NITROGEN REMOVAL FROM SECONDARY EFFLUENT
APPLIED TO A SOIL-TURF FILTER
Elizabeth Leigh Anderson
A Thesis Submitted to the Faculty of the
DEPARTMENT OF SOILS, WATER, AND ENGINEERING
In Partial Fulfillment of the Requirements
For the Degree of
MASTER OF SCIENCE
WITH A MAJOR IN SOIL AND WATER SCIENCE
In the Graduate College
THE UNIVERSITY OF ARIZONA
1 9 7 8
STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of re quirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made.
Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judg ment the proposed use of the material is in the interests of scholar ship. In all other instances, however, permission must be obtained from the author.
APPROVAL BY THESIS DIRECTOR
This thesis has been approved on the date shown below:
Assistant Research Scientist
I wish to express my gratitude to Dr. Ian L. Pepper for his willingness to work, counsel, guide and review during the course of this experiment. My thanks to Dr. Gordon V. Johnson for his original idea and his help in getting under way.
The author also wishes to extend thanks to Dr. Wallace H.
Fuller, Dr. William R. Kneebone, Dr. Donald F. Post, Dr. Jack L.
Stroehlein, Dr. Thomas C. Tucker, and Dr. Arthur W. Warrick for advice suggestions, and comments. The staff of the Soils, Water, and Engineer ing Department will be remembered for their help and encouragement.
This work could not have been completed without the help of many others. Thanks to Lyzi Cattany, Bob Drake, Monte Edlund, Evan
Karol, Don Leftwich, Charles McKown, Dave Morris, Dr. David B. Marx
Carl Michaud, Jim Mullins, Lewis Munk, and Michael Turner.
This work is dedicated to Dr. Kenneth K. Barnes and his wife
TABLE OF CONTENTS
LIST OF T A B L E S ................................................
LIST OF I L L U S T R A T I O N S ........................................ vi
Mechanisms of Nitrogen Removal . . . . . . ...............
Nitrogen in the Soil E n v i r o n m e n t .........................
Tertiary Treatment of Wastewater .........................
Land Disposal of Wastewater .......................
Effluent Use for Irrigation and C r o p s ...............
Renovation of Effluent by Low Rate Land Application .
High Rate Infiltration-Percolation ...................
Spray-runoff Land T r e a t m e n t ........................
M a t e r i a l s ............
M e t h o d s .............................
Effluent Application ............. . . . . . . . . . .
Sample Collection . . . . . . . . . . . . . . . . . .
Chemical Analysis ..................... . . . . . . .
Effluent Application Rates ........ . . . . . . . . .
Interpretation of Data . ...................
Water R e c h a r g e ...........................................
Utilization . . . . . . . . . . . . . ................. •
Nitrogen Removal ................. . . . . . . ........ •
Nitrogen Transformations in Leachate .................
The Soil's Role ....................... . . . . . . . . .
LIST OF REFERENCES 72
LIST OF TABLES
1. Effluent application rates used 3 June 1977 to 28 April
1978 . . ................................................
2. Analysis of variance for percent water recharge 11 July
1977 to 28 April 1978 .................................
3# Percent water recharge means . . . . . . . . . . . . . .
4. Total recharge volumes, total evapotranspiration volumes, and total grass yields for 30 April 1977 to 28 April
5. Analysis of variance for N utilization, 11 July 1977 to
28 April 1978
7. Total yields and percent N (organic -N and NH^-N) in bermudagrass clippings collected for 16 weeks, 11 July to 29 October 1977 ........................................
8. Total yields and percent N (organic -N and NH^-N) in ryegrass clippings collected for 22 weeks, 28 November
1977 to 28 April 1978
9. Analysis of variance for percent N removed, 11 July
1977 to 28 April 1978 ................................. 57
10. Percent N removed means and average ppm -N in leachate •
11. Total N removed by clippings and by soil processes from
30 April 1977 to 28 April 1978 .........................
LIST OF ILLUSTRATIONS
1. Effluent N composition during the year long study . . . . 26
2. Percentage water recharge for all effluent application rates on s a n d ................. .................... ..
4. Maximum and minimum air temperatures and rainfall measured at Rincon Vista Turfgrass Research Center during the year long study . ........... .......... ..
