u L ti ARV ■™~*^*^"'"** p V, 1— t r» ►-« cr c ►->■ o "-^ o 2. -< S -< < c o ^ O r. »" c a g o. a. n o J" ?; a fw *^ .^ " rt ►— s r. » 3 ?3 1^ rn i-» rt O 0 " 8 □8 S- ?5 c o a rt H - 5l e- g- n • 3 w 3 J- M rf M 3" o f K' a "i ft £ o n H 3* cr o ►-. p < rt Jl n 5? ^ , . O ci r» "O cr » S o m Z 3) 3) •D < 0) (0 0 C c 33 0 o 00 3- 0 r- c ^ w "0 ^ w 0 O) ^ 3) y 3' ' cr 0 » 0 I c H OT $ <-» m W 5" -< 0 0 Q> 3 H 10 0 cn 0 ^ (D Z o> t 0 > (0 0' 0 (A 3 m Z •< m Z < 3J -TJ o o z en ^ i ? I 2 ^ H m o 0 > > m O -< ;a WATER POLLUTION CONTROL RESEARCH SERIES 1 3030 E LY A/7\ -7 REC-R2-7l-r OWR NO. IT'V-IO BIO-ENGINEERING ASPECTS OF AGRICULTURAL DRAINAGE SAN JOAQUIN VALLEY, CALIFORNIA PR 8 REC'D EMOVAL OF NITRATE BY AN ALGAL SYSTEM APRIL 197-1 UNIVERSITY Oh CXuFORNlA DAVIS SEP 13 1973 GOV'T. DOCS. -LIBRARY CALIFORNIA DtPARTMENT OF WATER RESOURCES KEO-ENCTNEERING ASPECTS OF ACSRICULTURAL DRAIMACE SAN JOAQUIN VALLEY, CALIFORNIA The Bio-Engineering Aspects of Agricultural Drainage reports describe the results of a unique interagency study of the occurrence of nitrogen and nitrogen removal treat- ment of subsurface agricultural wastewaters of the San Joaquin Valley, California. The three principal agencies involved in the study are the Water Quality Office of the Environmental Protection Agency, the United States Bureau of Reclamation, and the California Department of Water Resources. Inquiries pertaining to the Bio-Engineering Aspects of Agricultural Drainage reports should be directed to the author agency, but may be directed to any one of the three principal agencies. THE REPORTS It is planned that a series of twelve reports will be issued describing the results of the interagency study. There will be a summary report covering all phases of the study. A group of four reports will be prepared on the phase of the study related to predictions of subsurface agricul- tural wastewater quality -- one repoirt by each of the three agencies, and a summary of the three reports. Another group of four reports will be prepared on the treatment methods studied and on the biostimulatory testing of the treatment plant effluent. There will be three basic reports and a summary of the three reports. This report, "REMOVAL OF NITRATE BY AN ALGAL SYSTEM", is one of the three basic reports of this group. The other three planned reports will cover (1) techniques to reduce nitrogen during transport or storage, (2) possi- bilities for reducing nitrogen on the fann, and (3) desalination of subsurface agricultural wastewaters. BIO-ENGINEERING ASPECTS OF AGRICULTURAL DRAINAGE SAN JOAQUIN VALLEY, CALIFORNIA REMOVAL OF NITRATE BY AN ALGAL SYSTEM Prepared by the California Department of Water Resources William R. Gianelli, Director The agricultural drainage study was conducted under the direction of: Robert J. Pafford, Jr., Regional Director, Region 2 UNITED STATES BUREAU OF RECLAMATION 2800 Cottage Way, Sacramento, CaUfornia 95825 Paul DeFalco, Jr., Regional Director, Pacific Southwest Region WATER QUALITY OFFICE, ENVIRONMENTAL PROTECTION AGENCY 760 Market Street, San Francisco, California 94102 John R. Teerink, Deputy Director CALIFORNIA DEPARTMENT OF WATER RESOURCES 1416 Ninth Street, Sacramento, California 95814 DWR-WQO Grant #13030 ELY DWR-USBR Contract #14-06-200-3389 A April 1971 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25 REVIEW NOTICE This report has been reviewed by the Water Quality Office, Environ- mental Protection Agency and the U. S. Bureau of Reclamation, and has been approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Water Quality Office, Environmental Protection Agency, or the U. S. Bureau of Reclamation. The mention of trade names or commercial products does not constitute endorsement or recom- mendation for use by either of the two federal agencies or the Cali- fornia Department of Water Resources, ABSTRACT An algal system consisting of algae growth, har- vesting and disposal was evaluated as a possible means of removing nitrate-nitrogen from subsurface agricultural drain- age in the San Joaquin Valley of California. The study of this assimilatory nitrogen removal process was initiated to determine optimum conditions for growth of the algal biomass, seasonal variations in assimilation rates, and methods of harvesting and disposal of the algal product, A secondary objective of the study was to obtain preliminary cost esti- mates and process design. The growth studies showed that about 75 to 90 per- cent of the 20 mg/1 influent nitrogen was assimilated by shallow (l2-inch culture depth) algal cultures receiving 2 to 3 mg/l additional iron and phosphorus and a mixture of 5 percent C02. Theoretical hydraulic detention times required for these assimilation rates varied from 5 to 16 days, de- pending on the time of the year. The total nitrogen removal by the algal system, assuming 95 percent removal of the algal cells, ranged from 70 to 85 percent of the Influent nitrogen. The most economical and effective algal harvesting system tested was flocculation and sedimentation followed by filtration of the sediment. The algal cake from the vacuum filter, containing about 20 percent solids, was then air- or flash-dried to about 90 percent solids. The market value for this product as a protein supplement was estimated to be about $80 to $100 per ton. Preliminary estimates Indicate that the removal of nitrate from tile drainage by an algal system will cost about $135 per million gallons of treated water. This figure includes engineering and contingency costs and recovery of some cost by the sale of an algal product. The estimate will be refined at the end of operational studies to be completed in 1970. Key words: Algae stripping, nutrients, tile drainage, nitrogen removal, treatment costs. BACKGROUND This report is one of a series which presents the findings of intensive interagency investigations of practical means to control the nitrate concentration in subsurface agricultural wastewater prior to its discharge into other water. The primary participants in the program are the Water Quality Office of the Environmental Protection Agency, the United States Bureau of Reclamation, and the California Department of Water Resources, but several other agencies also are cooperating in the program. These three agencies initiated the program because they are responsible for pro- viding a system for disposing of subsurface agricultural wastewater from the San Joaquin Valley of California and protecting water quality in California's water bodies. Other agencies cooperated in the program by providing particular knowledge pertaining to specific parts of the overall task. The ultimate need to provide subsurface drainage for large areas of agricultural land in the western and southern San Joaquin Valley has been recognized for some time. In 195^* the Bureau of Reclamation included a drain in its feasibility report of the San Luis Unit. In 1957> the California Department of Water Resources initiated an investi- gation to assess the extent of salinity and high ground water problems and to develop plans for drainage and export facili- ties. The Burns-Porter Act, in I960, authorized San Joaquin Valley drainage facilities as part of the State Water Facilities. The authorizing legislation for the San Luis Unit of the Bureau of Reclamation's Central Valley Project, Public Law 86-488, passed in June i960, included drainage facilities to serve project lands. This Act required that the Secretary of Interior either provide for constructing the San Luis Drain to the Delta or receive satisfactory assurance that the State of California would provide a master drain for the San Joaquin Valley that would adequately serve the San Luis Unit. Investigations by the Bureau of Reclamation and the Department of Water Resources revealed that serious drainage problems already exist and that areas requiring subsurface drainage would probably exceed 1,000,000 acres by the year 2020, Disposal of the drainage into the Sacramento -San Joaquin Delta near Antioch, California, was found to be the least costly alternative plan. Preliminary data indicated the drainage water would be relatively high ^n nitrogen. The then Federal Water Quality Administration conducted a study to determine the effect of 11 discharging such drainage water on the quality of water in the San Francisco Bay and Delta. Upon completion of this study in 1967, the Administration's report concluded that the nitrogen content of untreated drainage waters could have significant adverse effects upon the fish and recreation values of the receiving waters. The report recommended a three-year re- search program to establish the economic feasibility of nitrate-nitrogen removal. As a consequence, the three agencies formed the Interagency Agricultural Wastewater Study Group and developed a three -year cooperative research program which assigned specific areas of responsibility to each of the agencies. The scope of the investigation included an inventory of nitrogen conditions in the potential drainage areas, possible control of nitrates at the source, prediction of drainage quality, changes in nitrogen in transit, and methods of nitrogen re- moval from drain waters including biological-chemical processes and desalination. TABLE OF CONTENTS Page ABSTRACT i BACKGROUND ii CHAPTER I - CONCLUSIONS AND SUMMARY 1 Conclusions 1 Summary 1 CHAPTER II - INTRODUCTION 5 Algal Growth 7 Growth Characteristics 8 Light 9 Temperature 10 Carbon Source 11 Inorganic Macronutrients 12 Micronutrients 13 Algal Harvesting 13 CHAPTER III - MATERIALS AND METHODS 15 Laboratory - Chemical and Biological 15 Algae 17 Water l8 Algal Growth 22 Lightbox 22 Miniponds 24 Rapid Growth Pond (RGP) 26 Algal Harvesting 28 Laboratory 28 Pilot-Scale Separation Studies 29 Sedimentation Tank 30 Shallow Depth Sedimentation Tank 30 IV TABLE OF CONTENTS (Continued) Upflow Clarifler 31 Upflow Sand Filter 32 Microscreen 3^ Vacuum Filter 3^1 Centrlfugation 35 CHAPTER IV - RESULTS AND DISCUSSION 37 Algal Growth 37 Lightbox Studies 38 Nutrient Requirements 38 Temperature 42 Algal Species 44 Water Source 44 Miscellaneous Lightbox Studies 46 Minipond Studies 47 Nutrient Additions 47 Phosphorus 47 Iron 50 Carbon Dioxide 51 Mixing 54 Detention Time 58 Culture Depth 60 Soil Ponds 62 Biomass Control 66 Addition of Fish 67 Rapid Growth Pond 69 Biological and Chemical Observations 70 Biomass Production 70 Algal Genera Noted 73 Predatory Organisms 74 Dissolved Effluent Nitrogen 74 Dissolved Oxygen 75 pH and Alkalinity 77 RGP Evaporation and Temperature Data .... 78 Algal Harvesting 80 Laboratory 80 TABLE OF CONTENTS (Continued) Pilot Scale Separation Studies 87 Sedimentation Tank 87 Shallow Depth Sedimentation Tank (Water Boy) 88 Upflow Clarifier 92 Upflow Sand Filter (Sanborn Filter) .... 92 Vacuum Filter 95 Micros creen 95 Centrifuges 96 Drying 97 Regrowth Studies 97 Botulism Studies 100 CHAPTER V - DISPOSAL 103 Animal Food Supplement 103 Use of Algae as a Soil Conditioner 108 Miscellaneous Possible Markets for an Algal Product 110 Use of Algae to Produce Methane Gas Ill CHAPTER VI - PROCESS EVALUATION 113 Removal Efficiencies 113 Process Configuration 115 Cost Estimates II5 CHAPTER VII - AREAS FOR FUTURE INVESTIGATION 123 ACKNOWLEDGMENTS 125 LIST OF REFERENCES 126 PUBLICATIONS I3I vi FIGURES Number Page 1 Growth of Algal Population In Batch Culture (Hypothetical) 8 2 Photomicrograph of Scenedesmus l8 3 Seasonal Variation in Total Dissolved Solids and Nitrate-Nitrogen in Agricultural Drainage Water Available at Wastewater Treatment Center . , , 19 4 Algal Growth Facilities at Agricultural Waste- water Treatment Center 21 5 Minipond Designs Used at the lAWTC 25 6 Rapid Growth Pond Used for Large-Scale Algal Growth Studies 27 7 Typical Dilution Curve Used to Determine Total Solids from Percent Transmittance 29 8 Schematic and Flow Diagram of Water Boy Shallow Depth Sedimentation Unit 30 9 Effect of Tube Angle on Settling Path of Suspended Particle 31 10 Schematic Diagram of Upflow Clarlfler 32 11 Schematic Diagram and Principle of Operation of Sanborn Filter 33 12 Principle of the Belt Filter 35 13 Effect of Phosphorous Addition on Nitrogen Assimilation 39 14 Effect of Iron Addition on Nitrogen Assimila- tion by Scenedesmus Cultures 39 15 Effect of Type of CO2 Addition on Nitrogen Assimilation by Scenedesmus Cultures 40 16 Nit rate -Nitrogen Remaining at Day 11 in Scenedesmus Cultures with Various Combinations of Fe, P, and Carbon Dioxide 4l FIGURES (Cont'd) Number Page 17 Comparison of Nitrogen Assimilation by Scenedesmus Cultures Using Water from Three Different Tile Drainage Systems 45 18 Comparison of Nitrogen Assimilation In Various Tile Drainage Systems - October Sample ... 46 19 The Effect of Phosphate Addition on Nitrate Assimilation In Outdoor Growth Units .... 49 20 Comparison of Nitrogen Assimilation in Mini- ponds With and Without the Addition of 2 mg/l P04-P 49 21 Effect of Fe Additions on Algal Nitrate Assimilation in Outdoor Growth Units in Late Spring 50 22 Effect of Fe Addition (3 mg/l) on Algal Nitrogen Assimilation in Outdoor Growth Units in Late Summer 50 23 Effect of C02 Addition on Effluent Quality in Outdoor Growth Units in Late Spring ... 52 24 Effect of Different Levels of CO2 Addition on Nitrogen Assimilation in Ponds at 11.4 Days Detention Time 54 25 Effect of Continuous Mixing on Nitrogen Assimilation, MP Run 3 55 26 Effect of Mixing Duration on Average Total Nitrogen Assimilation in Minipond Run 4A , . 56 27 Effect of Time of Mixing on Total Nitrogen Assimilation 57 28 Effect of Detention Time on Nitrogen Assimila- tion, 5 Percent CO2 Added 59 29 Effect of Detention Time on Nitrogen Removal - Summary of Data from 12-Inch Ponds from July- December 1969 60 30 Maximum and Minimum Minipond Temperatures - July to December I969 6I viii FIGURES (Cont'd) Number Page 31 Plot of Detention Time required to Assimi- late Approximately 80 Percent of the Total Nitrogen vs Average Run Temperature .... 6l 32 Effect of Three Different Culture Depths on Total Nitrogen Assimilation 62 33 Average Total Nitrogen Assimilation for Various Culture Depths, Minipond Runs 6-9B 63 34 Slope of Depth vs Percent Removal Curves Compared with Average Solar Radiation During Corresponding Run 64 35 Comparison of Average Total Nitrogen Assimi- lation at Two Culture Depths, 8 and 12 Inches Inches, and Various Detention Times .... 64 36 Comparison of Total Nitrogen Assimilation in One Soil Pond with a Comparably Operated Minipond and Standard Minipond with Highest Nitrogen Assimilation 65 37 Comparison of Volatile Solids in Miniponds With and Without Biomass Control 67 38 Percent Total Influent Nitrogen Assimilated in Miniponds, With and Without Biomass Control 67 39 Proportionate Increased Nitrogen Assimi- lation Attributed to Fish, By Season ... 68 40 Operation of Rapid Growth Pond During 1969 69 41 Average and Range - Percent Volatile Solids of Total Suspended Solids at Different Levels of CO2 Addition 71 42 Diurnal Fluctuations in Effluent Volatile Solids 71 43 Effluent Nitrogen and Volatile Solids Used in Biomass Production Figures 72 44 Change in Nitrogen Forms During Minipond Run 75 IX FIGURES (Cont'd) Number Page 45 Diurnal Fluctuations in Dissolved Oxygen . , 77 46 Diurnal Changes in Bicarbonate Alkalinity. . 78 47 Average Daily Evaporation in Rapid Growth Pond During 1969 78 48 Daily Maximum and Minimum Temperatures in Rapid Growth Pond During 1969 , . 79 49 Concentrations of Mineral Coagulants Required for at Least 90 Percent Algae Removal, Jar Tests 80 50 Concentrations of Lime, Alum and Ferric Sulphate Required for 90 Percent Algae Removal - Before and After Addition of Ferric Chloride to Rapid Growth Pond ... 82 51 Percent Transmissions of Water Boy Influent and Effluent Samples During First Three Months of 1970 89 52 Relation Between Solids Loading and Run Length, Sanborn Filter 9^ 53 Algal Growth Responses of Various Combin- ations of Untreated Agricultural Tile Drainage and "Algal Stripped" Water Mixed with San Joaquin River Water (Delta) ... 99 54 Nitrogen Assimilation in Various Minipond Runs Il4 55 Flow Diagram of Algal Stripping Plant , . , 115 56 Algae Growth Pond II6 57 One of Three 12-Pond Groups Constructed per Phase 117 58 Predicted Seasonal Variation of Tile Drainage Flow and Nitrogen Concentrations from San Joaquin Valley, California II8 TABLES Number Page 1 Relation Between Incident Light Intensity, Depth and Algal Concentration at I,^ = 100 ft-c 9 2 Chemical Analysis Schedule l6 3 Concentration of Mineral Constituents in Tile Drain Water 19 4 Trace Metal Analyses of Alamitos Sump Water During I968-69 22 5 Total Milligrams Nitrogen Assimilated by Algal Cultures at Different Temperatures and CO2 Additions 43 6 Chemical Analyses of Various Sumps ^4 7 Minlpond Runs Conducted at the lAWTC, 5/15/68 to 12/31/69 ^8 8 Average Percent Total Nitrogen Assimilated from 20 mg/1 Influent (Minipond Run 8) . . . 53 9 Slope of Various Depth versus Percentage Removal Curves ..... 63 10 Dominant Algal Genera at the lAWTC 73 11 Effect of Carbon Dioxide and Aeration on Alkalinity in Minipond Run 8 76 12 Estimated Cost of Lime, Alum, and Ferric Sulfate to Remove 90 percent Total Suspended Solids 83 13 Results of Laboratory Tests of the Effect of Various Polyelectrolytes on Plocculation and Sedimentation of Algal Cultures at the lAWTC 84 14 Polyelectrolytes Which Were Experimentally Beneficial and Economically Feasible Aids in Algal Separation 86 xi TABLES (Cont'd) Number Page 15 Water Boy Operational Criteria and Results for May 1970 90 16 Cost Comparisons of Four Coagulcuits 91 17 Composition of Slurry from Water Boy .... 91 18 Summary of Operational Data for Sanborn Filter, 11/17-21/69 93 19 Number of Hours Required to Reach 80 Percent Solids, from Original Sample Containing 6.4 97 Percent Solids 20 Composition of Algae Grown on Sewage .... 10^ 21 Protein, Carbohydrate, and Lipid Contents of Some Freshwater Algae 106 22 Amino Acid Composition of Oven-Dried RGP Product Algae 107 23 Estimated Algal Production by an Algal Stripping Plant, 1975-2000 109 24 Rail Freight Rates for 100-Lb. Algae Packages from San Francisco, California 110 25 Algae Stripping Design Criteria 119 26 Treatment Costs for Algae Stripping 121 27 Capital Cost for Algae Stripping 122 XI 1 CHAPTER I - CONCLUSIONS AND SUMMARY Conclusions 1. Algal growth and harvesting is a technically feasible method of removing nitrate-nitrogen from subsurface agricultural tile drainage in the San Joaquin Valley, This method can be used to reduce 20 mg/l of influent nitrogen to 3-5 mg/l which includes organic nitrogen remaining in unharvested cells. 2. Laboratory algal growth assays comparing algae-treated and untreated tile drainage mixed with potential receiving waters showed that treatment lowered the bio- stimulatory nature of the waste, 3. Preliminary cost estimates for the system described in this report are about $135 per million gallons. This figure is based on results of feasibility studies and may be revised after 1970 operational studies. Summary Tile drainage will support extensive algal growth, providing environmental conditions are optimum for such growth, The effect of several chemical and physical factors on algal growth were studied in laboratory and outdoor cultures. The following summary of the factors Includes the best estimates of optimum levels for outdoor cultures, 1. Three nutrient additives appeared necessary to support sustained massive growth of Scenedesmus quadrlcauda in tile drainage. About 2 milligrams of phosphorus per liter of waste were needed regardless of season. The addition of 5 percent carbon dioxide was required during part of the year, as was 2 to 3 rng/l of iron. 2. Mixing data obtained before carbon dioxide was added routinely to the outdoor cultures consistently showed that rapid mixing was required for maximum nitrogen removal. During the time when additional carbon dioxide was a growth requirement, ponds with carbon dioxide addition only had nitrogen assimilation rates comparable to those with both carbon dioxide and mixing. The exact mixing requirements of an algal system have not been defined but it appears that the maximum requirement will be about four hours of mixing at an average velocity of about 0.25 feet per second. -1- 3. When phosphorus was available to the algae, detention time was normally the most important variable studied in individual outdoor culture experiments. The theo- retical hydraulic detention times required for maximum nitro- gen assimilation varied from 5 to l6 days and appeared to be directly related to pond temperature within the range of 12''C and 25 °C, and independent of temperature within the range of 25°C to 33°Co As shown in items 4 and 5, depth and biomass control also significantly affected detention time. 4. The optimum culture depth in these studies was eight inches, but by a three to four day increase in deten- tion time, comparable nitrogen assimilation rates could be obtained at a 12-inch depth. Comparison of the difference between nitrogen assimilation at two depths, 8 and I6 inches, showed that the difference varied seasonally and was directly related to available light. That is, depth had less effect during summer light conditions than during the winter months, 5. Some mechanism may need to be incorporated into the growth units to control algal biomass. This mechanism may consist of a settling area in the pond itself. The set- tling area would remove the heavier, older, and less meta- bolically active cells from the system. With such a system, a minimum summer detention time of five days at a 12-inch depth should be feasible. Without it, about eight days detention time may be needed. The two series of laboratory studies which compared rates of nitrogen removal in effluents from various tile systems in the Valley showed that the Alamitos sump water used in these studies provided results comparable to the re- maining systems studied. These studies also showed that the addition of iron, phosphorus, and carbon dioxide was required for maximum nitrogen assimilation by algae grown in water from any of the systems studied. During the investigation relatively little definitive work was accomplished on soil-lined ponds. This process probably involves a combination of algal and bacterial meta- bolic pathways. Data from the two soil ponds at the lAWTC showed that removal efficiencies often were comparable to those of the best algal stripping pond and required only the addition of phosphorus to achieve these results. Harvesting of algal biomass is divided into three steps or stages, namely, concentration, dewatering, and drying. These steps differ in the amount of moisture remaining in the algal product. Studies at the Interagency Wastewater Treatment Center demonstrated that algae can be readily separated from -2- agricultural tile drainage and concentrated to 1 to 2 percent solids (by weight) either by coagulatlon-f locculation and sedimentation with any of several chemical coagulants or by use of a rapid sand filter (Sanborn filter) with backwashing. The slurry resulting from the concentration process can then be dewatered to about 10 to 20 percent solids by vacuum filtration or by self -cleaning centrif ugation. Using these processes, the effluent algae concentration will be only 5 to 10 percent of the influent, at least within the influent range of 100-600 mg/l of algae. Laboratory jar tests were conducted to determine the effectiveness of various mineral coagulants (lime, alum, and ferric sulfate) to determine their effectiveness in achieving coagulation and flocculation of the alga Scenedesmus in growth pond samples. The studies showed that the addi- tions of these minerals could effect 90 to 95 percent removal of the algae (influent suspended solids concentration of 100 to 600 mg/l) during all seasons of the year; however, the required concentrations varied with changes in operation of the growth unit. When iron (FeCl3) was added to the rapid growth pond as an algal nutrient, the concentration of the reagents required to remove 90 to 95 percent of the algae was about 5 mg/l for ferric sulfate, 20 mg/l for alum, and 40 mg/l for lime. These are compared to about 80 mg/l, 100 to l40 mg/l, and l80 to 200 mg/l for the same compounds when iron was not present in the growth pond. Approximately 60 polyelectrolytes were tested alone and with the mineral coagulants to evaluate their effective- ness in algal separation. Of the compounds tested, both anionic and cationic, 17 polyelectrolytes were found to aid coagulation and to be economically comparable to mineral coagulants. With iron added to the growth unit and no carbon dioxide addition, almost complete (99 percent) algal separa- tion