on sewage (Oron et al., 1984), dairy barn wastewater (Whitehead et al., 1986) and coal mine drainage (Norecol, 1986). It is unliKely that leaKage through the fabric walls or bottom of the tanKs would have been a significant factor contributing to excess N and P in the ducKweed tanKs. Biogas bubbles were frequently observed breaKing the surface both inside and outside all enclosures. The bubbling pattern within the tanKs was suggestive of a slow leaK through the bottom fabric, and raised doubts about the impermeability of the tanK walls. However, since the potential contribution of nutrients from leaKage would have affected all tanKs equally, regardless of whether or not ducKweed was present, this factor can be discarded as a significant possibility. Based on the evidence available, it is not possible to explain with absolute certainty the causes of additional N and P import to the ducKweed tanKs. Certainly, all of the potential causes considered could have been involved. The most liKely cause appears to have been the addition of ducK excreta, since the nutrients contained therein would have originated from without the enclosures. Further research would be necessary to determine the relative importance of the allochthonous nutrient sources hypothesized above. It is clear that a clearer understanding of N and P (and other nutrient) dynamics in ducKweed systems is necessary before conclusions can be drawn regarding the full potential or limitations of using ducKweed for tertiary treatment. 142 On a full scale lagoon, the potential impact of faunal nutrient imports on the overall nutrient budget would be considerably less than that reported here for smaller-scate experimental systems. This hypothesis is based on the premise that in a full-scale system, the source of the "imports" would be the lagoon itself (i.e. the nutrients would be cycling from the water to plants to ducks/snails/aphids to excretion products and back into solution). The greater complexity of the duckweed ecosystem, in terms of the diversity and mobility of associated grazing organisms, would appear to provide greater opportunities for nutrient removal from the water, be it via the natural food web or through controlled biomass harvesting. 1^3 5. SUMMARY AND CONCLUSIONS The following conclusions can be drawn from the study: 1. Removal of suspended solids, chemical oxygen demand, total nitrogen and total phosphorus from sewage stabilization pond water was greater in the presence than in the absence of ducKweed. Removal of inorganic-N however was greater in the absence of ducKweed during warm sunny weather. 2. Cropping of ducKweed at a rate of 15%/weeK improved treatment performance relative to the uncropped condition, only in the case of VSS. Cropping at 50%/weeK improved treatment performance in the case of inorganic-N (ammonia-N, in particular) and dissolved ortho-P. Differences in influent quality, and possibly weather, between cropping periods precluded comparison of treatment performance on the basis of cropping rate alone. 3. Cropping of ducKweed marKedly increased the fraction of total-N and total-P removed via plant uptaKe. During Period 1, uncropped ducKweed accounted for 3.7%, and cropped ducKweed for 6.9% of the overall N mass removal. In the case of P, uncropped ducKweed accounted for 3.5%, and cropped ducKweed for 7.2% of the overall P mass removal. During Period 2, the corresponding values for ducKweed-related N mass removal were 6.4% and 12.6%, respectively, under uncropped and cropped conditions. The corresponding values for ducKweed-related P mass removal during Period 2 were 8.0% and 8.9%, respectively. 