3. Percentage water recharge for all effluent application rates on m i x .......................................... . 35
5. Percentage water recharge, showing all five effluent
6. Percentage N utilization for all effluent application rates on s a n d ............................................ 45
7. Percentage N utilization for all effluent application rates on mix ........ .. ........ . . . . . . 46
8. Percentage N utilization showing all five effluent application rates averaged for each soil . . . . . . . . 52
9. Percentage N removed for all effluent application rates on sand . . . . . . ................... . . . . . . . . . 54
10. Percentage N removed for all effluent application rates on m i x ......................................
11. Percentage N removed showing all five rates of effluent application averaged for each soil ................. .. 62
This study investigated the use of a soil-turf filter for renovation of secondary effluent# Lysimeter plots were filled with a sand, and a sand mixed with organic matter and soil, then seeded to annual ryegrass. In spring, the ryegrass was removed by close mowing and seeded to bermudagrass. Plots were drip irrigated twice a week with Tucson secondary effluent at rates of 10, 17, 22, 34 and 43 mm/day. Analysis of leachate, effluent and grass clippings allowed a total N balance to be computed. Percentage of leachate available for groundwater recharge was
# at the lowest rate and 72# at the highest rate when values were averaged over 42 weeks for both soils. Percent age of applied nitrogen removed by the sand-turf filter was 40 to 79# and 48 to
# on the mix-turf filter, and decreased as application rate increased. Percentage of applied nitrogen utilized by turf was
12 to 26# on sand and 18 to
# on mix, and decreased as rate of application increased. Results of the 42 week study showed that the
N concentration in municipal effluent could be reduced from 20 ppm -N to 10 ppm -N. The sand-turf filter could renovate 17 mm/day and the mix-turf filter 34 mm/day, and still yield leachate averaging less than 10 ppm NO^-N. Results also indicated ryegrass utilized more N than bermudagrass.
Tucson, Arizona is a city of 450,000 people totally dependent on groundwater for meeting their water needs. Fifty percent of the water pumped into the distribution systems appears as return flow to sewage treatment plants (Cluff, DeCook, and Matlock 1972). Groundwater levels have decreased 2 to 5 feet per year as use exceeds natural re plenishment (Groundwater Resource Management data 1977i from Department of Soils, Water, and Engineering, University of Arizona). Over 35 million gallons of secondary effluent per day are released into the
Santa Cruz river without consideration to nutrients in the effluent, groundwater pollution, or water conservation. Public Law 92-500 restricts discharge of wastewater into streams, lakes, and surface water (Bouwer 1976). Such restrictions have stimulated research on tertiary treatment methods which lower the concentration of pollutants found in wastewater.
Sidle and Johnson (1972) found turfgrass removed 90# of the
N applied in effluent. Their plots were irrigated with 2.84 inches of effluent whenever available soil moisture reached 40#. After N was removed by the soil-turf filter, 4096 of the applied effluent passed below the root zone and was available for groundwater recharge (Johnson
1973)• This study indicated the potential of a soil-turf filter for renovation of municipal sewage effluent. Wilson and Lehman (
) in vestigated the potential of a grass filter for removing settleable
materials and reducing organic pollution in sewage effluent. Results
2 showed overland flow using bermudagrass for filtration provided only partial treatment of the effluent. The grass height and density were important in order to ensure filtration. Bermudagrass was able to tolerate prolonged flooding with sewage effluent. Further studies by
Lehman (1968) using bermudagrass and soil columns indicated that effluent application required proper management for maximum removal of
N and heavy metals. Johnson (1973) cited several reasons for investi gating the use of turfgrass as a disposal system for nutrients in effluent. Turfgrass requires a relatively high constant level of N throughout the year and can be grown year round in the Southwest. Golf coursess greenbelts, playgrounds and parks could be watered with effluent; saving groundwater and promoting city-wide recreation. Be cause treatment plants are generally on the periphery of cities where acres of turfgrass could be developed for recreation, transmission of sewage effluent would not require extensive pipe lines.