was obtained with less than 0.2 mg/l of the cationic polyelectrolyte, Cat-Floc, The algal product was usually air dried to about 90 percent solids, although one sample was dried by a De Laval spray drier at the company's test facilities. Two to three days air drying was normally required to reduce the moisture content to the desired level. Literature review and market predictions indicate that a market can probably be developed for an algal product that will retail at about $80 to $100 per ton. The question of marketability can be answered more completely when the evaluations of the algal product are obtained from the com- panies that received representative samples. The high ash content of dried algae (30 to 50 percent) from a nitrogen -3- removal plant operated on tile drain effluent may preclude its use as a food for livestock but not as a protein supple- ment for fowl, or as a soil conditioner. Modification of the harvesting process to include in-pond settling, as well as chemical coagulation-f locculation may result in the pro- duction of two different by-products, one with 10 percent ash, and the other with about 50 percent ash. -4- CHAPTER II. INTRODUCTION This is the final report of field studies con- ducted by members of the Interagency Nitrogen Removal Group on the feasibility of using algal growth and har- vesting (algae stripping) as a method of removing nitrate- nitrogen from subsurface agricultural drainage in the San Joaquin Valley. These studies were conducted at the Interagency Wastewater Treatment Center (lAWTC) located near Firebaugh, California. Field work began in January 1968 and continued through December 1969; however, opera- tional studies will be conducted through December 1970. They will be discussed in a later (June 1971) report. The original Impetus for considering the algal process came from a formal feasibility report submitted to Department of Water Resources (DWR) (Oswald, et al 1964). Based on laboratory culture studies and a review of per- tinent literature, this panel of consultants concluded that algae stripping was a technically feasible means of removing nutrients from tile drainage and that removal of these nutrients would reduce the potential of the drainage for causing deleterious algal blooms in the receiving waters The consultants proposed that a two-stage field study be Initiated to determine the exact levels of nitrogen re- moval which could be realized by the process. The first stage involved construction and operation of a small (50 X 200 feet) prepllot plant. If the results of the prepilot studies show that algal stripping is a promising method, two pilot plants, each having a capacity of one million gallons per day (mgd), would be built. The pre- pilot plant was designed by engineers of the California Department of Water Resources following suggestions by the consultants and construction was started in July 1967. Algae stripping Is an asslmilatory removal pro- cess in which the nutrient in question is first incorpor- ated into cellular tissue and the cells then removed from the medium. The process Includes three distinct areas of activity: growing the algae, separation of the algae with the incorporated nutrient from the liquid phase (including drying of the algae), and disposal of the algal product. The effluent nitrogen from an algae stripping plant will consist of two fractions — the influent dissolved nitrogen not assimilated by the algae and the particulate cellular nitrogen not removed by the separation process. At the lAWTC the effluent total nitrogen limit was 2 milligrams per liter (mg/l) as recommended by the I967 Environmental Protection Agency (formerly the Federal Water Quality Administration) report entitled "San Joaquin Master Drain, -5- Effects on Water Quality of San Francisco Bay and Delta". This amounts to a 90 percent reduction of the 20 mg/1 aver- age nitrogen concentration predicted for a combined San Joaquin Valley tile drain effluent. To remove 90 per- cent of 20 mg/1 by the algal process will require 95 percent assimilation by the algae, with subsequent removal of 95 percent of the algal biomass. The concept of using algal growth to remove nutrients from wastewaters is not original with this project, but its application to agricultural wastewaters has received little consideration, Oswald and Gotaas (1957), Fitzgerald (i960), Hemens and Mason (1968) and North American Aviation (1967) are some of the workers reporting on the use of the algal process to remove nutri- ents, mainly nitrogen and phosphorus, from secondary sewage effluent. These studies all indicated that tertiary treat- ment by photosynthetic algae effectively removed nutrients but that some problems were encountered. Because algal growth is light and temperature dependent, growth and nutrient assimilation were reduced during periods of cloud cover and/or cold weather. The practical problem of eco- nomically removing the suspended algae was also noted. This investigation was intended to determine if these limitations were equally important in an area of moderate climate and with agricultural wastewater as the culture medium. The general objective of the feasibility phase of this study was to determine whether algae stripping would effectively remove nitrogen from agricultural tile drain- age effluent. In working towards this objective, several questions had to be answered before a true assessment could be obtained. Among the most important of these questions were: 1. Will agricultural tile drainage support sustained algal growth? 2. What levels of nitrogen assimilation can be expected from this growth? 3. What environmental conditions are needed for maximum nitrogen assimilation and growth? 4. Can the algae be readily and economically separated from the liquid phase? 5. What levels of total nitrogen removal can be obtained by the algal stripping process? -6- 6. Is there a potential market for an algal product, or how do you dispose of the algal product? 7. What Is the total cost of an algal treatment system. Including growth, harvesting, and disposal? 8. Does the process do the Intended Job; that is, does the treated effluent have less potential for causing noxious algal blooms in the receiving waters? 9. Are the results obtained from an isolated tile system applicable to a combined drainage facility? The discussion of the results of a study of the algae stripping process requires a basic understanding of various concepts of algal growth and separation. A brief review of some of the more pertinent facets of these two areas will be included in the following sections. Algal Growth Algae cells contain a group of pigments, the chlorophylls, which enable the organism to produce organic material through a series of reactions requiring light energy, water, carbon dioxide, and various inorganic nutri- ents. The photosynthetic process can be summarized by the following approximation (Jewell and McCarty, 1968): aCOg + CNO3" + ePO^i" + (c+3e)H''" + l/2(b-c-3e) HgO + sunlight ^C^H^jN^O^jPg + (a+b/4+5c/M-d/2-f5e/4)02 as indicated by this equation, the rate of nitrogen (or phosphorus) assimilation by algal cells is a function of the rate at which organic material is synthesized. In a nu- trient removal system, such as that studied at the lAWTC, the ultimate goal of systems design is to have the undesir- able nutrient -- nitrogen for example — to limit formation of cellular material. Using this concept, all other re- quired elements of photosynthesis should be available in optimum or excess amounts. To accomplish this goal, the effect of various factors on photosynthesis must be understood. -7- Growth Characteristics Although this category does not fit conveniently into the classification of factors affecting photosynthetic activity, some explanation of algal growth in two general types of algal systems may be helpful. A generalized algal growth curve for algae grown in a culture without nutrient replenishment (batch culture) is shown in Figure 1, The initial period of adjustment to the medium (lag phase) is followed by a period of rapid cell division (log or exponential phase ) . Cell growth is eventually lim- ited by nutrient availability or by light limitation caused by mutual cell shading. The algal biomass, or cell numbers, may then decline, al- though in multi- algal cultures the original species may be replaced by another alga with different nutritional requirements. The specific growth rate of cells during the exponential (log) growth phase in this type of system is a function of cell concentration and can be described by the equation: TIME -DAYS FIGURE I - GROWTH OF ALGAL POPULATION IN BATCH CULTURE (HYPOTHETICAL) dN St KN -1- where K is the specific growth rate (day "*■), N is the cell concentration (in any applicable unit), and t is the time in days. In most waste treatment systems utilizing algae, algal nutrients flow continuously through the system and thus are constantly renewed. The theoretical analysis of this type of cellular growth, continuous culture, has been comprehensively treated by Monad (19^9) and its application to algal systems reviewed by Retovsky (in Malek and Fencl, 1966) and Shelef, et al (1968), The importance of this type of process to algal cultures is that by manipulation of environmental characteristics (nutrient concentration. -8- light, and so forth) the algal population can be maintained at steady-state, or constant, population density. Con- versely, manipulation of population density can cause some factor (nutrient) to be rate-limiting. As will be shown in the Materials and Methods chapter, the growth units used in this study were designed to be approximations of a continu- ous-flow type of system, with nitrogen as the rate-limiting factor. Light Absorption of light energy by dense algal cultures follows the Beer-Lambert law: T T «-Ecd Id = ^o® where I© is the incident light intensity, I^^ is the inten- sity of light at any depth, d is depth in centimeters, c is the algal biomass in mg/1, E is the extinction coefficient in cm2/mg, and e is the base of natural logarithms. As shown by this equation, light penetration to I^ is directly affected by incident light and inversely affected by depth and culture density. Optimum light intensities for maxi- mum algal growth range from 200 to 400 foot-candles (ft-c), and the lower limit may be 100 ft-c. Using an extinction coefficient of 2 x 10-3 cm^/mg (determined experimentally), Bogan, et al (i960) calculated the depth at which various incident light levels would penetrate several different concentrations of algae and still leave 100 ft-c. These values are shown in Table 1. TABLE 1 RELATION BETWEEN INCIDENT LIGHT INTENSITY, DEPTH AND ALGAL CONCENTRATION AT I^ = 100 ft-c (Prom Bogan, i960) centra tioii Depth - cm for corresponding Ip (mg/1) 1,000 ft-c 2,000 ft-c 5,000 ft-c 10,000 ft-c 50 23 100 11.5 200 5.8 400 2.9 30 39 46 15 19.5 23 7.5 9.8 ^^•^ 3.8 4.9 5.8 -9- Using similar calculations Golueke and Gotaas (1958) deter- mined that 4.5 Inches was the theoretical optimum pond depth for domestic wastes, but practical experience Indi- cated that optimum culture depth below which light no longer limited algal growth was about 8 inches. The practical implications of the light-absorption equation are that if light-limiting conditions are to be avoided, high-rate algal ponds must be shallow and that depth may have to be adjusted seasonally. Even in the relatively shallow cultures used in chemostat (laboratory continuous- flow) studies, only a small percentage of the incident light energy is converted to cellular energy. Oswald (1963) re- ported that in laboratory studies with settled sewage, an average of 4 percent of the incident energy was fixed by the algal cultures. Conversion efficiency varied inversely with intensity, duration of light, and detention time, and directly with temperature and carbon dioxide concentration. Another possible method of increasing the avail- ability of incident light to individual cells is to move the algae into the light path by induced turbulence (mixing). Kok (1953) found that increased algal yield could be ob- tained in Chi ore 11a cultures by using intermittent or flashing light, and that the optimum proportion of light- to-dark period appeared to be about 1:9. In these studies the flash time varied from 3 to about 200 milliseconds. In outdoor cultures the flashing light effect can be achieved by mixing a sufficiently dense algal culture; however, the mixing would have to be such that the duration of exposure to light and dark were of the correct time intervals. Mixing of this type would not be economically feasible in conventional wastewater treatment systems. Temperature As with all organisms, temperature affects the growth rate of algae, normally following the Van't Hoff rule according to which the growth rate doubles for each 10 *C Increase in temperature, within the range of temperature tolerance. Oswald (1963), working with Chlorella pyrenoidosa, noted that optimum light conversion efficiency occurred at 20°C. Using mixed cultures of Chlorella and Scenedesmus, Witt and Borchardt (i960) found that the light saturation intensity level (lowest light level at which maximum growth rate was attained) was directly affected by temperatures between 20 and 30**C. Because it is normally impractical to heat an outdoor algal culture, seasonal tem- perature variations cause changes in the required detention times of the system. In periods of cold weather, the algae -10- are allowed a longer contact time with the nulrlent supply, which compensates for slower growth rates. Carbon Source Algae normally use free carbon dioxide as an in- organic carbon source, although some algae have been reported to use the bicarbonate ion, Osterlind (1950) stated that Scenedesmus quadricauda could freely use the bicarbonate ion but that Chlorella pyrenoidosa would only grow if provided with carbon dioxide (COp) . In poorly buffered systems, use of CO2 and bicarbonate (HC03) causes the equilibria in the following equations to shift to the right, accompanied by a rise in pH. CO2 + H2O ^— ^ HgCOo H"*" + HCO';;z±: iT*" + Co' The concentration of any of the components of the carbon dioxide-bicarbonate-carbonate buffer system is a function of temperature, pH, and total dissolved solids as well as the concentrations of the remaining components. The equilibrium equation for the formation of hydrogen and bi- carbonate ions from carbonic acid is: [HCO3 "^ ^ " ^1 where the dissociation constant, Ki, has been reported to be 3.5 X 10"7 at 18*'C (Chemical Rubber Publishing Co., 1951). At a pH of 8 the ratio of carbonic acid to bicarbonate ion is 0.0286, at a pH of 7 it is 0.286, and at a pH of 6 it is 2.86 (McKee_and Wolf^ 1963). A similar equllibrium_reaction between HCO3 ^""^ C0o~ has a reported K2 of 4.4 x lO" at 25°C (Chemical Rubber Publishing Co., 1951); thus at pH 7 the ratio of bicarbonate to carbonate ions would be 2,270 to 1, whereas at pH 11 the ratio would be 1 to 4.4 (McKee and Wolf, 1963) . At pH values above 9, carbonate precipitates as calcium and magnesium salts, thus decreasing total alka- linity. These precipitates also remove many algal nutrients, especially phosphorus and heavy-metal trace elements, by forming complex salts. In outdoor algal cultures exposed to the atmosphere, afternoon pH values may be as high as 10.5 to 11. -11- Inorganic Macronutrlents Inorganic macronutrient requirements Include nitro- gen and phosphorus. Most algae containing chlorophyll are able to use either ammonia or nitrate as a nitrogen source, and many species can use nitrite providing the concentration is low, about 0.001 molar (Fogg and Wolfe, 195^). Ammonium-N has been reported to be used in preference to nitrate, when both sources are provided to the same culture (Schuler, et al, 1953). As shown by the following equations, the use oT either of these nitrogen sources can cause undesirable changes in culture pH (Cramer and Meyers, 19^8). 1.0(N03)45.7(C02)+5.4(H20)-^C5^^Hg^802^3N^^Q+8.25 02+1.0(0H") 1.0(NH4"^)+5.7(C02)+3.4(H20)->C5^yH^^802^3N^^0"*^-25 02+1.0(H+) Nitrate assimilation results in the production of OH" ions which causes a rise in pHj whereas ammonium assimilation lowers the pH by formation of hydrogen ions. In addition to dissolved nitrogen sources, many blue-green algae can fix atmospheric nitrogen (Pogg, 19^7). A requisite for nitrogen fixation appears to be an extremely low concentration of available dissolved nitrogen in the growth medium. Dominance of nitrogen-fixing algae in a treatment plant would preclude the use of algae stripping as a means of removing nitrogen. Algae usually use phosphorus as orthophosphate (P04~^). The ratio of nitrogen to phosphorous concentrations in a typical algal cell is about 10:1. Phosphorus is es- sential to algal growth. Without it, no growth will occur, regardless of the algal species. Sawyer (1952) found that the N: P ratios in natural waters that were seemingly optimal for algal blooms varied from 30:1 to 15:1^ depending on the algal species involved. Some algae, when provided with quantities of phosphorus in excess of their requirements, can "store" the element up to a certain concentration (Ketchum, 1939a), Zabat (1970) found that, regardless of the availability of phosphorus, the maximum uptake by Chlorella soroklawa was 2 percent of the cellular dry weight. Phosphorus uptake can be influenced by light (Gest and Kamen, 19^8) and hydrogen ion concentration. Hydrogen ion concen- tration affects availability and has been mentioned previously. -12- Mlcronutrlents Many trace minerals are required for algal growth, although the actual concentrations required may be beyond the limit of detection by routine analytical procedures. The elements most commonly mentioned are iron, molybdenum, manganese, vanadium, cobalt, zinc, copper, sodium, and boron. In laboratory cultures, an accepted method of pro- viding these nutrients is by means of a soil extract (the supernatant from a sample of soil boiled in distilled water). The trace element concentration in agricultural tile drain- age may resemble a soil extract in that the water passes through several feet of soil before entering the drainage system. This brief summary shows that the growth studies at the lAWTC had to include the investigation of several variables, some of which (temperature and light, for example) were fixed by natural conditions. Algal Harvesting This phase of the study of the algae stripping process was designed to determine the most feasible method of removing the algae from the growth units' effluent while maintaining an acceptable concentration of algae in the plant effluent. The harvesting process can be divided into three distinct areas of activity, based on the amount of water retained in the algal product. The first step, con- centration, increases the solids from 0.015 to 0.040 per- cent by weight (depending on the concentration of algae in the growth unit) to 1 to 4 percent. The second step is de- watering which then brings the solids to 8 to 20 percent, and finally in the third step, the algal mass is dried to 85 to 92 percent solids by weight. At the latter moisture content, algae can be stored almost indefinitely without decomposition (Oswald and Gtolueke, i960). Relatively little work has been done on algal separation, especially at prepilot or pilot-scale levels. Oswald and Golueke (1968) present a comprehensive review of the results of several years of work on the problem of sepa- rating microscopic algae grown on secondary sewage effluent. In general, their results showed that the algae could be most economically concentrated by coagulation, f locculation, and sedimentation. (Coagulation is defined as the destabl- lization of colloidal particles, and flocculation as the mixing of the destabilized particles to encourage the for- mation of larger "floe" masses.) Dewatering was accomplished -13- by centrlfugation, with final drying in the open. An alter- native to separate dewatering and drying steps was to spread the concentrated algal slurry on sandbeds, which brought about the desired moisture content without the intermediate step of dewatering. North American Aviation (196?) also found sandbed dewatering and drying feasible in the harvest- ing of sewage-grown algae. In their studies, the algae were concentrated by sedimentation after coagulation and floc- culation and then spread on sandbeds. These studies were conducted in Central and Southern California, which are areas of moderate climate. A primary consideration in the algal harvesting process is the concentration of algae remaining after routine separation of the algal biomass from the liquid phase. Based on 2 mg/1 total nitrogen in the effluent, the maximum allowable concentration of algae in the effluent would be about 15 mg/1. The 2 mg/1 effluent nitrogen will probably contain about 0.5 mg/1 dissolved nitrogen with the rest as particulate nitrogen. Algae contain about 10 percent N; therefore, 15 mg/1 of algae can provide up to 1.5 mg/l N. Based on these figures, about 90 to 95 percent of the algae would have to be removed from the effluent of an operating algal treatment plant. Without this level of removal, organically-bound nitrogen would be regenerated by bacterial decomposition of the algae and would be available for algal growth downstream from the treatment site. Jewell and McCarty (1968) found that the amount of nitrogen and phos- phorus regenerated by the aerobic decomposition of algae varied with culture age, with almost no regeneration taking place in older cultures. The average amount of N regenerated after about 300 days of aerobic decomposition was about 49 percent of the original amount present in the cells. Foree and McCarty (1968) reported that after 200 days of anaerobic decomposition, algal cultures retained about 60 percent of the original particulate nitrogen. These studies indicate that algae cellular nitrogen must be considered to be part of the total nitrogen released from a nitrogen stripping plant. -14- CHAPTER III - MATERIALS AND METHODS The studies described in this report, with the ex- ception of regrowth studies, were conducted at the Inter- agency Wastewater Treatment Center. In both algal growth and harvesting, studies were conducted using laboratory "bench-scale" experiments and expanded prepilot and pilot- scale studies. In general, the laboratory tests in both phases of the algal system were designed to screen a series of compounds or variables before testing in the larger systems. The following sections describe in detail the various materials and methods. Laboratory - Chemical and Biological A chemical and biological laboratory was located at the site and contained the equipment needed to perform routine chemical analyses. This included a Kjeldahl distil- lation and digestion unit, gas chroma tograph, specific ion electrodes, spectrophotometer and other routine laboratory equipment. For special analyses, samples were sent to the Department of Water Resources' laboratory at Bryte, California, Samples for trace metal analysis were sent to the U, S. Geological Survey laboratory in Sacramento for analysis by emission spectrograph. Almost all chemical analyses used in these studies followed the procedures out- lined in Standard Methods for Examination of Water and Waste Water, American Public Health Association, 1955. Table 2 lists the routine chemical analyses used and the normal frequency of analysis. As shown in Table 2, the analytical emphasis was on nitrogen forms, especially nitrate and nitrite. Prelim- inary studies on all the listed constituents indicated that sampling and analysis at a greater frequency than shown was not necessary to follow changes in the algal system. Most of the analyses listed were completed within two to five hours of sample collection and were not preserved in any way. -15- TABLE 2 CHEMICAL ANALYSIS SCHEDULE Constituent Frequency Method Nitrate Nitrite Ammonia Organic Nitrogen Orthophosphate Iron Chemical Oxygen Demand Dissolved Oxygen PH Alkalinity Electrical Conduc- tivity Total Dissolved Solids 3 times/wk, 3 times/wk. once/wk, twice/wk. twice/wk. once/wk. Occasionally Occasionally Daily twice/wk. Occasionally Brucine, specific ion electrode Diazotization KJeldahl - Distil- lation KJeldahl Stannous chloride Phenanthroline Dichromate refluxing Winkler-Azide Modi- fication Electrode Titration - pH meter Wheats tone Bridge Occasionally Evaporation Precision and accuracy tests on all nitrogen forms and orthophosphate indicated that experimental error was within the limits suggested in Standard Methods, In spite of this, the routine use of the brucine test for nitrate, with some procedural