144 4. During the first experimental period (2 June to 4 August) the presence of ducKweed appeared to cause changes in the nutrient removal mechanisms not related to ducKweed. Cropping (at 15X of the enclosure surface area per weeK) was associated with a decrease in non-ducKweed nitrogen removal, relative to the uncropped and control treatments. There was no difference in non-ducKweed N mass removal between the uncropped ducKweed treatment and the control during this period. Non-ducKweed P removal increased in the uncropped treatment and decreased in the cropped treatment relative to the ducKweed-free control. 5. During the second experimental period (4 August to 15 September) the presence of uncropped ducKweed was associated with a decrease in non-ducKweed N and P removal relative to the control. Quantification of non-ducKweed N removal in the cropped ducKweed treatment during the latter period was not possible due to allochthonous nutrient additions, while non-ducKweed P removal was highest in the cropped ducKweed treatment, there being little difference between the uncropped and control treatments. 6. WeeKly ducKweed cropping of 15X of the enclosure surface area was insufficient to remove all of the weeKly biomass production, resulting in increased standing crop over time. A sustainable standing crop of approximately 80 g/m2 (dry weight) was obtained when the cropping rate was increased to 50Z/weeK. The uncropped ducKweed population attained a maximum standing crop of 230 g/m2 (dry weight) and a steady state condition of approximately 140 g/m2 (dry weight) during the late summer. 14-5 7. Sustainable duckweed yields were obtained at both cropping rates, despite a severe infestation of aphids. Dry matter yields of 2.0 g/m2.d and 6.4 g/m2.d were obtained at the 15X and 50X weekly cropping rates, respectively. 8. Duckweed dry matter nitrogen content ranged from 5.8X to 6.5X. Differences in biomass nitrogen content were not associated with presence or absence of cropping, nor with differences in cropping intensity. Phosphorus content ranged from 1.2X to 1.4X of dry matter; P content was higher in the uncropped population by the end of the experiment. Crude protein content ranged from 36.4X to 42.5X of the duckweed dry matter. Dry matter content of fresh duckweed ranged from 5.9X to 7.0X of fresh weight. 9. Duckweed harvest removed 0.14 g N /m2.d and 0.03 g P /m2.d at the 15X/week cropping rate and 0.31 g N /m2.d and 0.07 g P /m2.d at the 50X/week cropping rate. The crude protein production rate averaged 0.88 g/m2.d at 15X/week cropping rate and 1.94 g/m2.d at 50X/week cropping. 10. Duckweed yields based on 50X/week cropping and a 120 day growing season were estimated to average 7.6 tonnes/ha dry weight and 122.6 t/ha fresh weight, representing 370 kg N/ha, 84 kg P/ha and 16 t crude protein/ha. 146 6. RECOMMENDATIONS The results of this study confirm, for British Columbia conditions, the elevated productivity, though not the N and P treatment capabilities of ducKweed, Lemna minor, grown on municipal wastewaters. Contamination of the experimental enclosures with N and P from waterfowl excreta precludes at this time the quantification of the treatment effect of ducKweed on water quality. Further research is required therefore to ascertain whether or not the high nutrient removal capability demonstrated in this study translates into acceptable tertiary treatment. To develop guidelines for employing ducKweed culture as a functional unit process, specific areas of study are recommended as follows: - Conduct a full-scale trial involving cropping of ducKweed from a small sewage lagoon (<1 ha), with the objective of obtaining representative effluent water quality data on nitrogen and phosphorus concentrations. - Evaluate harvesting techniques and wind-control measures for full-scale lagoons. - Assess the re-use potential of the harvested ducKweed biomass, post-harvest processing methods, and associated public health risKs. Assess the economic aspects of duckweed culture as related to wastewater treatment, with the aim of quantifying the potential contribution of resource recovery to reducing wastewater treatment costs. 