This study was conducted in order to determine the maximum rate of effluent that could be applied to a soil-turf filter and yield recharge water meeting United States Public Health Service (USPHS) standards. The goal was to determine the feasibility of using rec reational areas for effluent disposal, conserving groundwater, and contributing quality recharge water. The pollutant considered in this experiment was nigrogen with NO^ being the major ion of concern.
Nitrogen is involved in the eutrophication of streams; nitrate (N0~) can cause methemoglobinemia in infants and is poisonous to animals; nitrite (N0~) is poisonous to humans (McKee and Wolf 1965). There
are no USPHS limits on ammonium (HH^) or nitrite in drinking water.
Some countries limit NH^ to 0.5 ppm and NO” limits vary from 0.1 to
2 ppm (McKee and Wolf 1963)• The NO” limit is 45 ppm or 10 ppm N0”-N.
Land application of wastes is a practice that has existed for hundreds of years. Yet only in the past few decades has man sought to evaluate the effects of such a disposal system on his environment.
Studies of wastewater disposal are currently concerned with renovation of effluent enabling direct discharge into streams, rivers, lakes, or oceans without degrading the quality of natural waters; or use of wastewaters for agricultural and recreational purposes in order to de crease demand on freshwater reserves.
Evans (1970) regarded waste as a valuable resource in the wrong place, form, amount or time. He stressed the need to convert waste to a reusable product and the importance of neutralizing or converting toxic substances. In
there were 2,400 land waste disposal systems • being used for liquid wastes. Nine hundred of these systems were in the food processing industry (Evans 1970). The traditional method of waste disposal into surface waters is no longer feasible as dilution water is less available. Stream self-purification does not work on many pollutants (Stephan and Weinberger 1968). The two major problems are water pollution and water supply. Renovation and reuse of water increases freshwater supply. P ratt, Thorne and Wiersma (1977) cited five factors that affect utilization of wastes: economics, public policies, research and development, quality control, and public accept ance. Crops, forest, cities and people can all benefit if waste is properly disposed.
Tofflemine and Farnan (1975) reviewed several studies showing
5 that land disposal of wastewater was more economical for small treat ment plants and plants needing tertiary treatment. They cited examples of high rate infiltration beds being used in Arizona, Massachusetts, and New York. Young and Carlson (1975) concluded that land application of wastewater was the lowest cost alternative for improving water quality in the southern United States. Land treatment has the advan tage of being less sensitive to price flux when compared to tertiary treatments involving large quantities of chemicals and energy.
Mechanisms of Nitrogen Removal
Nitrogen removal by soil mechanisms has been summarized by
Lance (1972): biological denitrification, volatilization of NH^, adsorption of NH^, fixation of NH^ by clay and organic matter, N assimilation by microorganisms and vegetation, chemo-dentrification, and leaching. Generally the primary form of N in effluent is NH^.
Cation exchange capacity (CEC), organic matter content, and the clay fraction determine a soil's capacity for NH^ retention. Calcium and magnesium ions will compete with NH^ for cation exchange sites.
Ammonium is held by the lignin fraction of organic matter. The amount of NH^ retained by organic matter or clay increases with pH due to pH dependent charge. Montmorillonite and vermicilite are both capable of fixing NH^ in their clay lattice. Volatilization of NH^ can occur in an alkaline aqueous medium:
NHj + H20
OH” -» N H ^ t + 2H20
Volatilization increases with temperature and when rapid evaporation of
water occurs. Nitrogen can be assimilated by plants and microorganisms as either NH^ or N0~.
Denitrification is a process involving microorganisms which contributes to nitrogen removal from wastewater, but nitrification is a prerequisite. Nitrosomonas sp. are obligate autotrophic bacteria re sponsible for the oxidation of ammonium to nitrite. This reaction is the first step in nitrification and requires oxygen:
2NHJ + 30
2ND” + 2H20 + 4H+
Numerous heterotrophic organisms can convert reduced N compounds to nitrite, including bacteria, actinomycetes, and fungi. Nitrite can result from the oxidation of amines, amides, hydroxylamines and oximes.