modifications, often gave results of ques- tionable validity. To avoid this problem, use of the nitrate specific ion electrode was begun early in 1969. The instru- ment was standardized against known concentrations of nitrate in denitrified drainage water, A plot of meter readings versus concentration showed a straight line between 0,5 and 50 mg/l N, Although some problems were encountered, mainly day-to-day variations in electrode response, the specific ion electrode proved to be a rapid (up to 150 analyses per hour), simple, and reliable method of nitrate analysis. The biology section contained microscopes, an incubator, bacterial culturing equipment, and other neces- sary biological supplies and equipment. The primary method used to determine changes in algal blomass was measurement of volatile suspended solids. Volatile suspended solids were normally run two to three times per week, with total suspended solids once a week. One of the volatile solids analyses was conducted on the day total organic nitrogen analyses were made. The procedure for suspended and volatile -16- solids involved filtering 50 to 100 milliliters of algal sus- pension onto a weighed, preignited Whatman GFA glass filter disk. The disk was then dried for one hour at 103**C, weighed, and ignited at 560*'C for 15 minutes. A final weighing was then made to determine volatile solids, which were assumed to represent algal biomass. Approximate cell counts were made at weekly inter- vals, using a hemacytometer counting cell. These cell counts were used mainly to observe the condition of the algae and changes in species composition. In some lightbox studies, cell growth was followed by measuring i^ vivo chlorophyll fluorescence. A Turner Model III f luorometer was modified by adding a blue light source and the proper combination of filters (Corning CS 5-60 primary and a Corning CS 2-60 secondary) for measure- ment of chlorophyll a. The fluorometer, with a different light source and filter combination, was also used to determine true detention times in the miniponds. Changes in effluent concentration of Rhodamine B dye were followed and the detention time and mixing characteristics of the unit determined. Adjacent to the laboratory a continuous recording analyzer monitored rapid growth pond (RGP) water temperature and pH, and sunlight. A weather station, located on one of the RGP baffles, was checked daily to determine maximum and minimum air and water temperatures, evaporation, precipita- tion, and wind. Algae The original inoculum for the rapid growth pond was obtained in January 1968 from the University of California's Richmond Field Station. Approximately 100 pounds (20 percent solids) of a sewage -grown green alga, Scenedesmus quadri- cauda, or closely related species, were added directly to the RGP. This amount was added to the RGP on two more oc- casions, in the summer of 1968 and early in I969 when other algal species became dominant in the pond cultures. During the remainder of 1969* Scenedesmus retained its dominance and the RGP was used as a source of inoculum for all growth studies. -17- A coenobluni of typical Scenedesmus quadrlcauda cells Is shown in Figure 2. This alga is normally a four- celled coenobium; however, it is also found in a single-cell form or with as many as 32 cells in a col- ony. The genus has a cosmopolitan distri- bution, and one or more of the 30 known freshwater species can be found in most freshwater environ- ments in this country. The choice of this alga was mainly a matter of its avail- ability in large quantities from the Field Station. Pre- liminary culturing studies using tile drainage demonstrated that the species would grow in this type of medium. Additionally, the use of Scenedesmus quadricauda has some inherent advantages over most other algal species. Because of its wide distribution, planktonic nature, and ease of culturing, much literature is available on its physiology and biochemistry. Also, this alga has been reported to use the bicarbonate ion (Osterlind 1950), a pathway not available to all algae. Water Subsurface agricultural drainage was obtained from a collecting system called the Alamitos tile system, located near the site. The Alamitos system was selected for its easy access from readily available Bureau of Reclamation land and because extensive sampling had indicated that the water quality was similar to that predicted for the combined drain. This particular tile system drained about 400 acres of in- tensively farmed land on which safflower or barley was grown in the winter and principally rice and cotton in the summer. This type of farming resulted in predictable changes in the quantity and quality of the drainage water, the effects of which are Illustrated in Figure 3. Decreases in nitrate- nitrogen and total dissolved solids (TDS) concentration in the late spring were caused by increased tile flows when the rice crop was flooded. Tile flows varied from more than 1,000 gallons per minute (gpm) in the summer to less than 20 gpm in the winter of 1967-68. -18- lOPOO £ 8,000 Q 6.ooa NITRATE-NITROGEN (mg/l) TOTAL DISSOLVED SOLIDS (mg/l) FIGURE 3- SEASONAL VARIATION IN TOTAL DISSOLVED SOLIDS AND NITRATE-NITROGEN IN AGRICULTURAL DRAINAGE WATER AVAILABLE AT WASTEWATER TREATMENT CENTER TABLE 3 CONCENTRATION OF MINERAL CONSTITUENTS IN TILE DRAIN WATER (in milligrams per liter) Master Drain, 1969 Constituent Samples Taken at Alamitos Sump 5/2/67 7/1^/67 (before flooding) (after flooding) s or Calcium Magnesium Sodium Potassium Carbonate Bicarbonate Sulfate Chloride Nitrate Hardness Total Dissolved Solids Boron 373 190 1,390 0 247 3,600 559 17.6 1,710 6,590 14 19^ 220 106 160 875 1,900 4.1 20 0 0 373 220 2,070 3,500 304 1,000 9 20 921 1,200 3,950 6,800 9.3 11 -19- In Table 3 results of analyses made before and after flooding are compared with quantities predicted for the Master Drain. From these and other data, the Alamitos sump water appears to be quite similar to that of the predicted combined drain effluent during most of the year. However, flooding of the rice fields had a marked effect on water quality. Analysis of the trace mineral results in Table ^ indicates that these constituents remained relatively constant throughout the year. Because of the semiquantitative nature of the emission spectrograph, probably no major differences (with the possible exception of molybdenum) are shown. These elements are probably of limited solubility; therefore, their concentrations may be independent of flow. Because of the fluctuations of water quality illus- trated in Figure 3^ a covered storage pond was used to provide a constant supply of tile drainage water for the lightbox and all outdoor units. This pond, which could store about 820,000 gallons, was lined with polyethylene plastic to prevent loss of water or contamination by the ground water. The cover and lining minimized biological activity, thus providing a reasonably constant 30- to 90- day water supply depending on depth and detention time of the rapid growth pond. -20- -21- TABLE 4 TRACE METAL ANALYSES OF ALAMITOS SUMP WATER DURING 1968-69 Element Concentration 12/10/68 2/12/69 - parts per Sample Date 3/20/69 billion 4/8/69 9/6/69 Aluminum 31 1.4 2.5 1.7 14 Cadmium 23 20 2.5 12 11 Iron 8.6 4.3 3.2 10 6.3 Manganese 12 18 4.4 15 43 Molybdenum 40 17 32 77 57 Nickel 8.0 8.0 7.2 7.7 6.6 Vanadium 1.9 3.1 2.7 2.0 2.3 Zinc 5.7 5.7 10 6.7 5.7 Copper 2.9 1.4 2.5 1.7 1.7 Cobalt 1.4 1.4 2.5 1.7 1.4 Algal Growth The effect of different variables was tested at three levels of culture size: lightbox batch assays, small 1,000-gallon outdoor growth units, and a large l/4-acre demonstration pond. Because of lack of replicability of the large pond, most of the actual experimental work was con- ducted using the lightbox and small outdoor culture units. An aerial photograph of the Treatment Center (Figure 4) shows the location of the various facilities used in this investigation. Lightbox The lightbox was located in an air conditioned, con- verted office trailer provided with forced-air heating. The trailer was partially insulated and temperatures were -22- maintained at 25± 4**C. The lightbox consisted of two shelves, 12 X 3 feet, which held 200 one-liter culture flasks. No means of automated, mechanical agitation was provided, but air (compressed only or enriched with CO2) could be intro- duced to individual flasks via a central manifold. Air volume was regulated by short sections of capillary tubing in each line and resulted in approximately equal amounts of air being delivered to each flask. Four 12-foot cool-white fluorescent fixtures, com- prising a total of eight lamps, were placed about 15 inches above each shelf. Ballasts were separated from the light fixtures to minimize local heating. Light intensities on the shelves ranged from 350 to 400 foot-candles, depending on location. The lighting was controlled by a timer to regulate the light-dark cycle. Laboratory growth studies were customarily conducted with 1,000 milliliter (ml) erlenmeyer flasks containing 500 ml of culture. Each variable was tested in triplicate. Drainage water from the storage pond was used most of the time. On occasion, water was taken directly from the sump to determine whether storage affected the water's potential to support algal growth, A culture containing predominantly Scenedesmus from the outdoor growth units, usually the rapid growth pond, was used to inoculate individual flasks. The initial inoculum was normally on the order of 2,000 to 3,000 cells per milliliter. No attempt was made to obtain uni- algal or bacteria-free cultures in these studies; however, the algae in the inoculum were normally 90 to 95 percent Scenedesmus, Almost all lightbox studies conducted in 1969 used continuous lighting. No detrimental effect appeared, and the experimental time was shortened, allowing more assays to be run. In the laboratory studies, various levels of COg enrichment were tried. The addition of CO2 was accomplished by methods ranging from hand swirling, accelerated diffusion of atmospheric CO2 into the culture medium, to injection of 100 percent CO2. Temperature studies were conducted by immersing culture flasks to a one-inch depth in water baths at controlled temperatures. Certain observations were made daily during the course of the studies. Because of the primary interest in nitrogen uptake, emphasis was placed on the collection of daily nitrate data obtained by means of the specific ion electrode. Other parameters that were often measured daily were pH, maximum and minimum temperatures, and fluorescence. At the end of a run, the triplicate flasks of each set of -23- variables were often pooled for volatile solids, total KJeldahl nitrogen, nitrite, and cell count determinations. When the effect of various nutrient additions was tested, the stock solutions used were prepared from analytical or reagent grade chemicals. Miniponds A total of 22 small, resin-coated plywood ponds, termed "miniponds", were operated at Firebaugh, Eighteen of these ponds were 8 feet wide x 16 feet long with a designed operating depth of 12 inches (3 inches of freeboard) Approximate capacity of these ponds was 1,000 gallons. Two other ponds, also 8 x l6 feet, were used to study operating depths of 8 and l6 inches. The remaining two ponds (12 inches operating depth) were modified in the spring of I969 to improve the hydraulic characteristics of the ponds, specifically to eliminate stagnant zones during mixing, A comparison of the two designs is shown in Figure 5. All but two of the miniponds had mixing pumps whose capacity was approximately 80 gallons per minute (gpm). This provided average in-pond velocities of 0,25 to 0,5 feet per second (fps). Timers were placed in the electrical circuits of the pumps to vary mixing during a 24-hour period. The flow of water from the storage pond was metered into ponds by means of individual flowmeters. All water lines were made of opaque material to prevent clogging by algal growth. Effluent was drawn from near the bottom of the storage pond and discharged through a "broken" siphon arrangement. This device was also used to maintain a con- stant water depth. Normal operation of a minipond run began with the removal of all algae from a previous run. The ponds were then started with a common source of algae, usually from the RGP, For the first few days of a run, the ponds were oper- ated either on a batch basis or at long detention times. This acclimatization period was designed to allow algal biomass to accumulate before being influenced by the experi- mental variables. Pond samples were collected daily from near the effluent at about O83O hours and taken to the laboratory for analysis, In-pond pH values were determined once or twice daily by a portable pH meter. All water delivered to the miniponds was taken from the storage pond and metered into the ponds by flow- meters (rotameters) that were adjusted daily. Periodically, the rotameters were calibrated by timing the flow of a known -24- INFLUENT TO ^INFLUENT (IIXING PUMP INFLUENT TO /INFLUENT MIXING PUMP (a^ r\ -.-oJi IIttI. ^ t\\i4'. w //^ //^ ii\\ -CENTER ISLAND -EFFLUENT 8-0" STANDARD POND MODIFIED POND FIGURE 5- MINIPOND DESIGNS USED AT THE I. A.W.TC. volume and, if the meters were Inaccurate, the glass tubes and stainless steel floats were cleaned in a potassium di- chromate -sulfuric acid cleaning solution and then recali- brated. Nutrients were added to individual ponds or to the storage pond. Technical grade sodium nitrate was added to the storage pond to increase low summer nitrate concentra- tions to near 20 mgN/1. Often fertilizer grade phosphoric acid was added to the storage pond, but when phosphorus was being studied, it was added directly to the miniponds. The phosphorous concentration for most of the runs was 2 mg/l, as P. Iron (FeCl3) was added daily to the growth units when tested as a growth factor. Except during special mixing studies, the miniponds were mixed twice daily, from 0800 to O830 hours (for sampling) and from 1200 to 1530 hours. During most of the study, the miniponds were also swept once daily to eliminate concen- trated algal deposits caused by eddy currents „ Carbon di- oxide (5 or 100 percent) and atmospheric air were injected -25- Into some ponds through the Intake side of the mixing pumps during the afternoon mixing cycle only. Bottled COg was mixed with compressed air to obtain the desired concentration of carbon dioxide. The true amount of CO2 injected was esti- mated by recording changes in cylinder weight. During some minipond runs in 1969* a settling tank consisting of a 50-gallon drum with an influent line entering near the center of the drum was attached to one of the mini- ponds. Influent water came from one of the nozzles of the mixing system and a surface overflow returned the water to the pond. The algal sludge was periodically removed by draining from an opening near the bottom of the tank. During 1968, Sacramento blackfish (Orthodon micro- lepidotus) were added to some miniponds to test the possible effect of fish on algal growth. The fish ponds with mixing had a screen placed over the intake side of the mixing pump to prevent the fish from being brought in the pump suction. The California Department of Fish and Game used two miniponds to determine whether Clostridium botulinum might grow in this type of aquatic environment. One pond received a one-inch layer of soil from an area with a known botulism outbreak (Tulare Lake), while the remaining pond received one inch of soil from the Treatment Center site. Personnel from the Department of Pish and Game made periodic visits to the Center to sample the two ponds for bacterial analyses. These two ponds were also routinely sampled with the other miniponds; however, the only controlled variables were water depth (12 inches), P addition (2 mg/l), and de- tention time. The ponds were neither emptied, mixed, nor swept during the study period. Rapid Growth Pond (RGP) The rapid growth pond (Figure 6) is an asphalt- lined pond with a 12,5-foot wide folded raceway approximately 800 feet long. The center baffles were originally constructed of sheets of aluminum attached to both sides of a wooden, upright frame J however, algae accumulated between the two sheets and one sheet was removed. The RGP could be operated at depths from 0,5 to 3 feet, with the effluent taken from either the top or bottom of the pond. With its four avail- able mixing pumps, velocities of 1 fps were theoretically possible at any pond depth. This mixing velocity was con- sidered necessary to achieve complete resuspension of settled algae and periodic mixing and aeration of the bottom sludge layer. Each pump had an individual timer which allowed an almost infinite variety of mixing schedules. -26- Influent from the storage pond was pumped into the RGP semicontlnuously throughout a 24-hour period. Detention time was regulated by varying the length of time the influent pump functioned during each specified timing interval and by the flow of the pump. The effluent pump, also on a timer, could be programmed to pump at a specific time or as a function of pond depth. Recording flowmeters were attached to both influent and effluent pumps to record daily the amount of water entering and leaving the pond, A recircula- tion pump was available to introduce the pond culture into the influent line. Operation of this growth unit resembled that of the miniponds. The optimum levels of experimental variables noted in minipond runs were maintained as nearly as possible in the RGP. Because of its uniqueness, the pond was used mainly as a demonstration unit and as a source of algae for separation studies. During the 1969 operation of the RGP, daily changes often were necessary in mixing, sweeping, and so forth, to provide algae for testing various pieces of separation equipment. -27- Algal Harvesting Laboratory The laboratory used in the harvesting studies was separate from the chemical and biological laboratories des- cribed previously. Standarized coagulatlon-f locculation tests (Jar tests") were conducted in this laboratory to screen potential coagulants in the concentration step of algal harvesting. The coagulants tested included such com- pounds as lime (Ca(0H)2)^ alum (Al2(S04)3) and ferric sulfate (Fe2(S04)3)* as well as various polyelectrolytes. These tests were used to estimate the effectiveness of these different compounds for coagulation-f locculation before test- ing in the larger, pilot-scale units. Routine tests of the most promising test compounds, or combinations of compounds, were also conducted over extended periods of time to deter- mine possible seasonal variations in their effectiveness. The experimental procedures used for these tests are out- lined In the following paragraphs. Suspensions of algae for the Jar studies were ob- tained directly from outdoor growth units, usually the rapid growth pond. The suspensions had volatile solids concentra- tions of from 50 to 800 mg/1. In making the dally compar- isons of various chemical coagulants, a common supply of algae was used. It was collected on the morning of the testing period. The procedure for the Jar tests was standardized as follows: 800 milliliters of algae suspension were placed in 1,000 ml "tall form" beakers and the calculated quantity of reagent was added. The samples were then mixed by means of a multimlxer at a specified paddle speed and time, depend- ing on the type of coagulant used. The speed and duration were 40 rpm for eight minutes with lime and 70 rpm for three minutes with alum. These values were found to be optimum for maximum f locculation. If polyelectrolytes were used in conjunction with alum or lime, the conditions optimum for the principal chemical were used. If polyelectrolytes were tested alone, the mixing speed and duration most effective for alum were employed, although mixing speed and duration were optimized for the most promising polyelectrolytes. After the mixing operation, the beakers were placed in a darkened area of minimum air circulation. At the end of one hour, the sample supernatant was decanted and a spectro- photometer was used to determine the percent transmlttance of the supernatant at 410 m^ . During the first part of these laboratory studies, the amount of algal biomass (and other suspended solids) removed by coagulation-sedimentation was determined by using a few random samples to establish the relationship between -28- weight of blomass in the flocculated material and percent transmittance (at ^10 m^i) of the supernatant. A plot of these values was then used to estimate removal in the re- maining samples. This method proved to be tedious and it was modified in the following manner. The suspended solids and percent transmittance were determined for a sample of algae-laden water from a growth unit. The sample was acidi- fied with two drops of 6n H2SO4 to remove precipitated calcium carbonate. Several quantitative dilutions were then made of the original pond sample and the percent transmit- tance measured for each of the acidified dilutions. The calculated suspended solids versus the recorded transmittance for each dilution were then plotted, A typical curve is shown in Figure 7. The percent transmittances of the acidified super- natant from the floccu- lated samples were then determined, and the removal percentage or total suspended solids remaining were then obtained from the dilution curve. Chemical anal- yses occasionally per- formed on the supernatant included nitrate, nitrite, organic nitrogen, ortho- phosphorus, pH, hardness, calcium, and magnesium. All analyses were per- formed as outlined in the section on growth, except that calcium and total hardness were determined by the EDTA (ethylenediaminetetraacetic acid) titrametric method, with magnesium calculated as the difference between the two ions. 10 00 TOTAL B90 SUSPENDED 690 SOLIDS - 400 m^ /I 200 0 u 80- a. < / / 0 < / / Z 2 40- - i / / < ^^^ / 0- -"^ PER CENT REMOVAL FIGURE 7- TYPICAL DILUTION CURVE USED TO DETERMINE TOTAL SOLIDS FROM PER CENT TRANSMITTANCE Pilot-Scale Separation Studies The equipment used in these studies, which re- flected the types readily available for lease or rental, dic- tated the testing procedures that were followed. (Nothing in the following discussion is intended as an endorsement of any type of equipment,) Descriptions of the units and testing procedures are contained in the following paragraphs. -29- Sedimentation Tank. A rectangular sedimentation tank was borrowed from the Los Angeles County Sanitation Dis- trict and was used in the removal of algae and sediment from the rapid growth pond. The unit was approximately 4 feet wide X 6 feet deep x 22 feet long with a capacity of about 4,000 gallons. Water was pumped from the RGP to the bottom of the tank and effluent was withdrawn from near the top and returned to the growth unit. The settled algal slurry was moved to one end of the tank by blades attached to a continu- ous chain and was periodically discharged onto drying trays. Detention time was the only controllable variable. Shallow Depth Sedimentation Tank. This was a self- contained water treatment plant called the "Water Boy" obtained from Neptune Microfloc, Corvallis, Oregon. The Water Boy was the only unit consistently operated on-line with the RGP. The plant treated a portion of the pond ef- fluent and discharged its waste from the site. A schematic flow diagram of the unit is shown in Figure 8. As illus- trated in the diagram, the unit consisted of a flocculation chamber, a settling chamber, and a mixed-media filter. CHEMICAL ADDITIONS FLOCCULATION CHAMBER 1 400 GAL BACKWASH STORAGE FIGURE 8- SCHEMATIC AND FLOW DIAGRAM OF WATER BOY SHALLOW DEPTH SEDIMENTATION UNIT In the flocculation chamber the algae-laden water was mixed with a coagulant (or coagulants) in concentrations determined by Jar test studies, and from there passed to the settling chamber. The chamber differed from the one described in the preceding paragraph in that it was equipped with a module of settling tubes inclined at an angle of Ta upwards in the direction of flow. The theory behind this type of sedimenta- tion unit has been described by Hansen, et al (1969). Essen- tially, the settling tubes reduce particle" settling distances and thus lower the time required for sedimentation (i,e., detention time). Figure 9 illustrates the effect of tube inclination on settling distances of discrete particles, (Diagram and following description are taken from Hansen, et al, 1969). -30- FIGURE 9 - EFFECT OF TUBE ANGLE ON SETTLING PATH OF SUSPENDED PARTICLE "The path traced by a particle settling in a tube is the resultant of two vectors: V, the