148 7 L I T E R A T U R E C I T E D Amborski, R.L. and J.M. Larkin. 1980. Human and animal health aspects, in: J.B. Frye, Jr. and D.D. Culley, Jr. (eds.). A feasibility study of the conversion of animal feedlot wastes to useful energy in Louisiana. Unpublished Final Report 1978 - 1980, to the Louisiana Department of Natural Resources, and the U.S. Department of Energy, Agriculture and Food Processes Branch. Grant No. (EM-78-G-01-5261). American Public Health Association (A.P.H.A.), American Water Works Association, and Water Pollution Control Federation. 1985. Standard Methods for the Examination of Water and Wastewater (14th Edition). American Public Health Association, 1015 Eighteenth Street NW, Washington, D.C, 20036. 1193p. Angerelli, N.P.D.C. and B.P. Beirue. 1974. Influence of some aquatic plants on the development and survival of mosquito larvae in British Columbia. Can. J. Zool., 52(7):813-815. Arber, A. 1963. Water Plants - A Study of Aquatic Angiosperms. Weldon and Wesley, Ltd., Codicote, Herts., and Hafner Publishing Co., New York, N.Y. 436p. Baker, J.H. and I.S. Farr. 1982. Organic secretions by duckweed and their relationship with epiphytic bacteria. in: M.E. Rhodes and F.A. Skinner (eds.). Bacteria and Plants (The Society for Applied Bacteriology Symposium Series No. 10). Academic Press. 264p. Bell, H. 1984. Natural products - chemical warfare among algae. The Seiche, Vlll(3):1. Bieleski, R.L. 1968. Effect of phosphorus deficiency on levels of phosphorus in Spirodela. Plant Physiol., 43:1309-1316. Bhantumnavin, K. and M.G. McGarry. 1971. Wolff ia arrhiza as a possible source of inexpensive protein. Nature, 232(5311):495. Blackman, G.E., G. Sen, W.R. Birch and R.G. Powell. 1959. The uptake of growth substances.l. Factors controlling the uptake of phenoxyacetic acids by Lemna minor. J. Exp. Bot., 10(28):33-54. Blackman, G.E. and J. A. Sargent. 1959. The uptake of growth substances. II. The absorption and accumulation of 2:3:5-tri-iodobenzoic acid by the root and frond of Lemna minor. J. Exp. Bot., 10(30):480-503. Blair, Russell. (1986). President, Slickbar, Inc., Southport, Connecticut. Personal communication. Burton, J.H., A.J. Mutzar, S.J. Slinger and J.H. Neil. 1977. Utilization of aquatic plants for animal feeds. Proc. 17th Annual Meeting, Aquat. Plant Manage. Soc, Minneapolis, Minnesota, July 17-20, 1977. 18p. 149 Caldwell, D.H., D.S. Barker, W.R. Uhte and R.J. Renquist. 1977. Upgrading Lagoons. EPA-625/4-73-001D. U.S. Environmental Protection Agency, Washington, D.C. 67p. Clatworthy, J.N. and J.L. Harper. 1962. The comparative biology of closely related species living in the same area. V. Inter- and intraspecific interference within cultures of Lemna spp. and Salvinia natans. J. Exp. Bot., 13(38):307-324. Cole, G.A. 1979. Textbook of Limnology (Second Edition). CV. Mosby Co., St. Louis. 426p. Coler, R.A. and H.B. Gunner. 1971. The response of a specialized aquatic ecosystem, the duckweed rhizosphere, to selected environmental influences. Water Res., 5:329-333. Craker, L.E. 1972. Influence of ozone on RNA and protein content of Lemna minor L. Environ. Pollut., 3:319-323. Culley, Dudley, Jr. 1984. Professor, Department of Forestry and Wildlife Management, Louisiana State University, Baton Rouge, Louisiana. 70803. Personal communication. and E.A. Epps. 1973. Use of duckweed for water treatment and animal feed. J. Water Pollut. Contr. Fed., 45(2):337-347. and R.W. Myers. 1980. Effect of harvest rate on duckweed yield and nutrient extraction in dairy waste lagoons, in: J.B. Frye, Jr. and D.D. Culley, Jr. (eds.). A feasibility study of the conversion of animal feedlot wastes to useful energy in Louisiana. Unpublished Final Report 1978 - 1980, to the Louisiana Department of Natural Resources, and the U.S. Department of Energy, Agriculture and Food Processes Branch. Grant No. (EM-78-G-01-5261). , E. Rejmankova and J. Kvet. 1981. Production, chemical quality and use of duckweeds (Lemnaceae) in aquaculture, waste management and animal feeds. (Priviledge Communication, Rough Draft in review, 1981, World Mariculture Society). Daborn, G.R., J.A. Hayward and T.E. Quinney. 1978. Studies on Daphnia pulex Leydig in sewage oxidation ponds. Can. J. Zool., 56(6):1392-1401. Dale, H.M. and T. Gillespie. 1976. The influence of floating vascular plants on diurnal fluctuations of temperature near the water surface in early spring. Hydrobiologia, 49(3):245-256. Datko, A.H., S.H. Mudd, P.K. MacNicol and J. Giovanelli. 1978a. Phytostat for the growth of Lemna in semicontinuous culture with low sulfate. Plant Physiol., 62:622-628. 150 Datko, A.H., S.H. Mudd, J. Giovanelli and P.K. MacNicol. 1978b. Sulfur-containing compounds in Lemna perpusilla 6746 grown at a range of sulfate concentrations. Plant Physiol., 62:629-635. DeBusk, T.A., J.H. Ryther, M.D. Hanisak and L.D. Williams. 1981. Effects of seasonality and plant density on the productivity of some freshwater macrophytes. Aquat. Bot., 10:133-142. , J.H. Ryther and L.D. Williams. 1983. Evapotranspiration of Eichornia crassipes (Mart.) Solms and Lemna minor L. in central Florida: relation to canopy structure and season. Aquat. Bot., 16:31-39. de Jong, J. 1976. The purification of wastewater with the aid of rush or reed ponds, in: J. Tourbier and R.W. Pierson, Jr. (eds.) Biological Control of Water Pollution. Univ. Pennsylvania Press, pp.133-139. deLange, L. 1975. Gibbosity in the complex Lemna gibba/Lemna minor'. literature survey and ecological aspects. Aquat. Bot., 1:327-332. Derksen, G. 1981. The effects of a sewage lagoon effluent on the water quality of the Cowichan River during the 1980 low flow period. Regional Program Rep. 82-5. Environmental Protection Service, Pacific Region, Environment Canada. 83p. Dewante and Stowell Consulting Engineers, Inc. 1981. An investigation of the potential of five aquatic systems for ammonia control and effluent polishing. Unpublished Final Report to the City of Roseville, California, January, 1981. 17p. Dinges, R. 1974. The availability of Daphnia for water quality improvement and as an animal food source. Proc. Conf. on Wastewater Use in the Production of Food and Fiber, EPA-660/2-74-041. U.S. Environmental Protection Agency, Washington, D.C. pp.142-161. 1976a. A proposed integrated biological wastewater treatment system, in: J. Tourbier and R.W. Pierson, Jr. (eds.) Biological Control of Water Pollution. Univ. Pennsylvania Press, pp.225-230. 1976b. Who says sewage treatment plants have to be ugly? Water Wastes Eng., 13(4):20-33. 1978. Aquatic vegetation and water pollution control: public health implications. Proc. 53rd Annual Meeting, Texas Public Health Association. 19 Feb., 1978. 13p. 1982. Natural Systems for Water Pollution Control. Van Nostrand Reinhold Co., Toronto. 252p. Duncan and Associates Engineering, Ltd. 1983. Sewage treatment and disposal options. Preliminary report to the Joint Utilities Board of the City of Duncan and the District of North Cowichan. December, 1983. 