ve- locity of flow through the tube, and vs, the settling velocity of the particle. It can be seen in Figure 9 that if the settling surfaces are inclined upward in the direction of flow, the settling path of the particle is altered be- cause the component of the settling velocity which is parallel to the tube wall, Vgw, is opposite in direction to the ve- locity vector V. If V Is greater than Vg, the required length of the settling surface decreases as the angle Increases from zero up to about 25 to 30 deg. (at V » 2.5 Vs) and then Increases, approaching infinity as the angle of inclination is Increased to 90 deg. For V< Vg the tray length continues to decrease with increasing angle, " Use of the tubes also reduces turbulence and short- circuiting, thus promoting laminar flow of the water. With periodic flow reversal, or drainage, the inclination of the tubes facilitates removal of the algal floe. The coagulation- sedimentation step was designed to remove most of the sus- pended solldsj however, a mixed-media filter provided a final polishing of the effluent. After passing the filter, most of the clarified water went to discharge with only a small portion being stored In the backwash tank for back- washing the filter. The settled algal suspension from this unit normally was discharged into a holding sump to provide material for study of dewatering equipment. The Water Boy was used only to remove algae from growth pond water, that is, the concentration step. Controllable variables Included hydraulic detention time, type and amount of flocculate, and frequency of sludge re- moval. The unit, as tested at the lAWTC, required manual operation of the filter backwash and sludge removal cycles, Upflow Clarifier. In the upflow clarlfler con- structed at the site (Figure 10) a coagulating aid, usually sodium hydroxide, was added to the Influent algae-laden water from the RGP. The mixture then entered the apex of an -31- inverted cone and flowed through a thin diffuser layer to mix the. coagu- lant and algae-laden liquid. The algal floe flowed upward and, as the cross- sectional area in- creased, the veloc- ity decreased, causing the floe to settle. A mat of algal floe formed at the point at which the upward velocity of flow equalled the settling veloc- ity of the floe. The algal sludge was drawn off from the mat, and the clari- fied liquid passed off at an upper level. Operational variables were de- tention time and concentration of co- agulant,. SLUDGE MAT DIFFUSER(WIRE MESH WITH PIECES OF PVC) FIGURE lO-SCHEMATIC DIAGRAM OF UPFLOW CLARIFIER Upflow Sand Filter. A Sanborn filter was construc- ted for the algae stripping project by Bohna Engineering and Research, Inc., of San Francisco, California, The principle of operation for this unit is illustrated in Figure 11. A description of the principle is quoted from an unpublished Bohna internal report dated February 6, 1970: "The fluid to be filtered is pumped into a feed chamber at the base of the filter from where it flows upward in a thin channel called the feed wick which is closed at the top. The fluid then flows horizontally across a polypropylene cloth surface followed by a thin layer of sand, A second layer of cloth backs up the sand and allows the clarified liquid to enter a second channel called the drain wick from where the filtrate flows upward into a backwash-holding tank and then to final disposition. -32- CO z< UJU . h s; z Q-LJ llJ o 50- _I GU 5o >- o n o < fj w i1 . X u. < o # ^ z u o < CD UJ Q. o li. O y a. o z q: Q. 3ivnd aN3 anos- ow^z >40IM NIVdQ 1 >IDIM QNVS li 3ivnd aN3 anbs~> 3| XDIM 0333 '' XliJ 5= -f ~q: mx lijO ii. -33- "In backwashing, the overflow effluent valve and feed valves are closed and a dump valve at the base of the feed chamber Is opened instantaneously, allowing the hydraulic head to force filtered water down the drain wick, through the sand and out the dump valve, thereby removing the accumulated solids from the filter cloth and sand bed." This filter was operated on water directly from the rapid growth pond, without any chemical pretreatment, and the filtered effluent was discharged from the site. Most of the data for this unit were obtained during two intensive studies with supervisory help from Bohna. Con- trollable variables included influent flow rate, baclcwash volume, and the pressure differential across the filter media at which the unit was backwashed. Microscreen. An automatic rotating drum filter leased from the Zurn Company consisted of a 2-foot wide, rotating drum having a diameter of 4 feet and mounted in a self-contained unit. Algal-laden water entered the interior of the partly submerged drum and was filtered as the water flowed outwards through a screen attached to the revolving drum. Screens of 25 and 35 micron mesh size were supplied with the unit. Algae caught on the screen were washed off by a row of water jets located above the drum; and the algal slurry then dropped into a trough located inside the drum, above the water level. The microscreen was tested primarily as a concentrating device, but also as a means to dewater an algal slurry from the Water Boy. The major vari- able tested was the amount of solids loading per unit area of screen. Vacuum Filter. A continuous belt vacuum filter was leased from the Eimco Corporation to examine its feasi- bility both as a concentrating and as a dewatering device, A schematic diagram of the unit is shown in Figure 12, The filter media, a continuous porous belt, was immersed in the filterable material, A vacuum applied on the inside of the drum caused the suspended particles to form a cake on the belt. As the belt passed over the discharge roll, a cake deflector scraped off the cake. The belt was then washed as it passed back into the feed tank. The influent algal suspension for this unit con- tained from 0.3 to 3 percent solids. The number of possible experimental variables was quite large and included flow, algal concentration, filter material, drum speed, and amount of vacuum. With the exception of drum speed, all these variables were tested at the lAWTC. -34- FILTER 'MEDIUM X<^CAKE ^ DEFLECTOR i- -FEED TANK FIGURE 12- PRINCIPLE OF THE BELT FILTER r 'SEPARATE DISCHARGE TO DRAIN OR RECYCLE Centrlf ugatlon , Three types of centrifuges were investigated as possible means of primary concentration and/ or dewatering. The first of these units was a Bird Machine Company continuous flow, solid bowl centrifuge consisting of two concentric rotating elements inside a stationary housing. The inner element was a scroll-like unit rotating at a slightly higher speed than the outer element. Variables tested, using both straight pond water and a concentrated slurry, included flow, centrifuge speed, and liquid depth in the bowl. The second variety of centrifuge was a De Laval yeast-type separator built by the De Laval Separator Company which was designed for continuous feed and removal with solid/liquid and solid/liquid/liquid extraction. Centrif- ugal force in this type of unit is constant and discharge of solids is controlled by varying the size and number of nozzles. The third centrifuge was a De Laval self -cleaning continuous flow unit. In the last-named type of centrifuge, suspended solids accumulate on the side of the bowl until they reach a predetermined level and then are discharged as the bowl opens. Once the solids have been ejected, which occurs in only a matter of seconds, the bowl closes and normal operation continues. During opening and closing of -35- the bowl, bowl speed remains unchanged. Experimental vari- ables in testing this unit were feed composition (used as both a concentration and dewatering device), throughput (flow), length of time the bowl was open, and amount of bowl discharge. -36- CHAPTER IV - RESULTS AND DISCUSSION Data are presented In this section which were ob- tained in algal growth and nitrogen assimilation and har- vesting studies at the Treatment Center. Also included are the data of two other studies which accompanied algal growth — regrowth and botulism. The regrowth studies were concerned with the effectiveness of treatment methods for removal of those substances from agricultural wastewater which promote or "stimulate" algal growth. In the botulism work, the reseairch was designed to evaluate the possibility of botulism outbreaks in algal growth units. Algal Growth The investigation of algal growth was divided into three levels of interest — lightbox, miniponds, and rapid growth pond. The major emphasis in these studies was to determine the effect of various factors on the rate at which nitrogen is assimilated by algae in laboratory and outdoor cultures. To achieve continuity of results in these studies, nutritional and physical factors were made most favorable for the growth of Scenedesmus, It should be pointed out that there are other algae which are probably equally suit- able for nitrogen removal. The last part of this section on algal growth will provide some general information on vari- ous biological, chemical, and physical factors recorded during the course of the investigation. In this section many of the results will be re- ported in terms of nitrogen assimilation, expressed either as mg/1 or percent. In general, "nitrogen assimilation" is expressed or measured as the amount of soluble nitrogen disappearing (i.e,, removed) from a medium as a result of algal growth. This method of expressing nitrogen assimila- tion is a simplification of the true system found in algal cultures where growth and decomposition are occurring simul- taneously and a constant flux of nitrogen forms takes place. For example, in most cultures dissolved organic nitrogen increased with culture age, probably from extracellular ex- cretion or cellular decomposition. As defined in this report, this would be considered to be nonassimilated nitro- gen. Nitrogen balances obtained by determinations of both influent and effluent values indicated that nitrogen loss through anaerobic denitrification and release as nitrogen gas was not significant. -37- Llghtbox Studies Laboratory cultures of Scenedesmus were used to determine optimum nutritional requirements of the algae , the effect of temperature on algal growth, possible effects of seasonal variation in mineral quality of the tile drainage, and comparisons of algal growth and nitrogen assimilation in tile systems other than the system used at the Center. In addition to these studies, the llghtbox was used to evaluate the growth rate and nitrate uptake in tile drainage of algal species other than Scenedesmus . Because of the large numbers of flasks Involved in the llghtbox culture studies, nitrate was the only nitrogen form determined routinely. Occasional analyses for other forms indicated that nitrite and ammonia usually were present in amounts less than 0.10 mg/l and organic nitrogen in amounts ranging from 0.5 to 0.8 mg/l. Changes in nitrate were thus valid indicators of the effect of a variable on total nitrogen assimilation. The results of the batch culture studies were then used in outdoor cultures to provide optimum growth conditions for the algae. The results of these studies are detailed in the following paragraphs . Nutrient Requirements. Batch assays were first used to determine the necessary nutrient additions for opti- mum algal growth and nitrogen assimilation in the tile drain- age. Because the ratio of nitrogen to phosphorus (p) in Alamltos sump water is about 100:1 (as compared to approxi- mately 10:1 in the algal cell), phosphorus was the first element tested. Figure 13 Illustrates results of a typical experiment which compares 0 and 2 mg/l P addition to Scenedesmus cultures. In this particular study, unconverted nitrate meter readings, which are proportional to nitrate concentrations, were plotted instead of actual nitrate values, The meter readings were often used with the assumption that relative changes were a reliable indication of a variable's effect. At the end of five days, the culture to which no phosphorus was added, remained at, or near, the initial nitrate level; whereas, the culture containing 2 mg/l addi- tional phosphorus showed a 50 percent reduction In dissolved nitrate . In general, all llghtbox studies indicated that little or no growth occurred without supplemental phosphorus. The 0.2 mg/l originally present in the tile drainage water was not adequate for attaining the blomass levels of Scenedesmus cultures required for maximum nitrogen assimlla- tion. All subsequent laboratory cultures received 2 mg/l PO4-P, unless phosphorus was one of the experimental variables . -38- NO PO4 ADDEO^J-~y After the neces- sity for adding phosphorus had been established, fur- ther experiments were con- ducted on the effect of additional iron on Scenedes- mus cultures. A typical example of this type of assay is shown in Figure l4 where FeClo was used as the source of iron and the lighting cycle was set for 12 hours light and 12 hours darkness. The combination of 2 mg/1 P plus 3 mg/l Fe led to an increase in the rate and amount of nitrate assimilation. Experiments with different concentra- tions of Fe indicated that 3 mg/1 (from FeClj) produced nearly optimum results. Two additional iron compounds, ferrous sulfate (FeSO^) and ferric citrate (FeCSii^Oj) were tested. The results of the test showed that while both were adequate sources of Fe, optimum con- centrations differed. With the use of FeSO/|, the req- uisite concentration was similar to that of FeCl3, or 3 mg/1 Fe. However, when ferric citrate was used, about twice as much iron (as Fe) had to be added in order to achieve the same rate of nitrate assimilation. The use of a chelating agent, the sodium salt of EDTA, decreased the concentration of iron re- quired. For example, in one study 3.2 mg/1 Fe alone was as effective as 0.4 mg/1 Fe combined with 5.0 mg/1 EDTA addition. EDTA alone also resulted in an increase in nitrate assimilation, presumably by making the small amount of iron in the sump water avail- able for algal growth. Cost analyses of the use of EDTA in large-scale algal cultures indicated that the use of this DAYS AFTER INOCULATION FIGURE 13- EFFECT OF PHOSPHOROUS ADDITION ON NITROGEN ASSIMILATION NOTE : NON AERATED LIGHT I2 0N-I2 0FF -2mg PO4-P ^•r"2(ngP04+3mq/l Fe (FeClj) 0+ 10 20 30 DAYS AFTER INOCULATION FIGURE 14- EFFECT OF IRON ADDITION ON NITROGEN ASSIMILATION BY SCENEDESMUS CULTURES -39- NOTE: 2niq/l P, NO Fe LIGHT 12 ON -12 OFF -SWIRLED TWICE DAILY compound was not economically practical. After several rep- lications of the iron studies, 3 mg/1 of Fe (from FeCl3) was routinely added to lightbox algal cultures. Based on theoretical considerations, Alamitos sump water did not contain enough carbon to support the growth of biomass required to assimilate 20 mg/1 nitrate-nitrogen. This deficiency indicated that CO2 addition to laboratory cultures could enhance nitrogen assimilation by Scenedesmus, Studies were then initiated using a mixture of 4 percent CO2 in compressed air, compresed air only, and surface re- aeration (flasks hand-swirled twice daily). The results of one of the experiments are illustrated in Figure 15, which shows that the rate of nitrate uptake was significantly increased by the addition of of 4 percent CO2* and that compressed air also was an effective means of provid- ing some additional carbon. The pH values were about 7.2 to 7.3 for the 4 per- cent CO2* 9.6 to 9.8 for the compressed air, and 10,0 to 10,5 for the swirled cultures. On two occasions, various amounts of sodium bicarbonate were added to inoculated sump water to explore the possibility of using this compound as a carbon source of nitrogen assimilation. These cul- tures were then compared to other cultures receiving 4 percent CO2. In both of these studies, the addition of bicarbonate did not en- hance nitrogen assimilation. Although 4 percent CO2 was conclusively shown to benefit algal growth in the cultures, in almost all of the subsequent experiments at least two levels of aeration were applied, namely injection of 4 per- cent CO2 and diffusion of atmospheric CO2 by swirling the flasks twice daily. The two levels of COg addition were used to make the data more applicable to outdoor cultures, some of which received additional inorganic carbon. ^COMPRESSED AIR '4%C02 IN COMPRESSED AIR 10 20 30 DAYS AFTER INOCULATION FIGURE 15- EFFECT OF TYPE OF CO2 ADDITION ON NITROGEN ASSIMILATION BY SCENEDESMUS CULTURES The three levels of nutrients — Fe, P, and COp — were then combined in a factorlally designed experiment to determine their effect on nitrogen uptake by Scenedesmus cultures. Five levels of Fe at three levels of P were compared in flasks that received either 4 percent C02 or -40- compressed air. In addition, five levels of Fe were tested in cultures to which 2.0 mg/l P had been added and which were swirled twice daily. The nitrate remaining in the var- ious cultures is shown in Figure l6. These data indicate that interaction occurred between the three variables tested, and that in general, the addition of CO2 and Fe increased the rate of nitrate assimilation. The level of iron addition which provided the maximum effect was about 3 mg/l of Fe, The minimum requirement for complete nitrate assimilation at 11 days was 4 percent C02> 2 mg/l P, and 3 mg/l Fe, The addition of higher levels of Fe and P probably resulted in SWIRLED TWICE DAILY COMPRESSED AIR 47o CO2 IN COMPRESSED AIR 2 mg/l PO4-P 4 mg/l PO4-P 8 mg/ I PO4 -P ; LIGHT 12 ON -12 OFF 3 6 9 12 Fe ADDITION mg/l FIGURE 16- NITRATE-NITROGEN REMAINING AT DAY II IN SCENEDESMUS CULTURES WITH VARIOUS COMBINATIONS OF Fe, P, AND CARBON DIOXIDE the formation of an unavailable iron-phosphate complex. By day 16, all aerated cultures (with and without CO2) receiving the addition of 3 mg/l or more Fe had reached zero nitrate concentration. Carbon dioxide undoubtedly played a dual role in these culture studies by supplying additional inorganic carbon and lowering the pH to such an extent that many elements that ordinarily precipitated in algal cultures re- mained in solution. In most of the laboratory culture work at Firebaugh, 4 percent bottled COg was used; however, on occasion 100 percent C02 was tested. The addition of -kl- 100 percent CO2 was never found to be more effective than 4 percent COg and often was detrimental to algal growth, perhaps because pH values were less than 6. This value falls below the optimum range of 7 to 8.5 for Scenedesmus . Oswald (1963) found that about 0.5 percent CO2 addition to mixed cultures of Chlorella and Scenedesmus was most favora- ble for maximum light conversion efficiency. Possible seasonal variations in iron and phos- phorous requirements resulting from changes in the influent water quality were checked by conducting periodic laboratory studies in which cultures received various concentrations of these elements. These tests were normally run when the storage pond was being refilled. In all but one of these studies, 2 to 4 mg/l of both elements were necessary for maximum nitrogen assimilation. In the one exception iron was not required. Also included in many of these checks were comparisons of nitrate assimilation rates in sump water and storage pond water. Little or no difference was shown between the two waters, provided that both had comparable amounts of N, P, and Fe . In conducting this type of study throughout the year, seasonal differences were noted in nitrogen assimilation rates in those cultures receiving the optimum nutrient additions . Although these differences may be attributable to variations in inocula, they could have been the result of changes in water quality. This possibility will be discussed more fully in the section on comparison of water from various tile systems. On several occasions trace minerals were added to the cultures to determine their effect on algal growth. These elements seldom had any significant effect on nitrogen uptake, either individually or in commonly used trace element solutions. The few instances in which a beneficial effect was noted, the results could not be duplicated in further experiments . Temperature . A preliminary study on the effect of temperature on nitrogen uptake indicated that, of the three temperatures first tested (10°C, 20°C, and 30"C), 20°C was optimum for nitrate assimilation, and that Pe and P were necessary at all temperature levels. At 30 C, the initial Scenedesmus culture was replaced by a blue-green alga, Oscillatoria. The 10°C cultures had an extremely long lag phase and did not enter the exponential growth phase until near the end of the study. Therefore, another experiment was designed in which temperatures of about 15°C, 20 C, and -H2- 30°C were used. The apparatus used to maintain the vaj?lous temperatures did not permit testing the effect of 10°C at the time of the test. This second study differed from the first in two important respects — (l) three levels of CO2 addition (none, 4 and 100 percent) were tested at all temperatures and (2) all of the cultores were allowed to grow at 20^0 for four days before being subjected to the various experimental temperatures. (in the preliminary study, all culture flasks had reached the experimental temperature when they were inoculated.) When the cultures had assimilated all the nitrate in the flask, an additional 20 mg/l of nitrate-nitrogen was added along with 2 mg/l P and 3 mg/l Fe . Growth of Scenedesmus at the high tempera- tures followed the same pattern as that observed in the previous study — rapid initial growth, then dying away, with eventual replacement by Oscillatorla. This species change was noted when either no COg, or 4 percent CO2 was added but not with 100 percent CO2. The use of pure CO2 at all temperatures gave the same result: namely, a pale culture of appsLrently viable Scenedesmus which did not grow. The lack of growth may have been due to the low pH levels of these cultures (5.7 to 5.9). The 15°C cultures exhibited a surprisingly high rate of nitrogen uptake, especially with k percent CO2 addition. Table 5 summarizes the total nitrate assimilated by the different cultures. With the exception of the 100 percent CO2 cultures, the only combina- tion of vEirlables that produced a notable effect was the 4 percent CO2 concentration at a temperature of 15°C. The Inoculum used in this study was obtained from outdoor growth units which had been at a temperature of about 15°C for several days. Perhaps the period of acclimatization in the laboratory was not long enough for complete adjustment to the experimental temperatures . TABLE 5 TOTAL MILLIGRAMS NITROGEN ASSIMILATED BY ALGAL CULTURES AT DIFFERENT TEMPERATURES AND CO2 ADDITIONS o : Percent CO2 Added Temp. ""C . 