151 Dykyjova, D. and J. Kvet (eds.). 1978. Pond Littoral Ecosystems - Structure and Functioning. Springer-Verlag, New York, N.Y. 464p. Ehrlich, S. 1966. Two experiments in the biological clarification of stabilization pond effluents. Hydrobiologia, 27:70-80. Environmental Protection Agency (E.P.A.). 1979. Aquaculture Sytems for Wastewater Treatment: Seminar Proceedings and Engineering Assessment. EPA-430/9-80-006. U.S. Environmental Protection Agency, Washington, D.C. 485p. Eversull, H, A.S. Achacoso and R.H. Sorbet, Jr. 1981. Ensiling duckweed with high dry matter corn crop. Paper presented at the 77th Annual American Dairy Science Association. June 27-30, 1981. Pennsylvania State University. 4p. Fekete, A., D.N. Riemer and H.L. Motto. 1976. A bioassay using common duckweed to evaluate the release of available phosphorus from pond sediments. J. Aquat. Plant Manage., 14:19-25. Ferguson, A.R. and E.G. Bollard. 1969. Metabolism of Spirodela oligorrhiza. I. Utilization of ammonium, nitrate and nitrite. Planta, 88:344-352. Glandon, R.P. and CD. McNabb. 1978. The uptake of boron by Lemna minor. Aquat. Bot., 4:53-64. Hillman, W.S. 1957. Non-photosynthetic light requirement in Lemna minor and its partial satisfaction by kinetin. Science, 126(3264):165-166. 1961. The Lemnaceae or duckweeds: A review of the descriptive and experimental literature. Bot. Rev., 27:221-287. and D.D. Culley, Jr. 1978. The uses of duckweed. American Scientist, 66:442-451. Hodgson, G.L. 1970. Effects of temperature on the growth and development of Lemna minor under conditions of natural daylight. Ann. Bot., 34:365-381. Hossell, J.C and J.H. Barker. 1976. The distribution and characterization of bacteria on the surface of some river macrophytes. J. Appl. Bacterid., 41(3):xiv-xv. (abstract only). Hudson, Alan. 1986. Operator, Sewage treatment lagoons, City of Duncan, British Columbia. Personal communication. Hutchinson, G.E. 1957. A Treatise on Limnology, Vol. I. Geography, Physics and Chemistry of Lakes. John Wiley and Sons, New York, N.Y. p. 1975. A Treatise on Limnology, Vol. III. Limnological Botany. John Wiley and Sons, New York, N.Y. 660p. 152 Hutchinson, T.C. and H. CzyrsKa. 1975. Heavy metal toxicity and synergism to floating aquatic weeds. Int. Assoc. Theor. Appl. Limnol. Proc, 19(3):2102-2111. Joy, K.W. 1969. Nitrogen metabolism of Lemna minor. I. Growth, nitrogen sources and amino acid inhibition. J. Plant Physiol., 44:845-848. Kenny-Wallace, G. and G.E. Blackman. 1972. The uptake of growth substances. XIV. Patterns of uptake by Lemna minor of phenoxyacetic and benzoic acids following progressive chlorination. J. Exp. Bot., 23(7): 114-127. Landolt, E. 1981. Distribution pattern of the family Lemnaceae in North Carolina. Veroff. Geobot. Inst. Eth. (Proc. of the 16th I.P.E., Vol.3). Stiftung Rubel, Zurich. pp.112-148. Lemna Corporation, The. (undated). Lemna Wastewater Treatment Process: An Overview. (Promotional brochure). The Lemna Corporation, 7800 Main St. N.E., Minneapolis, Minnesota, U.S.A. 55432. Telephone (612) 571-1334. Lewis, W.M. and M. Bender. 1961. Effect of a cover of duckweed and the alga Pithophora upon the dissolved oxygen and free carbon dioxide of small ponds. Ecology, 42(3):602-603. Little, E.C.S. 1968. Handbook of Utilization of Aquatic Plants. PL:CP/20. Food and Agriculture Organization of The United Nations, Rome. 123p. Lumpkin, T.A. and D.L. Plucknett. 1980. Azolla\ botany, physiology and use as a green manure. Econ. Bot., 34(2):111-153. Mclnerney, M.J., M.P. Bryant and D.A. Stafford. 1979. Metabolic stages and energetics of microbial anaerobic digestion. Ch. 5 in: D.A. Stafford, B.I. Wheatley and D.E. Hughes (eds.). Anaerobic Digestion. Applied Science Publishers, Ltd., London, pp.91-98. McLay, CL. 1976. The effect of pH on the population growth of three species of duckweed: S pi rod el a oligorrhiza, Lemna minor and Wolff ia arrhiza. F reshwater Biol. 6:125-136. Metcalf and Eddy, Inc. 1979. Wastewater Engineering: Treatment, Disposal, Reuse. Second Edition. McGraw Hill Co., New York, N.Y. 920p. Middlebrooks, J.E. and A. Pano. 1983. Nitrogen removal in aerated lagoons. Water Res., 17(10):1369-1378. Morris, P.F. and W.G. Barker. 1977. Oxygen transport rates through mats of Lemna minor and Wolff ia sp. and oxygen tension within and below the mat. Can. J. Bot., 55(14):1926-1932. National Academy of Sciences (N.A.S.). 1976. Making Aquatic Weeds Useful: Some perspectives for developing countries. National Academy Press, Washington, D.C. 175p. 153 Ngo, Viet. (1986). President, The Lemna Corporation, 7800 Main St. N.E., Minneapolis, Minnesota, U.S.A. 55432. O'Brien, W.J. 1980. Engineering assessment of aquatic plant systems for wastewater treatment, in: Aquaculture Systems for Wastewater Treatment -An Engineering Assessment. EPA-430/9-80-007. U.S. Environmental Protection Agency, Washington, D.C. pp.63-80. Odum, H.T. 1971. Partnership with Nature. Ch. 10 in: H.T. Odum. Environment, Power and Society. Wiley-lnterscience, New YorK, N.Y. pp.274-303. Oron, G., L.R. Wildschutt, and D. Porath. 1984. Waste water recycling by duckweed for protein production and effluent renovation. Water Sci. Tech., 17:803-817. Ozimek, T. 1985. Heavy metals content in macrophytes from ponds supplied with post-sewage water. Symp. Biol. Hungarica, 29:41-50. Polprasert, C. and B. Sikka. 1986. Aquatic weeds and their uses - an overview of perspectives for developing countries. Environ. Sanitation Rev. (21). Asian Institute of Technology, Bangkok, Thailand. 79p. Porath, D. and Y. Ben-Shaul. 1971. Structural and physiological changes during "heat bleaching" in Spirodela oligorrhiza. Israel J. Bot., 20:157-168. and . 1973. Growth, greening and phytochrome in etiolated Spirodela (Lemnaceae). Plant Physiol., 51:474-477. and J. Pollock. 1982. Ammonia stripping by duckweed and its feasibility in circulating aquaculture. Aquat. Bot., 13(2):125-131. , G. Oron and G. Granoth. 1985. Duckweed as an aquatic crop: edible protein recovery, production and utilization, in: Agricultural Waste Utilization and Management. Proc. Fifth Int. Symp. on Agric. Wastes. Dec. 16-17, 1985, Chicago, Illinois. CSAE. pp.680-687. Power, E.A. and P.L. Wentzell. 1985. Experimental Study of Log Storage Impacts on Babine Lake - 1985: effects on water quality, bacteria, zooplankton and sockeye fry. Westwater Research Centre, Univ. of British Columbia, Vancouver. 79p. Prescott, G.W. 1978. How to Know the Freshwater Algae. Wm.C. Brown, Co. Publishers. Dubuque, Iowa. 293p. Rakocy, J.E. and R. Allison. 1979. Evaluation of a closed recirculating system for the culture of tilapia and aquatic macrophytes. Paper presented at the Bio-Engineering Symposium, American Fisheries Society, Fish Culture Section. Traverse City, Michigan, Oct. 15-18, 1979. 22p. Reed, S., R. Bastian, S. Black and R. Khettry. 1984. Wetlands for wastewater treatment in cold climates. Paper presented at the Water Reuse Symposium III, San Diego, California, Aug. 26-31, 1984. 9p. 154 Reid, M.S. and R.L. BielesKi. 1970. Response of Spirodela oligorrhiza to phosphorus deficiency. Plant Physiol., 46:609-613. Rodgers, J.H., D.S. Cherry and R.K. Guthrie. 