0 4 IQQ- 15 11 32 1 20 16 17 3 30 11 11 1 -43- Al^al Species. During the spring of 1969* unialgal cultures of 30 species of algae were obtained from the culture collection of the University of California at Davis. These algae were then cultured in drainage water, with and without P and Fe additions, and their rates of nitrate uptake compared to that of Scenedesmus from the outdoor growth unit. Of the species tested (including green, blue-green, and diatoms), six grew well in sump water plus nutrients, and one grew in the P-deficient medium. The six species included three green algae — Ankistrodesmus, Eucapsis, and Gleocapsls; and three blue -green algae -- ArTacystis, Qscillatoria, and Anabaena. By far the best growth and nitrogen assimilation were noted in Ankis trodesmus and Anacystis cultures; however, growth and nitrogen assimilation were not appreciably better than those of Scenedesmus. With no apparent advantage to be gained from using these algae, no concerted effort was made to develop large-scale cultures of the species tested. Water Source. In a study of this type, where the water comes from an isolated system, one of the questions raised concerns the applicability of the data to an area- wide system. Although such combined drainage was not avail- able during the project, samples of other tile systems were collected and tested for their algal stimulatory potential. As the partial chemical analyses of the samples shown in Table 6 indicate, the sumps differed considerably in nitrate and electrical conductivity (EC). The differences in nitro- gen concentration were nullified by bringing all sumps to about 55 mgN/1 as nitrate. Part of this study was performed to determine whether or not iron and phosphorous additions were needed to produce the algal biomass required to assimi- late the nitrogen in each of these waters. The results showed that both elements were necessary and that there was TABLE 6 CHEMICAL ANALYSES OF VARIOUS SUMPS Sump umho/cm^ Concentration - mg/l Total KJeldahl|N03-N|N02-N|P02|-P[Total Pe DPS 1367 5890 0.28 HMV 7016 3880 0.30 GSY 0855 8100 0.25 BVS 7402 5200 0.40 (Alamitos) 45 0.003 0.06 0.03 55 0.003 0.00 0.04 16 0.004 0.00 0.07 13 0.006 0.038 0.05 -44- little or no algal growth without their addition to the media, The curves in Figure 17 show changes in specific ion meter readings with time for two of the sumps: GSY 0855 and BVS 7402 (Alamitos). Data from HMH 7OI6 and DPS 1367 were similar to those of Alamitos and were omitted from the graph for reasons of ease of reading. The curves illustrate definite differences in the rate of nitrate uptake, with the algae in GSY 0855 water removing all the original nitrogen in about 7 days. The algae grown in the other drainage water required about 12 days to remove the 55 mgN/l, As the concentration of nitrate in a flask reached zero, an additional 20 mg/1 NO3-N were added. In this particular study, GSY O855 water was "respiked" three times, and the cultures from the other three sumps only once. By the end of day 14, the algae growing in GSY O855 water had assimilated a total of 105 milligrams NO3-N, as compared to only 60 milligrams assimi- lated in the remaining sumps. The cells grown in GSY O855 tile drainage were practically all one-celled Scenedesmus, whereas a considerable number of Oscillatoria and diatoms appeared in the other water samples. ORIGINAL SAMPLES RESPIKED SAMPLES ALAMITOS (BVS 7402) NOTE : 4% COg 2mg/l P 3mg/l Fe CONTINUOUS LIGHT 5 10 15 DAYS AFTER INOCULATION 20 FIGURE 17- COMPARISON OF NITROGEN ASSIMILATION BY SCENEDESMUS CULTURES USING WATER FROM THREE DIFFERENT TILE DRAINAGE SYSTEMS -45- The water for the previous study was collected in early September when sump flows were characteristic of sum- mer irrigation practices, especially rice flooding. Another series of samples from the same tile systems was collected in October when field irrigation had been discontinued. The plots of nitrate meter reading versus time for cultures grown in the three tile drainages are snown in Figure l8. Complete nitrate assimilation in all the sumps required about 12 days or about the same length of time required by cultures in DPS 1367, HMH 7016, and BVS 7^02 water in the previous study. Results using water from GSY O855 were similar to those from the other sumps. Algae grown in a composite sample from the four systems to which iron and phosphorus had been added but which contained only about 30 mgN/l, assimilated essentially all of the nitrogen by day 7. The difference in time needed to reach zero nitrate between the composite and the individual sumps undoubtedly was caused by the original difference in nitrogen content of the samples . The individual drainage samples were all brought up to about 55 mgN/l as nitrate. More studies of this type are needed for confirma- tion; however, seasonal variation in the algal growth poten- tial of individual tile systems does occur, at least in labo- ratory cultures. The water used in this investigation appears typical of other systems, and the nutrient additions re- quired for growth of Scenedesmus probably will be required in a combined system. Miscel- laneous Lightbox Studies. Addi- tional lightbox studies were de- signed to answer specific questions about occurrences in the outdoor growth units. The results of these studies will be presented in later discussions, DAYS AFTER INOCULATION FIGURE 18- COMPARISON OF NITROGEN ASSIMILATION IN VARIOUS TILE DRAINAGE SYSTEMS- OCTOBER SAMPLE NOTE: 5% COj 2mg/l PO4-P 3nig/l Fe CONTINUOUS LIGHT -he- The use of laboratory batch assays can provide valuable Information about such items as nutrient require- ments, desirable algal species, and comparisons of various waters. However, some discretion is needed in interpreting the results of such studies, especially with respect to nutrient additives. Because environmental conditions in laboratory cultures are near optimum, growth rates are high and the effect of minor elements can often be dramatic. Outdoor cultures are often limited by light or temperature; therefore, the algal cultures do not achieve maximum poten- tial growth and the effect of a minor element is masked. This condition appeared to be especially true with iron, an element necessary for maximum nitrogen assimilation in prac- tically every llghtbox study but only seasonally required in outdoor cultures. Minipond Studies The minlponds were constructed during the spring of 1968 and were placed in operation that May. Listed in Table 7 are the minipond runs, their duration, and the re- spective areas of emphasis. During the first three runs, operational problems were being solved, and it was not until fall that adequate physical control of the ponds was achieved. Because of this, results of minipond runs befo2?e September 1968 will not be emphasized in this report. This section deals with the effect of nutrient additions and physical and bio- logical parameters on the uptake of nitrate-N. Nutrient Additions. Five nutrients which, based on the composition of the tile drainage, theoretically limited algal growth were added to the outdoor growth units -- phos- phorus, iron, inorganic carbon (CO2), manganese, and potas- sium. Of these five nutrients, only the addition of the first three enhanced algal growth and nitrogen assimilation. Manganese and potassium had no demonstrable effect. With all these additions, an attempt was made to maintain a culture of Scenedesmus quadricauda. The specific effects of each nutri- ent are described in the following paragraphs. Phosphorus. Preliminary laboratory culture studies had indicated that the phosphorous concentration in tile drainage water was limiting to algal growth. In the first three minipond runs, all of four to six weeks duration, the effect of four levels of P addition was evaluated. The results of these studies, shown in Figure 19, demonstrated that phosphorus also was necessary in outdoor cultures of Scenedesmus. The optimum phosphorous concentration was not determined. In runs 1 and 2, nitrate removal increased with phosphorous additions up to 2 mg/l P. However, in run 3, only the cultures to which 0.5 mg/l of phosphorus had been added assimilated more nitrogen than did the control which contained no additional phosphorus. During these three runs. -47- TABLE 7 MINIPOND RUNS CONDUCTED AT THE lAWTC, 5/15/68 to 12/31/69 Run No, Dates Main Areas of Study 1 5/15 to 6/28/68 2 7/1 to 7/23/68 3 8/5 to 9/17/68 4A 9/17 to 12/7/68 4B 12/8/68 to 1/25/69 5 2/10 to 4/7/69 6 4/16 to 6/27/69 7 6/27 to 7/29/69 8 8/8 to 9/29/69 9A 10/4 to 11/19/69 9B 11/19 to 12/31/69 Phosphorous addition, depth, detention time, fish, soil ponds Phosphorous addition, depth, detention time, fish, soil ponds Mixing, P addition, detention time, depth, fish, soil ponds Mixing, fish, soil ponds Mixing, fish, soil ponds Mixing, depth, detention time, soil ponds, fish CO2 addition, Fe addition, depth, soil ponds CO2 addition, Fe addition, biomass control, detention time, depth, soil ponds CO2 addition, Fe addition, biomass control, detention time, depth, soil ponds CO2 addition, Fe addition, biomass control, detention time, depth, soil ponds CO2 addition, Fe addition, biomass control, detention time, depth, soil ponds -48- influent flows were not regulated with the desired accuracy; therefore, the contradictory results in the third run were assumed to be caused by opera- tional problems. Beginning in the fall of 1968 and through July 1969* approximately 2 mg/1 P (added as H3PO4) were added to the storage pond. During the remainder of 1969, all of the miniponds except one received daily P additions. Total nitrogen assimila- tion for the control pond in one minipond run and a pond at the same depth, detention time, and so forth, are shown in Figure 20. The original inoculum for both of these ponds had been grown in a phosphorus -rich medium (drainage water plus 2 mg/1 P), and laboratory analyses of the filtered inoculum indicated that the residue contained about 4 percent phosphorus. Additional analyses of filtered and unfiltered pond samples indicated that a substantial portion of the phosphorus in the resi- due was composed of col- loidal precipitates caused by high in-pond pH values. The data in Figure 20 show that for a two-month period the nitrogen assimilation in the non-P pond was con- sistently 10 to 20 per- cent lower than that in the pond receiving supplemental phosphorus. By the middle of December, this dif- ference had increased to ADDITION (mq/l) FIGURE 19 - THE EFFECT OF PHOSPHATE ADDITION ON NITRATE ASSIMILATION IN OUTDOOR GROWTH UNITS ^^2mg/l PO4-P A '-^""^^ 1 ''' ^^'"WkJ^^^ NO ADDED j«>^. / \ ''°4 \ / \ \' ^ > OCT ' NOV ' DEC ' JAN ^ FIGURE 20- COMPARISON OF NITROGEN ASSIMILATION IN MINIPONDS WITH AND WITHOUT THE ADDITION OF Zmg/I PO4-P -49- 30 to ho percent, perhaps because all excess cellu- lar phosphorus, and all precipitated phosphorus In the pond, had been used. The phosphorus-deficient cultures were pale green and tended to settle more rapidly than those re- ceiving P. Microscopi- cally, the algae seemed to be viable although they lacked the dark green color of typical healthy cells. Iron. In almost all of the llghtbox studies, the addition of 2 to 4 mg/1 iron was neces- sary for maximum nitrogen removal. Results of the first test of this element in the miniponds, per- formed in the spring of 1969, indicated that sub- stantially more nitrogen could be removed by cul- tures with 6 mg/1 Fe added (using FeCl3 as the source) than by controls with no Fe added (Figure 21). The increase in removal amounted to about 3 mg/1 N at the de- tention times tested. The iron-supplemented cultures were characterized by a dark, luxuriant green color, and microscopic ex- amination of a pond sample showed large numbers of empty cell walls, indi- cating rapid cell division. As in the lightbox cultures, the use of EDTA seemed to decrease the amount DETENTION TIME (DAYS) FIGURE 21 - EFFECT OF FE ADDITIONS ON ALGAL NITRATE ASSIMILATION IN OUTDOOR GROWTH UNITS IN LATE SPRING I 2 IRON ADDITION (mg/l) FIGURE 22- EFFECT OF FE ADDITION (3 mg/1) ON ALGAL NITROGEN ASSIMILATION IN OUTDOOR GROWTH UNITS IN LATE SUMMER -50- of iron required for maximum effect, but its use was discon- tinued because of the dissolved organic nitrogen contributed by the compound. (Since EDTA contains about 10 percent nitrogen, the daily addition of 25 mg/l EDTA resulted in an increase of about 2 mg/l in effluent dissolved organic nitrogen.) Wide use of this chelating agent in an operating treatment plant probably would not be economically feasible. From July through December 19^9^ one minipond was always operated without added iron while the remaining ponds usually received a daily dose of 3 mg/l Fe, The addition of iron had little apparent effect on nitrogen uptake. Illus- trated in Figure 22 are the results of a typical minipond run during this period. Although the average total nitrogen assimilation was slightly higher when iron was added, the increase was less than 0.05 mg/l, and was of doubtful sig- nificance. Analyses for total (particulate plus dissolved) and dissolved iron in this and other runs provided inter- esting results. Since the dissolved iron in all samples, including one containing tile drainage only, and another from a pond which received 3 mg/l Fe were essentially the same, the solubility of the element must have been extremely limited. Conversely, total iron concentrations increased in those ponds receiving added iron with the result that more than 30 mg/l Fe was present in many samples, presumably ac- cumulating in the ponds as an iron phosphate precipitate. Because of the low solubility of Fe in tile drainage, the method of addition used in the studies reported herein was extremely inefficient and should receive more consideration before being used in future studies. A beneficial side effect resulting from iron addition was that the PeCl3 acted as a coagulant, thereby reducing the amount of coagulant required in separation studies. This effect is discussed further in the harvesting portion of the report. Carbon Dioxide. Addition of CO2 was first evalu- ated early in the summer of 1969^ with rather spectacular results. One hundred percent CO2 was added to two miniponds by bubbling the gas into the intake side of the mixing pumps during the afternoon mixing session. Results obtained with one of the ponds are illustrated in Figure 23. Almost im- mediately after COg injection, inorganic nitrogen and ortho- phosphate concentrations decreased to near zero. For the next few days the culture assimilated all the available inorganic nitrogen and then turned brown and appeared to die. Similar results were noted in the other minipond and in the RGP. In the second minipond, the period of complete assimi- lation was about twice as long as it was in the previous pond. Injection of COg was discontinued when the culture began to turn brown. Within four weeks after cessation of CO2 injection, the algae (Scenedesmus) had regained their -51- (I/6UJ) d-*'Od *N-20N lN3niJJ3 (I/6UJ) N-^ON lN3nndd3 -52- original color and metabolic activity. Monitoring the In- pond pH Indicated that the pH level never fell below seven, although It may have been much lower at the point of Injection, The mlnlpond runs following the Initial CO2 studies all Included various levels of CO2 addition — 100 percent, 50 percent, 5 percent, atmospheric air, and natural aeration. In general, the results of these runs indicated that the addition of CO2 increased nitrogen uptake and that a 5 per- cent CO2 concentration was preferable to 100 percent. Table 8 contains average nitrogen assimilation data from TABLE 8 AVERAGE PERCENT TOTAL NITROGEN ASSIMILATED FROM 20 mg/1 INFLUENT (MINIPOND RUN 8) Detention Time ; Type Addition Days ; 5^ CO2 ; 100^ COg : Air ; Natural Aeration" 3 43 52 46 5 64 62 60 49 8 78 74 70 mlnlpond run 8, which lasted from August 6 through September 29, I969. These data show that at a three-day detention time nitrogen assimilation was low regardless of the availability of CO2, Indicating that detention time, not carbon, was the limiting factor. At five days detention time, 5 percent CO2, 100 percent CO2, and atmospheric air were all somewhat more effective than was natural aeration. At eight days detention time, about 8 percent more nitrogen was re- moved in cultures receiving 5 percent CO2 than in those receiving atmospheric air only. Another example, based on mlnlpond run 9A (October 4 through November 19, 1969) is Illustrated in Figure 24. In this run, when the detention period was held at 11,4 days, an average of about 80 percent of the influent nitrogen was assimilated by cultures injected with 100 per- cent and 5 percent COg* whereas only about 55 percent was removed by cultures not receiving CO2 enriched air. When the detention period was reduced to eight days, nitrogen uptake seems to have been enhanced only when 5 percent CO2 was added. With further reduction of detention time to five days, assimilation was not affected by the presence or ab- sence of CO2 enrichment. An interesting aspect of the ex- periment in which 100 percent CO2 was added to one of the -53- ponds was the sudden decline in nitrogen uptake after November 9 (Figure 24). At this time the rate of algal growth in the pond seems to have de- clined, although the culture appearance remained green and healthy. This pond continued to receive 100 percent CO2 until December but never re- gained its former as- similation rate. COMPRESSED AIR FIGURE 24- EFFECT OF DIFFERENT LEVELS OF COg ADDITION ON NITROGEN ASSIMILATION IN PONDS AT 11.4 DAYS DETENTION TIME The appar- ently active toxic effect of 100 percent C02 demonstrated in the early summer of 1969 was never again noted in subsequent runs. Possible reasons for this effect may have been toxicity of the gas itself or toxicity of material brought into solution by the lower pH values. When the apparent toxic effect was first noted in the outdoor cultures, samples of pond water and CO2 were taken into the laboratory for lightbox culture studies. No toxicity was evident when filtered pond water was assayed for this factor with both 4 percent and 100 per- cent C02« Another possible explanation for the die-offs was based on the hypothesis that the algae died because of their rapid growth rate resulted in nutrient starvation. This hypothesis was tested by allowing laboratory cultures to remove all of the available nitrogen from the medium and leaving them in this nutrient-deficient condition for up to 10 days (with continuous lighting). At the end of this starvation period, the cultures were given nitrogen, iron, and phosphorus. The algae readily incorporated the additional nitrogen and assumed a healthy appearance. The true cause of the apparent CO2 toxicity is still unresolved and, since the effect was never repeated after the first few occurrences, the question may never be answered. Minipond runs in late 1969 did indicate that the ^iddition of CO2 was required to achieve maximum nitrogen assimilation in the outdoor cultures, but that the quantity of CO2 which is optimum will depend on season and drainage water quality. Mixing. The exact role that mixing plays in outdoor cultures may be a combination of various effects including: (1) moving the algae into the light; (2) reducing anaerobic -54- sludge banks caused by settled algae; (3) re- moving the algae from the system by keeping them in suspension and thus avail- able for discharge in the effluent; (H) replenish- ing of carbon dioxide by increasing the surface area exposed to the atmos- phere; and (5) by stirring the bottom deposits, making essential nutrients more available. In the first two minipond runs, mixing was limited to sweeping the ponds twice daily. In these studies, actively growing cultures of Scenedesmus often com- pletely settled from sus- pension during the after- noon, presumably as a result of entrapment in precipitated calcium and magnesium salts at pH values above 10. In minipond run St power and equipment became avail- able for mixing six minlponds Figure 25, indicated that the could be significantly increa cultures. DETENTION TIME -DAYS 0 24 HOURS OF DAILY MIXING RGURE 25- EFFECT OF CONTINUOUS MIXING ON NITROGEN ASSIMILATION, MP RUN 3 The results, shown in amount of nitrogen assimilated sed by continuously mixing the By the fall of 1968, all but two of the miniponds had mixing systems, and in run 4 various durations of mixing were investigated. The average percentages (each number is an average of three ponds) of total nitrogen assimilated for approximately 60 days operation are plotted in Figure 26. The two- and four-hour mixing periods took place in the afternoon, and the 12-hour period occurred during daylight hours. The results showed that total nitrogen assimilation was almost doubled by the optimum amount of mixing and that 4 and 24 hours had approximately equal effects. The unex- plained decrease in nitrogen removal in cultures mixed 12 hours daily noted in the Individual weekly averages was unexplained but statistically significant. During the run, total nitrogen removals were low in all the ponds; therefore, the study was repeated on a smaller scale in May and June, 1969. The main emphasis at that time was placed on deter- mining what portion of the daylight hours was optimum for -55- mixing the cultures. The results of the four mixing periods, four hours in the morning, four hours in the afternoon, 15 minutes of each daylight hour and for 14 daylight hours, are shown in Figure 27. After the initial period of adjustment. HOURS OF DAILY MIXING FIGURE 26-EFFECT OF MIXING DURATION ON AVERAGE TOTAL NITROGEN ASSIMILATION IN MINIPOND RUN 4A up to about May 20, four-hour mixing in the afternoon con- sistently increased the amount of nitrogen incorporated by the algal cultures. Full daylight mixing, either continu- ously or for 15 minutes of each hour, was detrimental to nitrogen assimilation. Again, in July 1969 cultures con- tinuously mixed during daylight hours assimilated less nitrogen (65 percent) than did cultures with afternoon mixing only (85 percent), both at eight days detention time. Based on the information from the runs cited above, the standard mixing regimen imposed was four daylight hours. For convenience in sampling, this period was divided into two intervals -- O8OO to O83O hours and 1200 to 1530 hours. One effect of this regimen imposed on the miniponds was to peri- odically change the type of flow through the system. Tracer -56- MIXING SCHEDULE 1500-1930-. I5min/hr -0600-2000- 5 10 15 20 25 30 4 9 MAY JUNE FIGURE 27-EFFECT OF TIME OF MIXING ON TOTAL NITROGEN ASSIMILATION Studies showed that the pond influent went through the pond as a discrete entity (plug flow) when the mixing pumps were off. During mixing periods, the ponds acted as almost completely mixed vessels. The average in-pond mixing velocities were deter- mined by measuring the velocity at various points in some selected ponds. These results showed that the average velo- cities varied from about 0.25 feet per second (fps) in the l6-inch pond to about O.5O fps in the 8-inch ponds. These velocities did not prevent the algae from settling, especially in some stagnant areas of the pond. The two ponds that were modified to eliminate these stagnant spots (see Figure 5) had higher velocities than comparable unmodified ponds, and they were relatively free of algal sludge banks. In spite of the increased mixing efficiency, no significant difference occur- red in rates of nitrogen assimilation between algal cultures grown in the two types of pond. After the addition of CO2 became routine, one mini- pond was left without CO2 and without mixing. The data from this unit, when compared to a similar pond that was mixed, showed comparable effluent dissolved nitrogen levels, a find- ing that Indicated mixing was not always essential to algal growth. Preliminary evaluation of the effect of CO2 addition to a nonmixed pond has indicated that CO2 alone may be ade- quate for achieving the desired rates of nitrogen uptake. The mixing requirements of an algal stripping system will receive more emphasis in the 1970 operational studies. -57- Detention Time. Detention times as used In this study are theoretical hydraulic terms computed by the formula: 9 = V P where 6 Is the detention time, F the flow, and V Is the volume of the container. Three separate dye tracer studies Indicated that the true and the theoretical hydraulic deten- tion times were In reasonable agreement. For example, the calculated theoretical detention time for one of the ponds was 5 days and the measured true detention time about 5.2 days. Because of pond design, mixing, and natural algal settling rates, cell detention time was significantly longer than the hydraulic detention time. In one Instance, when the two were compared by use of tracer studies, the values were about 30 days for the cells, as compared to 5 days for a hypothetical water molecule (periods of mixing, 0800 to 0830 hours and 1200 to 1530 hours). The effect of detention time can be best evaluated from results obtained after July 1969 because standard mix- ing, nutrient additions, and detention times were used during this period. Detention times of 3i 5t 8, 11.5* and 16 days were studied, although only three of these times were com- pared in any particular run. An example of a detention time study is illustrated in Figure 28. The ponds were operated on a batch basis until August 15, 19^9* when the desired detention times were set. After reaching steady-state, the pond with an eight-day detention time consistently assimi- lated 80 percent or more of the Influent nitrogen. At five days, steady-state was reached at a level less than 75 per- cent, and at three days the level was closer to 50 percent total nitrogen uptake. The sudden decrease in removal that occurred about September 12 may have been caused by a sudden temperature drop during this period, (See Figure 48.) The overall effect of hydraulic detention time is best illustrated in Figure 29, in which the data from four mlnlpond runs from July through December 1969 are summarized. The data are averages of total nitrogen assimilations for 12-inch ponds in which CO2 was the only other variable tested, i,e,, depth, mixing, etc, , were identical. As will be shown in later sections, detention times can also be affected by physical changes in the system. The influent total dissolved nitrogen concentration was about 20 mg/1 during this period. The results demonstrate clearly that detention times must be varied seasonally to achieve the desired algal growth. The average maximum removals of dissolved influent nitrogen for each run (which included start-up) were consistently near -58- 55100 3 10 20 AUGUST SEPTEMBER 25 30 A DETENTION TIME - DATE FIGURE 28- EFFECT OF DETENTION TIME ON NITROGEN ASSIMILATION, 57o CO2 ADDED 80 percent, while in the actual runs the steady-state levels were in the 85- to 95-percent range. In these 12-inch depth ponds, eight days was the shortest detention time which would allow for the assimilation of an average of 80 percent of the influent nitrogen. Although not shown in this figure because of differing mixing schedules and the absence of carbon di- oxide, data for May and June also indicated that detention times of less than eight days were too short for maximum nitrogen assimilation. Required detention time for a given rate of nitro- gen assimilation is intimately related to water temperature and must be adjusted to ensure that the rate of nitrogen assimilation is not significantly different from nitrogen input. Smooth curves of the maximum and minimum daily minipond temperatures are shown in Figure 30. Actual maxi- mum and minimum water temperatures recorded were 98°F (36.6°C) and 34°F (l,l°C) respectively. The average maximum temperature for each run was calculated from these curves. (Because the curves had generally similar shapes, minimum temperatures could also have been used.) In Figure 31* shows the average maximum temperatures versus the detention time required to remove about I6 mg/l of the total iijfluent nitrogen. -59- MINIPOND RUN No. AND DURATION 8 H H 9A -+-— 9B NOTE ; NUMBERS ABOVE BARS INDICATE THEORETICAL HYDRAULIC DETENTION TIME 11.5 11.5 AUG SEPT OCT MONTH DEC FIGURE 29- EFFECT OF DETENTION TIME ON NITROGEN REMOVAL- SUMMARY OF DATA FROM 12 INCH PONDS FROM JULY- DECEMBER 1969 From these data it appears that detention time was indirectly proportional to maximum average pond temperature between 15 and 25 "C, the range of temperatures during the selected study period. With standardization of pond operation and shorter run times during the operational studies, such a curve will be extended to cover the entire range of temper- atures found on the west side of the San Joaquin Valley, Culture Depth, Water depths of 8, 12, and l6 inches were studied as part of the investigation. Depth primarily affects amount and quality of available light energy received by algal cultures and indirectly affects mixing. As shown previously, the average water velocity in the l6-inch deep pond was about 0,25 fps as compared to 0,5 fps in an 8-inch pond. Experiments conducted during the period of May through December I969 were taken as the source of data for studying this variable, because of the uniform operating conditions during that time. Because of the limited pond availability, only the 12-lnch depths were replicated in any particular run. The data for the 8- and l6-inch depths were from single ponds. In all minipond runs, nitrogen -60- assimilation in the 8- and l6-inch cul- ture depths differed dis- tinctly. The curves in Figure 32 show the typical effect of depth. All these ponds were operated at eight days detention time and received 5 percent CO2. The overall average assim- ilations of total influent nitrogen were 80, 69, and 55 percent for 8, 12, and I6 Inches, respectively. FIGURE 30- MAXIMUM AND MINIMUM MINIPOND TEMPERATURES -JULY TO DECEMBER 1969 10 15 20 25 AVERAGE MAXIMUM POND TEMPERATURE 30 FIGURE 31- PLOT OF DETENTION TIME REQUIRED TO ASSIMILATE APPROXIMATELY 80% OF THE TOTAL NITROGEN vs AVERAGE RUN TEMPERATURE -61- note; detention time, 8 days with 5% 002 15 25 4 14 OCTOBER NOVEMBER FIGURE 32- EFFECT OF THREE DIFFERENT CULTURE DEPTHS ON TOTAL NITROGEN ASSIMILATION All of the average total assimilations for each depth were plotted for the five minipond runs. Figure 33* and the slopes (change in percent removal per change in depth. ^5^R/AD) of the straight lines from the percent- age at 8 inches to that at l6 inches were calcu- lated, and the values listed in Table 9 were ob- tained. The data show that the slope of the line increased as the days became shorter, and that the difference in nitrogen assimilation between algae cultures grown at 8 and l6 inches of depth was more pro- nounced when available light energy per day increased. In Figure 34 the slopes from Table 9 have been plotted as a function of the average daily solar radiation during each of the runs. A straight line fitted by eye indicates that the effect of depth within the limits tested was directly correlated to average daily solar radiation. As illustrated in Figure 35, one effect of de- creased depth is to lessen the detention time required for algal assimilation of the available nitrogen. In the four runs shown, the average amounts of nitrogen assimilated were practically identical; however, in each run the neces- sary detention times were three to four days shorter in the 8-inch culture depths. The savings in detention time are about balanced by the additional land area required for treatment at the shallower depth. Soil Ponds. The two ponds which had layers of soil on the bottom received a minimum of operational atten- tion, consisting mainly of regulation of the influent flow and addition of phosphorus, and they were never drained or refilled. In spite of, or perhaps because of, the lack of attention, both ponds maintained cultures of organisms which removed surprisingly large amounts of nitrate-nitrogen. Both ponds had a mixed biota consisting of algae, zooplankton. -62- 90i d 80- a 40- 5 DAYS D.T. 8 DAYS D.T. ^..^x - \ \ \ 8 DAYS-T-^\,\ D.T. Xq DATES OF RUN □ MAY- JUNE • JULY "-^5-11 DAYS D.T. O AUG -SEPT ^x X OCT -NOV V •^ A NOV-DEC D.T DETENTION TIME POND DEPTH - INCHES FIGURE 33- AVERAGE TOTAL NITROGEN ASSIMILATION FOR VARIOUS CULTURE DEPTHS, MINIPOND RUNS 6-9B TABLE 9 SLOPE OF VARIOUS DEPTH VERSUS PERCENTAGE REMOVAL CURVES Run : Approximate : : Dates of Run : Slope (A^R/AD)* 6 7 8 9A 9B May-June July August-September Oc tober-November November-December 1.5 1.5 2.5 3.1 5 ♦Percent change In removal per change in depth from 8 to l6 inches. -63- higher aquatic plants, bacteria, and flying Insect larvae (mainly dlp- teran) , Dissolved oxygen determin- ations made at dif- ferent times of the year Indicated that the soil ponds were always aerobic, at least above the soil-water Inter- face. The amount of algae was usually low (less than 100 mg/l)* S"<3 the algae were often of the flagellated variety, e.g., Euglena and Carte rla. An indi- cation that the nitrate removal NOTE; CHANGE IN PER CENT REMOVAL AS DEPTH CHANGES FROM 8 TO 16 INCHES SLOPE OF LINES (A%R/CHANGE IN OEPTK+) FROM TABLE 9 FIGURE 34-SLOPE OF DEPTH vs PER CENT REMOVAL CURVES COMWRED WITH AVERAGE SOLAR RADIATION DURING CORRESPONDING RUN lOCh 80- 70- 60 5 8 CULTURE DEPTHS NOTE; NUMBERS ABOVE BARS INDICATE THEORETICAL HY- DRAULIC DETENTION TIME 11.5 8 11.5 JULY AUG SEPT OCT TIME NOV DEC FIGURE 35- COMPARISON OF AVERAGE TOTAL NITROGEN ASSIMILATION AT TWO CULTURE DEPTHS, 8 AND 12 INCHES, AND VARIOUS DETENTION TIMES -6^- mechanism was not entirely algal was provided in a study in which a sample of the soil, water, and organxsms was en- closed in a darkened bottle and floated jn tht pond. At the end of one week, the original 21 mg/l of total dissolved nitrogen had been reduced to 1^.5 mg/l ( JO3-N, 12 mg/l; NO2-N, 2 mg/l; and NH3+0rganic-N, 0.5 mg/l). Because light was not available during the incubation period, the nitrogen reduction was assumed to be the result of anaerobic bacterial decomposition. The dissolved oxygen content of water in the bottle decreased from about an initial 12 mg/l to a non- detectable amount at the end of the test period. The organic carbon source was presumably the product of algal and bac- terial decomposition, as well as extracellular organic products of algal metabolism. A comparison of nitrogen removal data from one of the soil ponds with a minlpond of comparable depth and de- tention is shown in Figure 36. The remaining soil pond followed essentially the same pattern as the one shown and was omitted for graphical clarity. Also included in the figure are the removal data for the best 12-inch depth non- soil pond in each minlpond run. These data support the fol- lowing conclusions: soil pond cultures consistently removed more nitrogen than did a mixed minlpond at a similar deten- tion time, and the removals in the soil ponds were often higher than in the best routinely operated minlE>ond, /STANDARD MINIPQND WITH <~^HIGHEST NITROGEN ASSIMILATION il DEC JAN FEB MAR APR MAY JUN JUL AUG MONTHS FIGURE 36-COMPARlSON OF TOTAL NITROGEN ASSIMILATION IN ONE SOIL POND WITH A COMPARABLY OPERATED MINIPOND AND STANDARD MINIPOND WITH HIGHEST NITROGEN ASSIMILATION -65- The operational studies ^n "1970 were designed to define the mechanisms involved in reduction of nitrogen in these ponds, which we have called "symbiotic ponds". The nitrogen removal mechanisms for these units appear to resemble that in grass plots described by Williford and Cardon (l97l) in which water flowing through fields of water grass showed marked nitrogen reduction. In the grass plots, relatively little planktonic algal growth took place; dissolved oxygen concentrations were always above zero; and no organic carbon was added. Net influent nitrate concen- trations ranging from 30 to 300 mg/1 (as NO3) were reduced by amounts ranging from 50 to almost 100 percent. The authors tentatively concluded that the reduction was the result of anaerobic denitrification in local strata near or at the soil-water interface. Biomass Control. The term biomass control is used to describe the process of removing settleable algae and other settleable materials which tend to accumulate in the ponds during long periods of operation. Mixing schedule, algal settling, and pond design each contributed to the problem of accumulation of algal cells in the small outdoor ponds. The possibility that the older uells might be detri- mental to growth of the younger cells was considered in this study. The detrimental effect could result from light ex- clusion by older, nonreproducing cells, by the production of some autoinhibltory substance, or by competition for scarce nutrients. Although not a part of the algal biomass, particles of suspended inorganic precipitates also tended to accumulate in the ponds and substantially reduced light penetration into the cultures. To determine whether or not such effects were present and their influence, a settling tank (described previously) was attached to the mixing pump of one of the two redesigned miniponds, while the other pond was operated as a control. As shown by the curves in Figure 37* the settling tank reduced the volatile solids within the pond by about 50 to 60 mg/1 during mlnipond run 9A, The total nitrogen removal for the two ponds is shown in Figure 38, both at eight days detention time and with 5 percent CO2 addition. The average total removal for the pond with biomass control was about 80 percent; for the control pond, 61 percent. During this run the best of the remaining 12-inch ponds also removed about 80 percent but at a detention time of 11.5 days. In another run involving five days detention time, the pond with the settling tank removed an average of 71 percent compared to 59 percent re- moved by the control. These data indicate that accumulation must be prevented in outdoor cultures, and that the construc- tion of larger units may have to include some mechanism for controlling the accumulation of inactive algae in the pond. -66- 400- _ 3CX>J \<-!*° BIOMASS CONTROL E S ~— — ^^^ O J 20O \ ^BIOMASS CONTROL § 100- FIGURE 37-COMPARISON OF VOLATILE SOLIDS IN MINIPONDS WITH AND WITHOUT BIOMASS CONTROL such as in-pond settling areas with sludge re- moval, or other suitable methods. By providing control of excess blo- mass and Inert materials the required detention times for maximum nitro- gen assimilation may be lowered significantly. The data from this study of blomass control Indicated that, during the period from July through November, the rate of nitrogen assimi- lation In the blomass- controlled ponds were comparable to other 12- Inch ponds which were being operated with a about three days of ^ additional deten- | tion time. Thus « with blomass control, g five days may be i | the minimum summer ' 5 detention time to z achieve 80 percent 3 or more assimilation. s Addition 2 of Fish. Some pre- ^ limlnary studies S Indicated that fish £ may possibly en- hance the growth of algae in tile drain- age waters and aid algal assimilation of nitrogen. Their beneficial effect was thought to be due in part to their stirring action and in part to their production of some waste product essential to algal metabolism. For these reasons, and because fish might also add a true recreational value to drain waters, a quantity of Sacramento blackfish weighing from 0,5 to 1.5 pounds each were obtained in May I968 and placed in some ponds during the first four minlpond runs. Some ponds received 2.5 pounds of fish and some received 5 pounds of -BIOMASS CONTROL V NOTE : DETENTION TIME - 8 DAYS FIGURE 38-PER CENT TOTAL INFLUENT NITROGEN ASSIMILATED IN MINIPONDS, WITH AND WITHOUT BIOMASS CONTROL. -67- fish. This particular species was selected for Its reported active consumption of planktonlc algae. The fish survived in the miniponds, although some mortality was noted. Unfortunately, no record of growth was maintained; thus we do not know if the fish actually gained weight in the ponds. The fish did have a beneficial effect on nitrogen uptake. In Figure 39, which summarizes their impact in the first four runs, the best minipond without fish is compared to a pond with comparable operating cri- teria containing fish. In all four runs, the algae in the ponds with fish assimilated more nitrogen than did the con- trol ponds. The increased nitrogen assimilation attributed to the presence of fish ranged from 13 to 43 percent with an average of 28 percent. After minipond run 4 A, all fish were removed from the miniponds, chiefly because iron additions had proved to be more effective at increasing nitrogen re- moval but also because the screens covering the Influent to the mixing pumps (to keep the fish out) were reducing the mixing efficiency. Large algal ponds can apparently support fish growth and fish apparently do enhance algal growth and nitrogen assimilation. lOO- o UJ 1- < 2 75- 'mm ^P m>. CO < z UJ o ^:^^^T 1 s H,^^^^^ CENT TOTAL NITRO BEST POND 1 CONTROL 1 '///////// ERAGE PER o MAY JUNE JULY 1968 AUG SEPT OCT FIGURE 39- PROPORTIONATE INCREASED NITROGEN ASSIMILATION ATTRIBUTED TO FISH, BY SEASON -68- f Rapid Growth Pond This pond was of limited value as a research tool because of the lack of a comparable unit to use as control; consequently, it was used principally as a demonstration unit and a source of algae for separation studies. During the first few months of operation we were unable to maintain Scenedesmus cultures and the pond was reseeded twice, once in the summer of I968 and again in January I969. The pond normally was operated with a 12-inch water depth and at vary- ing detention times, on the basis of minipond data. During these first few months, the pond occasionally was operated at less than 20 mg/l influent nitrogen and with different mixing cycles. After January I969, the pond was operated for almost five months without being drained and with little variation in operating procedures. Plots of effluent nitrogen, influent nitrogen, and detention time for the RGP during the calendar year 1969^ are presented in Figure 40. The period up to June 12 was characterized by a relatively stable effluent that averaged about 5 mg/l total N (75 percent assimilation). On June 12, 100 percent 002 was added to the pond, whereupon the effluent inorganic nitrogen decreased to near zero. Ten days later the pond turned brown and nitrogen removal ceased. On July 7 -INFLUENT NITROGEN MAR APR MAY JUN JUL AUG SEP OCT MONTHS FIGURE 40- OPERATION OF RAPID GROWTH PONO DURING 1969 -69- the pond was drained and cleaned, and on July l4 the pond was refilled and inoculated with the contents of two of the miniponds. The 100 percent C02 was again added but with slightly different results than before. Scenedesmus began to grow actively but were soon replaced by diatoms and blue- green algae, which floated to the surface one afternoon and were removed by surface overflow. The Scenedesmus then returned in late July and have since remained as the domi- nant algae. The possible explanations for this change in species composition could be the apparent C02 toxicity discussed earlier in the section on addition of C02 to the miniponds. After July 1969 we were hesitant to try CO2 again because the Scenedesmus culture in the pond was the principal source of material for the separation studies; and, inasmuch as the addition of 100 percent CO2 to the pond might have proved lethal or at least inhibitory to the population of Scenedesmus, no CO2 was added to the pond thereafter. Also, after July the pond was often operated to conform to the needs of the individual separation studies. For example, occasionally the pond was mixed all day to provide a uniform algae supply even though full-day mixing had been found to reduce nitrogen assimilation. Because of this, the nitrogen uptake figures are minimal approximations of the unit's potential. The 1970 operational studies will provide a more realistic evaluation of the potential nitrogen removal capa- bilities in this pond. Biological and Chemical Observations Biomass Production, As used in the present study, the term biomass refers to the results of volatile suspended solids determinations. In many algal growth studies, cell counts, packed cell volume, or light transmittance (or absorbance) are used to estimate changes in biomass. The primary interest at Firebaugh was in biomass because of its possible effect on the growth system and on the market value of the algae produced; therefore, estimates of biomass pro- duction were needed. Volatile suspended and total suspended solids of algal cultures can often be used interchangeably since the ash portion of algae is usually only 10 to 15 per- cent of the total weight. At Firebaugh the volatile portion was usually only 50 to 70 percent of the total suspended material. The nonvolatile solid material consisted of in- soluble complexes of magnesium, calcium, phosphorus, potassium, and iron resulting from the high pH levels attained in the growth units, as well as of some windblown soil particles. Although CO2 addition to the growth units did lower the pH, -70- I apparently no signifi- cant change occurred in the percentage of volatile material in the effluent solids (Figure 4l). The mixing system finally selected for general use in the outdoor growth units caused obvious diurnal fluctuations in the effluent volatile solids. A typical diurnal cycle is illus- trated in Figure 42. In this particular in- stance, the 0830 vola- tile solids (normal sampling time) of 374 mg/1 indicated an unrealistically high value for the amount of biomass produced by the system. Graphical in- tegration of this curve indicated that the aver- age volatile solids concentration for the 24-hour period was ap- proximately 120 mg/1. Because of the diurnal fluctuation caused by algal settling, the total pond biomass increased Indicating that it was accumulating at a rate faster than that at which it was being removed. Volatile solids concen- trations of more than 500 mg/1 often were noted in the growth units and the solids seldom reached steady-state conditions. Actual biomass production during a 19- day period of August- September 1969 was calcu- lated for a minipond in aoi 60- I I 0 5 10 PER CENT CARBON DIOXIDE ADDED FIGURE 41 -AVERAGE AND RANGE- PER CENT VOLATILE SOLIDS OF TOTAL SUSPENDED SOLIDS AT DIFFERENT LEVELS OF COg ADDITION -MrXING PERIODS § 100- SAMPLING TIME FIGURE 42- DIURNAL FLUCTUATIONS IN EFFLUENT VOLATILE SOLIDS -71- FIGURE 43-EFFLUENT NITROGEN AND VOLATILE SOLIDS USED IN BIOMASS PRODUCTION FIGURES run 8. Effluent (0830 samples) volatile solids and total nitrogen are shown in Figure 43. An overnight study of the pond during this period indicated that the aver- age volatile solids in the effluent were about one -half of the O830 values. Both curves in Figure 43 were divided into 48-hour intervals and the average effluent nitrogen and volatile solids calculated for each interval. During this 19-day period, an average of 240 mg/1 volatile solids was dis- charged in the effluent of the pond. Prom this figure, the actual pro- duction of algae per liter of influent can be calculated from evapo- ration data by assuming that one -fourth of the influent was lost through evaporation. Based on this calcu- lation, about 180 grams of volatile material were shown to be produced from each liter of influent. If it is further assumed that algae are about 90 percent volatile, then the actual algal production was approximately 200 mg/1. For the 19-day period, an average of approximately 17 mg/1 of influent nitrate as N was used in the production of the biomass. Chemical analyses showed that the volatile solids were about 8,3 percent nitrogen. These data result in the following nitrogen balance: Cl should = Cde + Cae - (Cai " CAe) + Cx 185 should = 23 + 146 + 6 + Cx 185 = 175 + Cx where Cj = grams N entering system Cde = grams N dissolved in effluent Cae = grams N in effluent algae -72- Cm = grams N In in-pond algae-initial CAe = grams N in in-pond algae -final Cx = grams N not accounted for Although an exact balance was not obtained, the difference, Cx> could be in the range of experimental and sampling error, or could be a measure of nitrogen lost by some mechanism such as denitrification. Based on the information accumulated thus far, it is estimated that the overall biomass yield for the algal system will be on the order of 1 ton of volatile solids per million gallons of waste plus 1/2 ton of inert materials. Algal Genera Noted. The number of algal genera which were noted in the pond cultures was relatively low. The only genera that dominated the cultures are shown in Table 10. As previously noted Scenedesmus quadricauda was originally introduced, but occasionally Scenedesmus dimorpha was noted although never as the dominant algae, ^enedesmus was typically in the two- or four-celled stage, although in very active cultures the alga took a unicellular form with four small spines. The soil ponds went through several algal successions, with some flagellated algae (usually Carte ria, Phacus, or Euglena) as the dominant genera. The soil ponds also often contained large numbers of an extremely small uni- cellular green algae, either Chlorella or Nannochloris. TABLE 10 DOMINANT ALGAL GENERA AT THE lAWTC Genus - Type Scenedesmus green Oscillatoria blue -green pysmorphococcus green Schroederia green Lagerheimia green Navicula diatom Scenedesmus was originally seeded into the RGP in the spring of 1960 and dominated the cultures until early summer of that year. At that time, species of other genera, mainly of Pysmorphococcus and Oscillatoria, began to dominate the cultures and, by August, Scenedesmus had almost entirely -73- disappeared. The RGP was then drained, cleaned, and refilled, and another 100 pounds of Scenedesmus sp. obtained from the Richmond Field Station were added. From then on, Scenedesmus was the dominant species in the RGP, although the culture was killed by CO2 in the summer of 1969 and was reseeded from the miniponds. Summer conditions appeared to be particularly suitable for growth of the blue-green alga, Oscillatoria. During both 1968 and 1969* this alga was prevalent in some minlpond cultures. Scenedesmus reestablished dominance with the onset of cooler weather. All of the algae listed in Table 10, with the ex- ception of diatoms and Lagerhelmia, appeared capable of re- moving large quantities of nitrate— nitrogen. The appearance of large numbers of diatoms or Lagerhelmia was usually ac- companied by a decrease in nitrogen assimilation. Predatory Organisms. Relatively little emphasis was placed on this aspect of pond ecology. However, during the spring and early summer of 1969, some data were collected on potential algal predators. Rotifers were by far the most common of such organisms with populations of 30,000 organ- isms/liter noted on occasion. In ponds with actively growing cultures of algae, rotifers were always present in numbers fewer than 500/llter, Large numbers of rotifers usually were found in ponds with senescent or dying algal populations, but it is not possible to s tate whether rotifers preceded or followed the senescent conditions, Planktonic crustaceans (Copepoda, Daphnla, etc) were notably absent from the algal ponds, and only in the two soil ponds were these organisms prevalent. Although usually fewer than 1,000 Copepod nauplll and adults per liter appeared in the cultures, on one occasion more than 5^000 of these organisms per liter were noted in a soil pond. Dissolved Effluent Nitrogen. In general, the dis- solved effluent nitrogen contained only nitrate, nitrite, and organic nitrogen. Ammonia, if detectable, was almost always less than 0,1 mg/l, compared to an undetectable amount occurring in the influent. Biological activity in the mini- ponds did cause dissolved organic nitrogen to increase from about 0,4 mg/l to 0,7 - 0,8 mg/l. Dissolved organic nitrogen usually was higher in the soil ponds than in the remaining miniponds. In minlpond run 8 conducted during August and September 1969, dissolved organic nitrogen in all the mini- ponds was consistently above 1 mg/l and occasionally greater than 1,5 mg/l. The concentration of nitrite-nitrogen followed a consistent pattern in the outdoor growth studies. After the ponds were filled, nitrite generally increased to 1 to 2 mg/l -74- ^ and then, as the nitrate decreased, nitrite followed the same trend. This pattern is illustrated in Figure 44, which shows a minipond with a high percentage of total nitrogen removal. The ponds with low effluent nitrate usually also had a low nitrite concentration. On several occasions, effluent nitrate-nitrogen concentrations were determined at two- or three-hour inter- vals for a 36-hour period. The variations in concentration were beyond the limits of the accuracy of the analysis; thus, a diurnal change in the concentration of this constituent could not be demonstrated. Dissolved Oxygen, This parameter was not routinely monitored, although determination of dissolved oxygen con- centration was a part of most diurnal studies. An example of the diurnal fluctuations in dissolved oxygen in four mini- ponds (data from September 10-11, I968) are shown in Figure 45. Ponds 1, 4, and 22 each showed approximately 75 percent total nitrogen assimilation, whereas pond 10 showed only about 40 percent. The greatest fluctuation was noted in pond 22, an unmixed soil pond. Undoubtedly this change occurred because the afternoon mixing tended to deoxygenate the pond water in the remaining ponds. The 5 mg/l minima noted in ponds 1, 4, and 22 were near the minimum concentra- tions recorded in any of the overnight studies. It is Interesting to note that the deoxygenation factor for the ponds is on the order of 1,25 mg/l/hr, whereas the oxygenation factor is a net of 2,5 mg/l/hr, or a gross of DAYS OF RUN FIGURE 44- CHANGE IN NITROGEN FORMS DURING MINIPOND RUN -75- a CJ » c o •H ^- •P 0) Kl E E >> (1> -H cd -P -P Q a o o> un in on o ^ o on t^ in m m oj on in ^- CO CO in co ON 00 t^ o OJ OJ a^ CO Lr\ -=1" OJ cv o 0\ 00 • • t>- 00 00 in on on in lO ^ on I^ ^ en \o vo o> l>- r-l iH iH iH 0^1 in cr\ CM OJ CM ^ • • • • • CTv o\ c^ (y\ t~ in in in CM OJ o o rH o CJ g u •H ^ ^ ■P rt o^ in C o o o iH o ■P C > D ffi faO CO E 0) o D -P o E CO •H ti-H >J E rH !h 3 3 05 rH CO E < •H • • •• & E O s z rH \ t>0\ E 0O| ■" — »: ^ \ +J E CVJ O O CD ^ Q-i 4J Sm CD W ?H 4J E <; xx x;xx!