1978. Cycling of elements in ducKweed (Lemna perpusilla) in an ash settling basin and swamp drainage system. Water Res., 12:765-770. Rusoff, L.L., E.W. BlaKeney, Jr. and D.D. Culley, Jr. 1980. DucKweeds (Lemnaceae family): a potential source of protein and amino acids. J. Agric. Food Chem., (28):848-850. Ruttner, F. 1963. Fundamentals of Limnology (Third Edition). Translated by D.G. Frey and F.E.J. Fry. Univ. of Toronto Press, Toronto. 295p. Said, M.Z.M., D.D. Culley, Jr., L.C. Standifer, E.A. Epps, R.W. Myers and S.A. Boney. 1979. Effect of harvest rate, waste loading, and stocKing density on the yield of ducKweeds. Proc. World Maricul. Soc, 10:769-780. Schmidt-Nielsen, K. 1983. Animal Physiology: adaptation and environment (Third Edition). Cambridge Univ. Press. 619p. Schumann, G.E., M.A. Stanley and D. Knudsen. 1973. Automated total nitrogen analysis of soil and plant samples. Proc. Soil Sci. Soc Amer., (37):480-481. Sculthorpe, CD. 1967. The Biology of Aquatic Vascular Plants. Edward Arnold (Publishers) Ltd. London. 610p. Seidel, K. 1976. Macrophytes and water purification, in: J. Tourbier and R.W. Pierson, Jr. (eds.). Biological Control of Water Pollution. University of Pennsylvania Press, pp.109-121. Serfling, S.A. and C. Alsten. 1979. An integrated, controlled environment aquaculture lagoon process for secondary or advanced wastewater treatment, in: Performance and Upgrading of Wastewater Stabilization Ponds. EPA-600/9-79-011. U.S. Environmental Protection Agency. Washington, D.C. pp.124-145. So, M.L. 1986. Growth characteristics of ducKweeds and their potential use as organic fertilizers in Hong Kong. Paper presented at the Conf. on Aquatic Plants for Water Treatment and Resource Recovery. Orlando, Florida, July 20-24, 1986. 14p. Stangenberg, M. 1968. Bacteriostatic effects of some algae and Lemna extracts. Hydrobiologia, 332:88-96. Stanley, R.A. and C.E. Madewell. 1975. Optimum dilution of swine wastes for growth of Lemna minor and Euglena sp. in: Managing Livestock Wastes. Proc. American Society of Agricultural Engineers, 275:331-333. 155 Stephenson, M. G. Turner, P. Pope, J . Colt, A. Knight and G. Tchobanoglous. 1980. The Environmental Requirements of Aquatic Plants. Appendix A of The Use and Potential of Aquatic Species for Wastewater Treatment. Publication No. 65, California State Water Resources Control Board, Sacramento, California. 655p. Stowell, R., R. Ludwig, J . Colt and G. Tchobanoglous. 1981. Concepts in aquatic treatment system design. J. Env. Eng. Div. - A.S.C.E., 107(NEE5):919-940. Stowell, R.E. et al. (in press). An Introduction to Aquatic Treatment Systems. Appendix E of The Use and Potential of Aquatic Species for Wastewater Treatment. Publication No. 65, California State Water Resources Control Board, Sacramento, California, p. Sundstrom, D.W. and H.E. Klei. 1979. Wastewater Treatment. Prentice-Hall, Inc., Englewood Cliffs, N.J. 444p. Sutton, S.L. and H.W. Ornes. 1975. Phosphorus removal from static sewage effluent using ducKweed. J. Environ. Qua I., 4(3):367-370. and . 1977. Growth of Spirodela polyrrhiza in static sewage effluent. Aquat. Bot., 3:231-237. TanaKa, O. and A. TaKimoto. 1978. Ef fect of nitrate concentration in the medium on the flowering of Lemna paucicostata 6746. Plant Cell Physiol. 19(4):701-704. Toerien, D.F., P.R. Cary, CM. Finlayson, D.S. Mitchell and P.G.J. Weerts. 1983. Growth models fo r Salvinia molesta. Aquat. Bot., 16:173-179. Tourbier, J . and R.W. Pierson, Jr. (eds.). Biological Control of Water Pollution. University of Pennsylvania Press, Philadelphia. 340p. Trewavas, A. 1972. Control of protein turnover rates in Lemna minor. Plant Physiol., 49:47-51. Wetzel, R.G. 1975. Limnology. W.B. Saunders Co., Philadelphia. 743p. Whitehead, A.J., K.V.Lo and N.R. Bulley. 1986. E f fec t of hydraulic retention time and ducKweed cropping rate on nutrient removal from dilute dairy manure. Paper presented at the Conf. on Aquatic Plants for Water Treatment and Resource Recovery. Orlando, Florida, July 20-24, 1986. 14p. Wile, I., G. Miller and S.A. BlacK. 1982. Design and use of artificial wetlands. Paper presented at the WorKshop on Ecological Considerations in Wetland Treatment of Municipal Wastewater. University of Massachusetts, Amherst, Massachusetts, June 23-25, 1982. 10p. Wolverton, B.C. 1979. Engineering design data for small vascular aquatic plant wastewater treatment systems, in: Aquaculture Sytems for Wastewater Treatment: Seminar Proceedings and Engineering Assessment. EPA-430/9-80-006. U.S. Environmental Protection Agency, pp.11-12. 156 and R.C. McDonald. 1979a. Energy from aquatic plant wastewater treatment systems. NASA Tech. Memo. M-X-72733, September 1979. 16p. and . 1979b. Upgrading facultative wastewater lagoons with vascular aquatic plants. J. Water Po/lut. Contr. F ed., 51(2):305-313. and . 1980. Secondary domestic wastewater treatment using a combination of duckweed and natural processes. Unpublished report, July 1980. National Space Technology Laboratories, NASA, Bay St. Louis, Mississippi. 16p. Zar, J.H. 1974. Biostatistical Analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. 620p. Zuberer, D.A. 1982. Nitrogen fixation (acetylene reduction) associated with duckweed (Lemnaceae) mats. Appl. Environ. Microbiol., 43(2):823-828. PUBLICATIONS The e f f e c t o f h y d r a u l i c r e t e n t i o n time and duckweed c r o p p i n g r a t e on n u t r i e n t removal from d a i r y barn wastewater, by A . J . Whitehead, K.V. Lo and N.R. B u l l e y . j_n: A q u a t i c P l a n t s f o r Water Treatment and Resource Recovery. K.R. Reddy and W.H. Smith (Eds.) M a g n o l i a P u b l i s h i n g I n c . 1987. pp.697-703. Methane p r o d u c t i o n a t 22° C o f l a b o r a t o r y - s c a l e f i x e d - f i l m r e a c t o r s r e c e i v i n g s c r e e n e d d a i r y manure, by K.V. Lo, P.H. L i a o and A . J . Whitehead. 1985. Energy i n A g r i c u l t u r e , 4:1-13. M e s o p h i l i c a n a e r o b i c d i g e s t i o n o f s c r e e n e d and unscreened d a i r y manure, by K.V. Lo, P.H. L i a o , A . J . Whitehead and N.R. B u l l e y . 1984. A g r i c u l t u r a l  Wastes. V o l . 11, No. 4: 269-283. Methane p r o d u c t i o n s c r e e n e d d a i r y manure u s i n g a f i x e d - f i l m r e a c t o r , by K.V. Lo, A . J . Whitehead, P.H. L i a o and N.R. B u l l e y . 1984. A g r i c u l t u r a l  Wastes V o l . 11, No. 9: 175-188. The e f f e c t o f l i q u i d - s o l i d s s e p a r a t i o n p r e t r e a t m e n t on b i o g a s p r o d u c t i o n from d a i r y manure, by K.V. Lo, N.R. B u l l e y , P.H. L i a o and A . J . Whitehead. 1983. A g r i c u l t u r a l Wastes V o l . 10, No. 9: 155-165. La P i s c i c u l t u r a : Una nueva f u e n t e de p r o t e i n a animal ( F i s h C u l t u r e : A new s o u r c e o f animal p r o t e i n ) , by A . J . Whitehead. Informe Primer S e m i n a r i o N a c i o n a l de S a l u d R u r a l , Q u e z a l t e n a n g o , 3-7 D i c i e m b r e , 1979. Documento No. 4. Guatemala ( I n S p a n i s h ) . I n f o r m a c i o n General Sobre La P i s c i c u l t u r a En Guatemala, by A . J . Whitehead. 1979. DECA/DIGESA, M i n i s t e r i o de A g r i c u l t u r a , Guatemala. 27p. Informe sobre l a fauna s i l v e s t r e de l a Hacienda E l Prado. by A . J . Whitehead. 1979. R e v i s t a AGA Ano 21, Epoca V, NO. 92: 12-24. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0058829/manifest

Comment

Related Items