>< X XI XXX xxxxx XX X X I oo o oo o e • • • • • rH rH rH rH rH rH 1 1 1 1 1 OJ OJ ifMnLnmLn i i inOJ OJ CM OJ OJ rHrH o o o o o o o O 'O .^ • ^ I rHCX) I in I I LHOJ in-::^- o o o o o o o oo • • o • • • • ^ r-{ T-\ r-i t-\ t-{ CO I I rH I I I I I inLPv I LnifMntn ^ CM OJ inoj OJ OJ OJ • •••••• o o o o o o o LTvLn CO o m O 1 1 1 rH 1 rH mm ^ 1 rH 1 • • m mo o m o • o o I m o o * * * * * * * o < <: < momm^-^ rH o mvo O rH I I I rH mVD OJ OJ CVI OO CV^ ^ MD o o o CO m'm^ moo oo oo 0)0)0000 oooo •P -4-3000000000000 O >5>,rHrHrHrHOOOrHrHrHrHO O rH rH <;-i <^ <;^ ti-i rH rH rH (m CD CO CO E E rH CO CO rH >5 >3 •H O O E E rH CO CO CD O O ^ -H -H 0) Sh ?^ E (1) 0) 0) E E o <; < E E CO CO E E CO CO • >j >j & O O ?H O E E O CO CO O O E •H -H O U U W 0 0) rH E E CO ..^ — » iHO) >j O tiD-P+^+>-PHH Pi 0) a o v H •H ■H • O O O fH ^^ •r4 -H O O ft C C C CO CO H to CO CO CO 0) 0) (U bO bO^ CO ^ fi ^^^■g r-j O 43 (0 ^ CO to s S p< s ? ^ CD W 3 3 3 3 o o o > rH H rH H rH 0) (U o o '^ £^ (U 0) 0) 0) (U t) Ti o o >? ^ fi t, Jh fn P 0) o fn rH 0 U a d CO -io6- TABLE 22 AMIND ACID COMPOSITION OF OVEN-DRIED RGP PRODUCT ALGAE : Sample 1 - 10/15/69 : Sample 2- 1/5/70 Amino Acid : Percent of Total :mg N/mg Algae ,: X 10"^ rPercent of Total :mg N/mg Algae : Amino Acid Cone. : Amino Acid Cone: x 10"^ Lysine HistidineV Argininel/ Tryptophani/ 7.6 3.2 5.7 1.1 2.2 1.3 1.9 0.5 7.8 5.5 5.1 1.6 1.1 0.3 1.0 0.1 Aspartic Acid 10.5 2.4 11.9 1.3 Threonine!/ h.9 1.3 5.4 0.6 Serine h.l 1.4 6.0 0.8 Glutamic Acid 10.5 2.2 13.1 1.2 Proline 5.7 1.5 5.2 0.7 Glycine 6.6 2.7 7.1 1.3 Alanine 9.5 3.2 8.4 1.3 Valine!/ 5.8 1.5 6.1 0.7 Methionine!/ Isoleucine!' Leucine!/ Tyrosine!/ Phenylalanine!/ 1.8 0.5 1.8 0.2 3.7 0.9 4.2 0.4 8.8 2.1 8.4 0.8 3.5 k.8 0.6 1.0 3.4 0.2 5.6 0.5 Total 99.5 31.6 100.3 13.3 —Essential amino acid. 20 percent more of the crude protein than could swine. The study concluded that, although algae were not a high energy food, its high protein content and available minerals made it a potentially useful food supplement. Erchul and Isenberg (1968) evaluated the protein quality of various algal bio- masses produced at a water reclamation experimental pilot plant. The products were mainly a mixture of Chlorella and Scenedesmus fed to rats to compare protein efficiency ratios (PER) to casein, soybean, and fish meal supplements. The algal biomass with the highest PER compared favorably with soybean meal. The variation in PER was apparently due to differences in digestibility. Leveille, et aj^ (1962), reported on the use of various algae (Chlorella, a mixture of Chlorella and Scenedesmus and Spongiococcum) as the sole protein source in rat and chick feeding studies. All the algae were inferior to soybean meal as a source of protein. The data showed them to be deficient in methionine for rats and chicks and the mix- ture to be deficient in glycine for chicks. Grau and Klein -107- ^unpublished progress report, 1956) fed sewage-grown algae (Scenedesmus-Chlorella mixture) to chicks and found that growth was generally slower when algae was used in place of soybean meal. The use of alum-flocculated algae appeared to be especially detrimental to chick growth when the concentra- tion of aluminum in the feedstuff exceeded 0.5 percent of the chickens' total food intake. Researchers from the North American Aviation Company (19^7) reported on the feeding of mixtures of Scenedesmus, Closteridlum, and Chlorella (grown on sewage) to chickens as a substitute for soybean meal. Their results indicated that the protein value and feeding efficiencies of algae were comparable to soybean meal. The algae may actually be preferred by the chicken rancher because the algal pigment, xanthophyll, gives a deeper golden color to the skin, meat, and egg yolks. The algae grown at Pirebaugh may be more suitable for poultry production because the salts included with the algae might provide some essential nutrients to the fowl. (Most poultry diets contain a small percentage of certain salts.) Information in the preceding paragraphs indicates that algal meal may find a market as a substitute for such protein supplements as soybean, cottonseed, or fish meal, especially in poultry production, although more data are needed. The next step was to determine whether the potential market could absorb the production of the proposed algal stripping plant. In Table 23 the estimated maximum and mini- mum algal production is given for each five-year interval from 1975 to 2000. The data are based on projected drainage flows from the entire San Joaquin Valley and estimates of algal production per unit flow. The basic assumption is that the total drainage from the Valley will eventually require treatment. The data indicate that peak annual production may be between 66 x 103 and 83 x 103 tons of dry algae. Accord- ing to the California Department of Agriculture (19§5)* the California poultry industry used more than 56O x 10^ tons of protein supplement in 1963* while the total used by all feed- ing operations was almost one million tons. The prices paid for these animal feed supplements varv from $8o/ton (cotton- seed meal^ and $100/ton (soybean meal) to $110/ton (meat and bone meal). If, as the tests indicate, algae can readily be substituted for soybean meal, the value of the by-product will be as shown in Table 23. Use of Algae as a Soil Conditioner Another market which can absorb the production of a full-scale algal stripping plant is the retail market for soil supplements to lawns, golf greens, etc. Its high -108- TABLE 23 ESTIMATED ALGAL PRODUCTION BY AN ALGAL STRIPPING PLANT, 1975-2000 Year Thousands of Tons of Algae per Year Maximum Minimum Approximate Value as Substitute for Soybean Meal Maximum : Minimum 1975 13,300 8,410 $1,330,000 $ 841,000 1980 27,200 18,000 2,720,000 1,800,000 1985 44,300 29,610 4,430,000 2,961,000 1990 62,000 42,400 6,200,000 4,240,000 1995 75,300 54,100 7,530,000 5,410,000 2000 82,610 65,510 8,261,000 6,551,000 nitrogen content and various salts in the product, combined with its slow rate of decomposition, make algae a desirable lawn conditioner. The current wholesale price of a soil conditioner manufactured in San Jose, California, is about $100/ton. This figure is the cost of 50-pound sacks of the product Intended to retail at $l4o/ton, 1970 prices, in Fresno, California. Railroad freight agents were hesitant about quoting shipping costs: however, the maximum prices would be as shown in Table 24 (shipped in 100-pound sacks). The exact shipping costs, which will be determined when algal product is given a freight classification, will have to be deducted from the value of the algae. Sewage plant operators have tried to develop mar- kets for their dried activated sludge, which is mainly com- posed of nonbiodegradable polysaccharide, protein, fat, and Inorganic complexes (Hurvitz, 1957). According to a report by Foster D. Snell, Inc. (1969), the market for this product has decreased because such material is not competitive with other nitrogen sources, and many plants are unable to recover freight costs, let alone process costs. An algal product with higher nitrogen content (6 to 8 percent) than activated sewage sludge (l to 2 percent) may find a more receptive market. -109- TABLE 24 RAIL FREIGHT RATES FOR 100 LB. ALGAE PACKAGES PROM SAN FRANCISCO, CALIFORNIA Destination : Amount Per 100 Lbs. : Amount Per Ton Los Angeles $1.30 $26.00 Portland 1.76 35.20 New York 4.31 86.20 Kansas City 3.23 64.60 Fresno .82 l6.40 Tacoma, Washington 1.98 39.60 Atlanta, Georgia 3.94 68.80 Miscellaneous Possible Markets for an Algal Product Some other possible markets exist for algae, but current technological difficulties preclude their consider- ation as important market outlets. A brief outline of several of these uses follows: 1. Some interest has been expressed in feeding algae to organisms reared commercially in aquacultures. These organisms Include freshwater fish and marine bivalves (clams and oysters). Development of these aquatic farms is now mostly in the talking and planning stages and could develop into a substantial market for algae. 2. Borgmann and Feeney (1948) isolated a sterol from Scenedesmus obliquus which had charac- teristics identical to those of chondrillasterol, a sterol isolated from a sponge. This algal sterol is one of the few naturally occurring compounds having the necessary configuration to serve as a starting point in the manufacture of cortisone, an important drug. More research will be required to determine whether this sterol is present in commer- cially valuable quantities in Scenedesmus quadricauda used in this project. 3. Use of algal protein in the adhesive industry has been suggested as a market for the product. Samples of Chlorella sent by Fisher and Little (1953) to a private laboratory to investigate -110- the possible uses of the protein fraction in algae showed that the protein was difficult to separate and that the molecular weight was too low to serve as a substitute for casein. Most chemical adhesive manufacturers are using synthetic compounds for adhesive production and show little Interest in natural organic sources. Use of Algae to Produce Methane Gas Suggestions have been made that an algal product can be converted to usable energy through methane fermentation, with subsequent use of the methane gas to provide heat and/or electric power. In a comprehensive paper on this process, Oswald and Golueke (196O) concluded that this could be a practical means of converting solar energy to thermal or elec- trical energy. The problem with this type of system in tile drainage is the high concentration of sulfate present. This amounts to about 3^000 mg/l in the Alamltos tile system. Foree and McCarty (1968) studied the anaerobic decomposition of algae and concluded that, as long as dissolved sulfate is available, organic matter decomposes through the reduction of the available sulfate ion. Further stabilization of the material then occurs by methane fermentation, as long as the sulfide concentration is not toxic to the bacteria. Little or no methane was produced in the decomposition of algal cultures containing about 2,000 mg/l dissolved sulfate. The presence of this inhibitory action may preclude the use of this method of algal disposal in the projected agricultural wastewater treatment plant, but an investigation would be re- quired to ascertain the feasibility, methodology and problems associated with such a system. -111- CHAPTER VI - PROCESS EVALUATION The studies at the lAWTC have shown that the algal stripping process can be a reliable means of reducing the nitrogen concentration in agricultural tile drainage. Because the algal process involves more than one step, and because each step may have alternatives, this section of the report will assign tentative estimates of nitrogen removal at each step, as well as total nitrogen removal by the system. Also, the section will include some preliminary design criteria and estimates of treatment costs based on facilities built and operated according to these criteria. Removal Efficiencies As shown in Figure 29* the algal cultures assimi- lated an average of 80 percent of the total influent nitrogen (80 percent of 20 mg/l, or about I6 mg/l) from July through December I969. This period, which covered the temperature range noted at the site, was selected because of standardized conditions in operation of the growth units. This 80 percent average is somewhat misleading because it includes data from the start-up period of the run — algae growing at steady- state conditions assimilated about 90 percent, or I8 mg/l, of the influent nitrogen with an effluent of 2 mg/l, or less, total nitrogen. The amount of nitrogen assimilated by the algae increased steadily as the study progressed, probably because of more complete understanding of the system and natural selection of a strain of algae more adapted to the waste and to the environment of the San Joaquin Valley. The average amounts of influent nitrogen assimilated by the (best) pond during steady-state conditions for all minlpond runs are shown in Figure 54. During I969 from I6 to 21 mg/l of nitrogen were assimilated, as compared to 7 to 10 mg/l in the earlier runs. Preliminary (1970) operational data obtained at the time this report was being prepared indicate that the algal system can remove 27 to 28 mg/l from a 30 mg/l nitrogen influent. Based on data from the feasibility phase of the project, minimum nitrogen assimilation (20 mgN/1 influent) will be 16 mg/l and maximum assimilation about I8 mg/l. If 95 percent of the algae can be separated from the liquid phase (a level consistently achieved bv the f locculation- sedimentation unit tested at the lAWTC), then the total nitro- gen remaining in the effluent would be on the order of: -113- Nip = Nj) + N^ where Nrp = total effluent N; Np = dissolved effluent N; and N^ = nitrogen in effluent algae or (5^ algae remaining) (original algal conc.)(^ N in algae ) Maximum Nrj, = ^ mg/l + (O.05 remaining) (250 mg/l) (0.08 N in algae) = 5 mg/l N, Minimum = 2 mg/l + (0.05 remaining) (250 mg/l)(0.08 N in algae) = 3 mg/l N 22n leT 20- 18- w (^j (6) ■■('i')" 16- (^B) (6) £ 14- (9A) 0 kJ 5 12- < *(IA) 2 '^ 0 0 0- 0 m-'i^f — WV" "4B" q: z 6- 4B -J < 0 4- 1968 & 1969 NOTE; NUMBERS IN PARENTHESIS ARE RUN NUMBERS 0- r 1 — 1 r — — T 1 r 1 — 1 — FEB MAR APR MAY JUN JUL AUG MONTH SEP OCT NOV FIGURE 54- NITROGEN ASSIMILATION IN VARIOUS MINIPOND RUNS -114- These figures were based on 20 mg/l total N in the Influent, Thus, overall nitrogen removal efficiencies will range from 75 to 85 percent. Depending on water temperatures, growth unit detention times of 5 to I6 days will be required to achieve these removal rates. Process Configuration From the data contained in this report, a tentative process configuration can be drawn. The configuration is preliminary in that data from the operational phase of the study were not considered. Figure 55 Illustrates the flow diagram of an algal stripping plant. Shallow growth units are followed by a sedimentation process which removes 95 percent of the suspended biomass. The effluent from the sedimentation tanks leaves the plant site and the algal slurry goes to a vacuum filter for dewatering to about 20 percent solids. The effluent from the dewatering device is recycled to the head of the sedimentation area while the sludge is dried to 85 to 90 percent solids by air or flash driers. The dried product is then re- moved from the site to be ^_^ sold as a marketable by- product. a C02 ADDITION Aolm™ z:i"s PLANT SEDIMENTATION INFLUEN f TANKS '— DE^ATEHINC Figure 56 illus- trates a schematic design of an algal growth unit. The flow enters and leaves the unit by gravity •"«?^«5tf through concrete pipes. Iron IDhOSDhorUS and figure ss-flo* diagram of algal stripping plant CO2 are added to individ- ual growth units at the mixing pumps. Outdoor propeller- type pumps perform four hours mixing per day. To reduce the required number of mixing pumps, it is proposed that the ponds be constructed in units of 12 with a central mixing station capable of mixing two ponds at a time for four hours. This type of design is shown in Figure 57. A series of auto- mated valves would rotate mixing to the desired ponds. An in-line sedimentation area Included in the design of the growth unit will remove much of the heavier suspended solids. Cost Estimates The feasibility studies conducted at the lAWTC were not designed to provide definitive cost data for an algal stripping system; however, some preliminary process cost esti- mates can be made. These estimates will undoubtedly be revised as more data become available from 1970 operational studies. -115- This is especially true in the case of pond mixing. Results demonstrated that in- organic carbon addition may be more important than mixing. Predicted seasonal variation in flows and nitro- gen concentrations of tile drainage from the San Joaquin Valley (see Figure 58) and climatic factors such as temperature and sunlight indi- cated that April was the criti- cal month for sizing the growth units. During this month, flows will be high, nitrogen con- centration should be about 30 mg/l, and sunlight and temper- ature conditions will not be optimum. Several factors were used in obtaining these esti- mates: 1, All costs were based* on January 1970 dollars. 2, Debt service was calculated for 50 years at 5 percent. STRUCTURE^. ^ IN-LINE SEDIMENTA- TION TANK 20Cf- ^ PLAN VIEW FIGURE 56-ALGAE GROWTH POND 3, Costs per million gallons treated were calculated by dividing the total annual cost (debt service plus oper- ations and maintenance costs) for a full-capacity plant by the estimated annual flow of the San Luis Drain. 4, An engineering and contingency factor of 57 percent was assumed for capital cost items whose design and operation was based on experimental data obtained at the treatment center, A lower engineering and contingency factor was used for items common to waste treatment facili- ties and for those items which can be purchased from the manufacturer. These engineering and contingency factors are shown in Table 27, 5, The land required for a full-capacity plant is to be purchased at the beginning of the project at $500 per acre. -116- Tf IN-UNE SEDIMENTATION TANK Vt 100 TYPICAL POND NUMBER ■MIXING PUMPS rCl2) FIGURE 57- ONE OF THREE 12-POND GROUPS CONSTRUCTED PER PHASE -117- I/6UJ -N0llVdlN3DN0D N390ailN O o ir t z «< 11 uj ZJ s< >- UJ 5 U- 3 z o O -3 < < tt: {/) Si < ll_ o < Ul (/) O UJ I- o o UJ q: Q. 00 m UJ oc (O bi o AAOTJ "IVnNNV IVIOI JO 1N30 hid -118- 6, The laboratory and office buildings and main- tenance and storage areas are basically the same as those at the San Jose, California, Sewage Treatment Plant. Capital costs for these facilities are based on information published by Guthrie (1969). 7. Electric power costs were calculated on the basis of $.Ol/lcw-hr. 8, General plant operation and maintenance (O&M) costs were based on curves published by Smith (I967 ) for trickling filter O&M. It was assumed that the general plant O&M costs included replacement of power costs for all plant operations except for the items indicated in Table 26. 9. The cost of reaeration was taken from figures published by Smith (1967). The individual ponds are designed according to the criteria presented in Table 25, with a mean channel length of approximately 6,500 feet. The influent and effluent struc- tures, the in-line sedimentation tank, and flow-directing vanes are all made of concrete. In addition, wherever a con- crete structure and the growth pond proper meet, a 20-foot concrete apron is built. Although an in-line sedimentation tank was not built into the rapid growth pond operated at the lAWTC, evaluation of a sedimentation unit used in conjunction with the growth pond indicated that a device of this type would be beneficial to algal growth. The in-line sedimenta- tion system included in this proposed pond design is routinely used in operations requiring sedimentation facilities but may not be the best solution of the problem of removing excess suspended material from the growth unit. With the exception TABLE 25 ALGAE STRIPPING DESIGN CRITERIA April Growth Pond Hydraulic Detention Time 8 days Growth Pond Depth 1 foot Growth Pond Width 200 feet Mixing Velocity 1 fps Duration of Mixing 4 hrs/day In-line Sedimentation Tank Surface Loading 5,000 gpd/ft^ Detention Time 15 minutes Growth Chemical Pe 2 mg/1 P 2 mg/1 CO2 800 lbs/million gal. -119- of the structures and aprons, the pond bottoms are assumed to be 22 percent soil cement and 78 percent compacted native soil by volume. The ratio is intended to prevent seepage into or out of the ponds. The levee which divides the pond is 2 feet high and has a crest width of 8 feet wide and 1.5-to-l side slopes. It is constructed from the soil excavated from the influent and effluent structures and the in-line sedi- mentation tank. The flow-directing vanes are 2 feet high and 8 Inches thick and built on 50-, 100-, and 150-foot radii. Ponds are separated from each other on the long sides by a 3- foot high levee with a 20-foot crest and 1.5-to-l side slopes. The surface loading on the f locculation-sedimentation tank is considered to be 900 gpd/ft^, with 10 mg/l ferric sulfate and 0.5 mg/l cationic polyelectrolytes (Calgon - Cat- Floc) as the chemical additions. Loading on the vacuum filter is 0.2 gpm/ft2, with capital and O&M cost estimates supplied by the Eimco Corporation. One-half of the sludge from the vacuum filter goes to 7*000 pounds H20/hr driers, and the remaining half is air-dried. Cost estimates for the driers were provided by the De Laval Separator Company. Table 26 lists the capital costs of the various com- ponents of the algae stripping system and Table 27 illustrates the total costs of these components for treating a million gallons of tile drainage. The estimated net cost of treat- ment is $135 per million gallons, which includes the recovery of some money through the sale of an algal product. The product recovery figure was based on producing l400 pounds of algae per million gallons of water and selling the product at $60 per ton. The costs shown in Tables 26 and 27 were obtained by assuming that the plant would be built to treat the estimated maximum annual quantity of about 50,000 million gallons for the San Luis Drain. Because the estimated ultimate annual flow of the facility will not be attained until about 30 years from the time the initial drainage reaches the treatment plant, another approach would be to construct the plant in stages paralleling the buildup of drainage flows. For a treatment plant constructed in phases of equal size (each additional phase to be constructed during the year when drain flow reaches 90 percent of existing plant capacity), the average net cost for treatment over the first 50 years of operation would probably be higher than the estimate shown above. This in- creased price is because part of the plant would be idle at all times. These preliminary cost estimates were obtained to determine which aspects of the process were the most costly. With this information, the operational studies of 1970 were -120- TABLE 26 TREATMENT COSTS FOR ALGAE STRIPPING Item Dollars Per Percent Million of Gallons Total Cost Capital Production Growth Pond 55 30 In-line Sedimentation Tanks 5 3 Chemical Feed Pumps 1 1 Mixing System 39 22 Harvesting Separator 1 1 Dewatering 1 1 Drying 6 3 Other Land, Building, Miscellaneous 6 3 Annual O&W Costs General 21 12 Drying, Dewatering & Growth Chemicals 27 15 Mixing Power 16 9 TOTAL $177/ing 100 Minus By-product Income -42 Net Cost of Treatment $135/Million gallons directed toward reducing the cost of the more expensive items (mixing, land preparation, etc.)* while maintaining the maximum level of nitrogen removal. The cost estimate presented in this section was based on high-rate algal growth units. The other algal system studied at the site — the "symbiotic", soil-lined ponds — has not had a cost estimate, primarily because of the unknown nature of the systems involved. Data obtained during the 1970 operational studies will provide some information on the ques- tions of reliability and the mechanisms involved. With this data cost estimates can be prepared. Because the system is simple and because it requires no mixing and little chemical addition for growth, costs for this process will be consider- ably lower than for the high-rate algal growth system. -121- CVI S < K O Eh CO O O < Eh H Oh < a o u -P •H ■D C 0 O. X w -p •H ft CtJ o o o oo o o oo o oo o oo o o o o •\ •> •> •! oo o o o mo •\ •» •^ oo OJ o o o in o>o r-i in inincvj •» n •% OJ OJ iH o in i>- •V •> in CO pHVO O OO o o o o o o O O C-- o in in O C^rH ii c a EH Kl E a O E •H 3 c +3 O, o K) •H Jh -D E ■P Kl W CO > J^ E 3 Kl E E 0L^ rH CO bO O "H CT bO td o 0) cd E -P ?M O E pH •H x: E O bO •H td 0 bO < -H H 4^ ^ ^ ^ E +> {^H-P E -D 0 O 5 iH E -H K) cd Cd "H TJ rH O 3 O 1 (U X 0 ft 3 >, ?H E -H Kl •o ?^ E x: -H > 0 s U 0 cd 3 ^ 0OHO2 ?H CO « Xi fn cd +^ li. W o rH OJ cn^ invD r- 00 a\o •D E • E O •P (d m o bO •o cd o E E B tH cd 05 'O U rH E 0 bOo (0 o 0 E 0 rH o E •HfH o ^ U O •» bO 0- E 0 O in 0 E T3 o •H 3 0 •> •P bOcd tH t- E E 0 rH o 0 0 U ft rH O 3 ft -«3- U ■ppq (0 I E 0 • OcO CQ Cd in ?H ? OJ fttZ) (U rH (d o (d t^ 0 -p -P 0 inx: o p -o 4-» • cd Eh 3 Cd 0 s-P O 0 JQ CO >, E •D ^ O •H 3T3tH r-i 0+> E 0 ON ox: M bO E CQ 0 B -D •HtH ^ E C: -P (d o u 0 E rH 3 O O C^ <»H E O •V • -o cd VJD ?H E U CO •o •\ O cd o CQ J3 LA-P ■PnH cO •» o -=r o CO oo cd cd C bO •* > >>o ^ • o • o 0 0 Kl E 03 E U 0 o 0 O 0- E s bO S bO Ti E E w bO CO •H D3 nH+> E E -P E -P-H 0 0 E 0 E •P O •P O E O H O H o-H ^ m 0 P O -122- CHAPTER VII - AREAS FOR FUTURE INVESTIGATION Many of the questions concerning ^ chnlcal feasi- bility have been answered by the study at ^'^ .ebaugh and other answers will result from current operatlondl studies. There are, however, areas of Interest which cannot be fully Investi- gated during the current project and which warrant further study. The proposed studies will be more meaningful when a combined drainage facility provides larger quantities of water. Among the areas of Interest are: 1. A redesigned algal system which may reduce the land area and modify the type of growth units necessary for nitrogen removal. The basic idea for this type of system was derived from personal communication with Dr. William J, Oswald, Professor of Environmental Health Sciences and Sanitary Engineering at the University of California, Berkeley. The type of unit suggested by Dr. Oswald would consist of a number of ponds in series, each containing different depths of water. The first pond would be relatively deep, with light as the factor limiting algal growth. This pond would only receive phosphorous addition and would not require mixing. The next pond would be shallower but light would still be the only limiting factor. The last pond in the series would be similar to the rapid growth pond used in the present study (shallow, with mixing, etc), and nitrogen would be the constituent limiting growth. Nitrogen assimilation in the earlier ponds would permit the use of short detention times in the final unit. Theoretical considerations indicate that a system of this type could reduce the land area required in a conventional system as much as 50 percent, 2, Use of the drain as a treatment system has been described by Goldman, e_t al (1969), in which the drainage canals and holding reservoTr are part of the process. Because of nutrients contained in the water, the optical clarity of the drainage and the use of relatively shallow, lined canals, some natural algal growth (either planktonlc or sessile), estimated at 15 to 60 mg/1, will occur to remove some nitrogen. If this growth is encouraged by nutrient additions and/or modification to the canal to in- crease turbulence, as much as 100 mg/1 of algae could possibly be grown in the canal prior to Kesterson Reservoir (a holding reservoir). This could remove about 8 to 9 mg/1 nitrogen. Then if Kesterson were redesigned to include both holding capability and improvements for increasing algal growth, it could conceivably remove the remaining nitrogen. A separation -123- plant would then be placed at the outlet of the reservoir. If these facilities are Included in the drainage system, considerable savings in land costs may be realized. 3, Use of the algae growing capability of the drain and of Kesterson Reservoir with the anaerobic denitrif ication filter process as a final polishing step. In the anaerobic system, an organic carbon source is added to the drainage and the water passed through a column containing some type of aggregate. Anaerobic bacteria attached to the media reduce the nitrate to nitrogen gas, which is evolved to the atmosphere. Studies at Firebaugh have demonstrated that algae entering the columns can increase dissolved organic and ammonia nitrogen, presumably by bacte- rial decomposition of algal protein. Since algae undoubtedly will grow in the drain, some provision for their removal will have to be incorporated in the design of an anaerobic system. If algal growth is encouraged, the dual system may reduce costs and provide a more reliable system for nitrogen removal. An unpublished communication from Dr. Oswald contains refer- ence to the following possible benefits from the dual system: a. Back-up coverage is provided in case of breakdown of one of the systems. b. The algal system can remove ammonia nitrogen produced by decomposition of bacterial dells in the filter. c. Recreational use may be provided by the ponds. -124- ACKNOWLEDGMENTS The following list includes the personnel actively involved in the algae stripping project at the lAWTC. Algae Stripping Project Under the Direction of Louis A. Beck .... Engineer, California Department of Water Resources Donald Swain Engineer, U. S. Bureau of Reclamation Percy St. Amant Engineer, Environmental Protection Agency Conducted by Randall L. Brown . . Biologist, California Department of Water Resources Bruce A. Butterfield . Engineer, California Department of Water Resources Joel Goldmaji Engineer, California Department of Water Resources James F. Arthur Biologist, Environmental Protection Agency James R. Jones Engineer, U. S. Bureau of Reclamation Assisted by Robert G. Seals Chemist, Environmental Protection Agency William R. Lewis . . . Chemist, California Department of Water Resources Dennis L. Salisbury Engineering Technician, California Department of Water Resources William L. Baxter . . . Laborer, California Department of Water Resources Clara P. Hatcher Laboratory Aid, California Department of Water Resources Elizabeth J. Boone Laboratory Aid, California Department of Water Resources Linda S. Harrington Laboratory Aid, California Department of Water Resources Consultants to the Project Dr. William J. Oswald University of California, Berkeley Dr. Clarence G. Golueke University of California, Berkeley Dr. Perry L, McCarty Stanford University, Palo Alto Report Prepared by Randall L. Brown California Department of Water Resources -125- LIST OF REFERENCES 1. Aach, H. G. 1952. Waschstrum und Zusammensetzung von Chlorella pyrenoldosa. Archiv fur Mikrobiologie, 17:213-246. 2. American Public Health Association. 1965. Standard Methods for the Examination of Water and Waste Water, 12th edition. 3. Bergmann, W. and R. J. Feeney. 1950. Sterols of Algae. I - the occurrence of chondrillasterol in Scenedesrnus obliquus. Jour. Org. Chera. 15:8l2-8l4. 4. Black, A. P., F. B. Birkner, and J. J. Morgan. I965. Destabilization of dilute clay suspensions with labeled polymers. Journal American Water Works Assoc. 57:1547. 5. Bogan, R. H., 0. E. Albertson, and J. C. Pluntz. I96O. Use of algae in removing phosphorus from sewage. Pro. Am. Soc. Civil Engrs., J. Saht. Engr. Div., SA5, 86:1-20. 6. Bohna Engineering and Research, Inc. 1970. Unpublished report entitled - Filtration of algal suspension. Submitted to California Department of Water Resources. 7. Brown, R. L., R. C. Bain, and M. G. Tunzi. I969. The effects of nitrogen removal on the algal growth potential of San Joaquin Valley agricultural tile drainage effluents. Presented at the Fall Meeting of the American Geophysical Union, December 15- I8, I969. 8. California Crop and Livestock Reporting Service. I965. Poultry and Hatchery Production in California, Summary for 1963. 9. California Department of Agriculture. 1970. Applied Botulism Research, including Management Recommendations Progress Report. 87 pages. 10. Chemical Rubber Publishing Company. 1951. Handbook of Chemistry and Physics. 32nd edition. 11. Cramer, M. and J. Meyers. 1948. Nitrate reduction and assimilation in Chlorella pyrenoldosa. J. Gen. Physiol. 32:93-102. 12. Erchul, B. A. F. , and D. L. Isenberg. I968. Protein quality of various algal biomasses produced by a water reclamation plant. J. Nutrition. 95:374-380. -126- 13. Fair, G. M., J. C. Geyer, and D. A. Okum. 1968. Water and Wastewater Engineering, Volume 2, Water Purification and Wastewater Treatment and Disposal. JoJ n Wiley and Sons, Inc~ New York. 14. Fisher, A. W. and J. S. Burlew. 1953. Nutritional Value of Microscopic Algae. In Algal Culture from Laboratory to Pilot Plant. J. S. Burlew, (ed.). Publ. bOO, Carnegie Inst, of Wash. pp. 303-310. 15. Fisher, A. W. and A. D. Little. 1953. Algae as Industrial raw materials. In Algal Culture from Laboratory to Pilot Plant. J. S. Burlew, (ed. ) . Publ. bOO, Carnegie Inst, of Wash. pp. 311-315. 16. Fitzgerald, G. P. I96I. Stripping effluents of nutrients by biological means. Transactions of the I96O Seminar on Algae and Metropolitan Wastes. Robert A. Taft, San. Engr. Ctr. Cincinnati, pp. 136-139. 17. Fogg, G. E. 1947. Nitrogen Fixation by Bluegreen Algae Endeavor. 6:172-175. 18. Fogg, G. E. And D. M. Collyer. 1953. The accumulation of lipides by algae. In Algal Cultures from Laboratory to Pll^"t Plant. J. S. Burlew, (ed. ). Publ. bOO, Carnegie Inst, of Wash. pp. 177-181. 19. Fogg, G. E. and M. Wolfe. 195^. The nitrogen metabolism of the blue-green algae (Myxophyceae) . Symposium Soc. Gen. Microbiol. 4:99-125. 20. Foree, E. G. and P. L. McCarty. I968. The decomposition of algae in anaerobic waters. Technical Report No. 95 j Stanford University, Dept. of Civil Engr. 21. Foster D. Snell, Inc. I969. Feasibility of hydrolysis of sludge using low pressure steam with SOo as a hydrolytic adjunct and utilization of resulting hydrolysate. Report prepared for the Fed. Water Qual. Admin. 22. Gest, H. and M. D. Kamen. 1948. Studies on the phosphorus metabolism of green algae and purple bacteria in relation to photosynthesis. J. Biol. Chem. 176:299-318. 23. Goldman, J. C, J. F. Arthur, W. J. Oswald, and L. A. Beck. 1969. Combined nutrient removal and transport system for tile drainage from the San Joaqain Valley. A paper presented before the National Pall Meeting of the American Geophysical Union, December I5-I8, I969. San Francisco. -127- 24. Golueke, C. G. and H. B. Gotaas. 1958. Recovery of algae from waste stabilization ponds - II. U. of California (Berkeley) lER Senes 44 - Issue 8. 25. Golueke, C, G., W. J. Oswald, and H. K. Gee. Harvesting and Processing Sewage-Grown Planktonic Algae. 1964. San. Eng. Res. Lab., U. of California, Berkeley. SERL Report No. 64-8. 26. Grau, C. R. and N. W, Klein. I956. Use of algae as a poultry feedstuff. Unpublished Progress Report for Fiscal Year 1955-56. Submitted to U. of California, Berkeley. 27. Guthrie, K. M. I969. Capital Cost estimating. Chem. Engr,, 76:6. 28. Hansen, S. P., G. L. Culp and J. R. Stukenberg. I969. Practical application of idealized sedimentation theory in wastewater treatment. J. Water Poll, Contr. Fed., Vol, 4l, No, 8, Part 1: l421-l444, 29. Hemens, J. and M, H. Mason. 1968. Sewage nutrient removal by shallow algal stream. Water Research. 2:277-287. 30. Hintz, H. F., H. Heitman, Jr., W. C. Weir, D. T. Torell, and J. H. Meyer. I966. Nutritive value of algae grown on sewage. J, Animal Science Vol. 25. 3^675-681. 31. Hurvitz, E, 1957* The use of activated sludge as an adjuvant to animal feeds. Engr. Bull., Ext. Series No. 94:395-414. 32. Ives, K, J, 1956. Electrokinetic phenomena of planktonic algae. Proc. Soc. Water Treat., Exam. 5t4l. 33. Jewell, W. J. and P. L. McCarty. I968. Aerobic decomposi- tion of algae and nutrient regeneration. Technical Report No. 91. Stanford Univ. Dept. of Civl Engr. 34. Ketchum, B. H. 1939a. The absorption of phosphate and nitrate by illuminated cultures of Nitzschia closterium. Am. J. Botany. 26:399-407. 35. Ketchum, B, H, 1939b. The development and restoration of deficiencies in the phosphorus and nitrogen composition of unicellular plants. J. Cellular and Comp, Physiol. 5:55-74. 36. Ketchum, B. H. and A. C. Redfield. 1949. Some physical and chemical characteristics of algae growth in mass culture. J. Cellular and Comp. Physiol. 33:281-299. 37. Kok, B. 1953* Experiments on photosynthesis by Chlorella in flashing light. In Algal Culture from Laboratory to Pilot Plant. J. S. Burlew (ed.). Publ. bOO, Carnegie Inst, of Wash. pp. 63-75. -128- 38. Leveille, G. A., H. E. Sauberlich and J. W. Shockley. I962. Protein value and the amino acid deficiencies of varioas algae for growth of rats and chicks. Jour, of Nutrition, 76:423-428. 39. McKee, J. E. and H. W. Wolf. I963. Water Quality Criteria. Publication 3A, the Resources Agency of Calif., State Water Quality Control Board. 40. Milner, H. W. 1948, The fatty acids of Chlorella. J. of Biol. Chem. 176:8l3-8l7. 41. Monod, J, 1949* The growth of bacteria culture. Ann. Rev; Microbiol. 3:371. 42. North American Aviation, Inc. I967. A study of the use of bioraass systems in water renovation. Final Report. 43. Osterlind, S. 1950. Inorganic carbon sources of green algae. II - Carbonic anhydrase in Scenedesaius quadricauda and Chlorella pyrenoidosa. Physiologic Plantarum. 3:430-434, 44. Oswald, W. J. 1963* Light conversion efficiency of algae grown in sewage. Trans. Amer. Soc. Civil Engrs. Vol. 128 part III: 47-83. 45. Oswald, W. J., D. G. Crosby and C. G. Golueke. 1964. Removal of pesticides and algal growth potential from San Joaquin Valley drainage water (A feasibility study). Unpublished report submitted to the Calif. Dept. of Water Resources. 46. Oswald, W. J. and C. G. Golueke. I96O. Biological trans- formation of solar energy in Advances in Applied Microbiology. Academic Press Inc., New York"! Vol. II: 223-2bl. 47. Oswald, W. J. and C. G. Goiueke. I968. Harvesting and processing of waste-grown microalgae. In Algae, Man and the Environment. D. F. Jackson (ed«). Proceedings of a Symposim held June I8-3O 1967. pp. 371-389. 48. Oswald, W. J. and H. B. Gotaas. 1955. Photosynthesis in sewage treatment. Proceedings - Separate No. 636. Amer. Soc. Civil Engrs. 49. Retovsky in Malek, I. and Z.Fencl. I966. Theoretical and Methodological Basis of Continuous Culture of Microorganisms. Academic Press, New York. 50. Sawyer, C. N. 1952. Some new aspects of phosphates in relation to lake fertilization. Sewage and Industrial Wastes. Vol. 24, No. 6:768. -129- 51. Schuler, J. P., V. M. Diller, and H. J. Kerslen. 1953. Preferential assimilation of ammonium ion by Chlorella vulgaris. Plant Physiol., 28:299-303. 52. Shelef, G., W. J. Oswald, and C. G. Golueke. Light inten- sity and nitrogen concentration as growth limiting factors. U. of Calif. Sanit. Engr. Res. Lab. Report No. 6o-4. 53. Smith, R. 1967. A compilation of cost information for conventional and advanced wastewater treatment plants and processes, USDI, FWPCA. Cincinnati, Ohio. 54. Spoehr, H, A. and H. W. Milner. 1949. The chemical com- position of Chlorella; Effect of environmental conditions. Plant Physiol., 24:120-149. 55. Stumm, W. and J. J. Morgan. 1962. Chemical aspects of coagulation. J, Amer. Water Work Assoc, 54:971. 56. Tenney, M. W. , W, P. Echelberger, Jr., R. G. Schuessler, and J, L. Paroni, 1969* Algal flocculation with synthetic organic polyelectrolytes. J. of Applied Microbiol. Vol. 18, No. 6:965-971. 57. Williford, J. W. and D. R. Cardon. 1970. Potential changes in the nitrogen concentration of drainage water during transport. Report No. 4. Agricultural Wastewater Study Group . 58. Witt, V. M. and J. A. Borchardt. I96O. The removal of nitrogen and phosphorus from sewage effluents through the use of algal culture (Scenedesmus and Chlorella in mixed or unialgal cultures). J. Biochem. Microbiol. Technol. Engr. 2:187-203. 59. Zabat, M., W. Oswald, C. Golueke, and H. Gee. 1970. Kinetics of algal systems in waste treatment, phosphorus as a growth limiting factor. SERL Publication, University of California, Berkeley. -130- PUBLICATIONS SAN JOAQUIN PROJECT. FIREBAUGH, CALIFORNIA 1968 "Is Treatment of Agricultural Waste Water Possible?" Louis A. Beck and Percy P. St. Amant, Jr. Presented at Fourth International Water Quality Symposium, San Francisco, California, August l4, 1968; published in the proceedings of the meeting. "Biological Denitrification of Wastewaters by Addition of Organic Materials" Perry L. McCarty, Louis A. Beck, and Percy P. St. Amant, Jr. Presented at the 24th Annual Purdue Industrial Waste Conference, Purdue University, Lafayette, Indiana. May 6, I969. "Comparison of Nitrate Removal Methods" Louis A. Beck, Percy P. St. Amant, Jr., and Thomas A. Tamblyn. Presented at Water Pollution Control Federa- tion Meeting, Dallas, Texas. October 9, I969. "Effect of Surface/Volume Relationship, CO2 Addition, Aera- tion, and Mixing on Nitrate Utilization by Scenedesmus Cultures in Subsurface Agricultural Waste Waters'' Randall L. Brown and James F. Arthur. Proceedings of the Eutrophication-Biostimulation Assessment Workshop, Berkeley, California. June 19-21, I969. "Nitrate Removal Studies at the Interagency Agricultural Waste Water Treatment Center, Firebaugh, California" Percy P. St. Amant, Jr., and Louis A. Beck. Presented at 1969 Conference, California Water Pollution Control Association, Anaheim, California, and published in the proceedings of the meeting. May 9, 1969. "Research on Methods of Removing Excess Plant Nutrients from Water" Percy P. St. Amant, Jr., and Louis A. Beck. Presented at 158th National Meeting and Chemical Exposition, American Chemical Society, New York, New York. September 8,-1969. "The Anaerobic Filter for the Denitrification of Agricultural Subsurface Drainage" T. A. Tamblyn and B. R. Sword. Presented at the 24th Purdue Industrial Waste Conference, Lafayette, Indiana. May 5-8, 1969. -131- PUBLICATIONS (Continued) SAN JOAQUIN PROJECT, FIREBAUGH, CALIFORNIA 1969 "Nutrients in Agricultural Tile Drainage" W, H. Pierce, L. A. Beck and L. R. Glandon, Presented at the 1969 Winter Meeting of the American Society of Agricultural Engineers, Chicago, Illinois. December 9-12, I969. "Treatment of High Nitrate Waters" Percy P. St. Amant, Jr., and Perry L. McCarty. Presented at Annual Conference, American Water Works Association, San Diego, California. May 21, I969. American Water Works Association .Journal. Vol. 61. No. 12. December 1959. pp. b59-bfe2. The following papers were presented at the National Fall Meeting of the American Geophysical Union, Hydrology Section, San Francisco, California. December I5-I8, I969. They are published in Collected Papers Regarding Nitrates in Agricultural Waste Water. USDI, FWQA, #13630 ELY December 19^9 • "The Effects of Nitrogen Removal on the Algal Growth Potential of San Joaquin Valley Agricultural Tlie Drainage Effluents" Randall L. Brown, Richard C. Bain, Jr. and Milton G. Tunzi. "Harvesting of Algae Grown in Agricultural Wastewaters" Bruce A. Butterfleld and James R. Jones. "Monitoring Nutrients and Pesticides in Subsurface Agricultural Drainage " Lawrence R. Glandon, Jr., and Louis A. Beck. "Combined Nutrient Removal and Transport System for Tile Drainage from the San Joaquin Valley" Joel C. Goldman, James F, Arthur, William J. Oswald, and Louis A. Beck. "Desalination of Irrigation Return Waters" Bryan R. Sword. "Bacterial Denitrification of Agricultural Tile Drainage" Thomas A. Tamblyn, Perry L. McCarty and Percy P. St. Amant. "Algal Nutrient Responses in Agricultural Wastewater" James F. Arthur, Randall L. Brown, Bruce A. Butterfield, Joel Co Goldman. -132- Jj Accession Number Siibjerl Field &. Group 05 D SELECTED WATER RESOURCES ABSTRACTS INPUT TRANSACTION FORM ^ I Department of Water Resources San Joaquin District Fresno, California REMOVAL OF NITRATE BY AN ALGAL SYSTEM y Q I Attthoi(s) Brown, Randall L. IX, Project Designation 13030 ELY 91 Note Available from Department of Water Resources Post Office Box 2385 Fresno, California 93723 22 23 Citation Agricultural Wastewater Studies Report No. I303O ELY 4/?! - T Pages 132. Figures 58, Tables 27, References 59 Descriptors (Starred First) ^Agricultural Wastes, *Water Pollution Control, Biological Treat- ment, Nitrates, Treatment Facilities oc Identifiers (Starred First) *Algae Stripping, Scenedesmus, Algal Growth and Harvesting 07 Abstract An algal Bystea eonslsting of algae growth, harvesting and disposal was evaluated as a possible near of removing nitrate -nitrogen fro» subsurface agricultural drainage In the San Joaquin Valley of California the study of this asslallatory nitrogen removal process was Initiated to determine optiaua conditions for growth of the algal bloaass, seasonal variations in assiailation rates, and aethods of harvesting and disposal of the algal product. A secondary objective of the study was to obtain preliminary cost esti- mates and process design. The growth studies showed that about 75 to 90 percent of the 20 mg/1 influent nitrogen was assimilated by shallow (l2-inch culture depth) algal cultures receiving 2 to 3 mg/1 additional iron and phosphorus and a mixture of 5 percent COg. Theoretical hydraulic detention times required for these assimilation rates varied from 5 to li days, depending on the time of the year. Ibe total nitrogen removal by the algal system, assuming 95 percent removal of the algal colls, ranged from 70 to 85 percent of the Influent nitrogen . the most economical and effective algal harvesting system tested was flocoulation and sedimentation followed by filtration of the sediment. The algal oalce from the vacuum filter, containing about 20 pereei solids, was then air- or flash-dried to about 90 percent solids. The market value for this product as a protein supplement was estimated to be about $80 to $100 per ton. Abstractor gpQWn 'Department of Water Resources SEND TO: WATER RESOURCES SCIENTIF U S DEPARTMENT OF THE IN WASHINGTON. D C 20240 ITION C ENTER 989-359-339 U Vt rt a- B HH C M* c 2- •< V •< < n o o ««• c O g ti CL D. n o *' P ^1 3 2 3 n ?• ?; e- n K o rt ►-• S K g ?3 n> 3 5' tn O rt O o D r. o ►5 n o ti rt ••■ c. p K p a c •0 re' 2v H n> o o re re i. (i> 3 3 J- w tn » s- o tJ »-• 3 O s- n> o re 0 tr p 5 E. o- '^ o o p < rt a o R r rt •o cr g" S n z 30 3D TJ < » o 5. 55 2. S ? o (O (D O Z — • m o o -* ^ X 3- _i i D I- _k CO o o> :: X fi> If 3 •< lO O CT o c H <-» m o' O 3 H CO O (D O * ^ ':*. > o •sj to o en 3 m o -< z < 3D o z m z -o o (/) H > 30 -n O 2 m O (/> 5 > i® 3764 > O m z o •< THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW BOOKS REQUESTED BY ANOTHER BORROWER ARE SUBJECT TO RECALL AFTER ONE WEEK. RENEWED BOOKS ARE SUBJECT TO IMMEDIATE RECALL MUU'M ^%h J^'^ECO )AN 5 1383 NOV 2 , LIBRARY, UNIVERSITY OF CALIFORNIA, DAVIS Book Slip-Series 458 3 1175 00499 7766 TC California. 82I4 Bulletin, C2 A2 ^ /7V 9'/J PHYSICAL SCIENCES LIBRARY Dept. of Water Reooxirces.