The conversion of nitrite to nitrate is the second and final step in nitrification and is performed mainly by Nitrobacter sp. which are obligate autotrophic bacteria. This reaction is represented as:
2N°2 + 02 -» 2N0"
A few other heterotrophs, mostly fungi, can produce nitrates. The availability of oxygen in the soil, temperature, soil moisture, and soil pH will influence this microbial activity. The pH range for nitrification is 5«5 to 10, with an optimum around ?• Temperature range for nitrification is 5 to 4o°C with an optimum around 30 to 35°C.
Dentrification occurs in an anaerobic environment where species of the genera Pseudomonas, Achromobacter and Bacillus are able to use nitrate and nitrite as terminal electron acceptors in the absence of oxygen. Autotrophs capable of reducing nitrates are Thiobacillus dentrificans and Thiobacillus thioparus. A possible pathway for den trification is:
2 HNOj -> 2 HN02 -» H 2N 202 h
- Ng or NgO
At pH 4.9 to 5*6 nitrous oxide is usually formed. At pH 7*3 to 7.9
nitrogen gas is the main form of N released. Waterlogging of soil or a decrease of oxygen in the soil pores induce dentrification. In creasing oxidizable carbonaceous material increases the loss of nitrogen gas. Dentrification can occur in a temperature range of 2 to 60°C with an optimum around 25°C.
Lance (1972) reviewed the process of chemo-dentrification and noted that it was unlikely this process would contribute to any sig nificant loss of N in land disposal of secondary effluent. Effluent generally contains 20 to 25 ppm-N and chemo-dentrification reactions usually occur when ammonia-producing fertilizers are banded or when large quantities of nitrite salts are added to the soil. There are several proposed reactions by which chemo-dentrification could occur.
All of these reactions assume an accumulation of NOg in the soil. In alkaline soils where volatilization could occur, NH^ accumulation is toxic to Nitrobacter and could lead to NOg accumulation. One proposed reaction for chemo-dentrification is:
NH5 + HN02 -*N2 + 2H20
Two other proposed reactions are:
3 HN0_ -»2N0 + HNO, + H o0
2 3 2
R-NH2 + HN02 -+ R-OR + H 20 + N2
Both of these reactions occur at pH 5 or less and have not been shown to be of importance in N loss from soils.
Leaching of N from the soil could occur, depending on the cations present in the water applied and their concentration. Ammonium could be replaced on the cation exchange site by another cation and then travel below the root zone with soil water. Nitrate is an anion and tends to travel with soil water, being easily leached below the root zone. Vegetation and microorganisms incorporate N in their tissue, removing N from the soil solution. Vegetation may also contribute to gaseous loss of N by promotion of denitrification in the rhizosphere
Nitrogen in the Soil Environment
Oxygen is necessary for the process of nitrification, which must occur before effluent N can be denitrified. Klausner and Kardos
(1975) found oxygen diffusion rates were not significantly different when effluent was applied at rates of 0, 2.$, and $.1 cm per week.
Oxygen concentration was not affected by crop cover. Preul and
Schroepfer (1968) point out that the major part of nitrification occurs in the upper two feet of soil. Adsorption will be the dominant mode of
NH^ removal in saturated flow until the capacity of the soil is ex ceeded. Unsaturated flow can increase nitrification of effluent N.
Lance and Whisler (1972) flooded soil columns with effluent to evaluate the optimum flood/dry cycle for N removal. In this process, NH^ is held on the soil's cation exchange sites and nitrified during the dry period.
When the column is flooded, NO” is leached from the soil. This creates a peak of high nitrate water at the beginning of the flood period and water with a very low N content after the NO” is leached. Lance,
Whisler and Bouwer (1973) determined that a five day dry period allowed enough oxygen to enter the soil to oxidize all the NH^ applied when flooding the soil with effluent for six days. The highest oxygen move ment in the soil was obtained on a one day flood/one day dry, effluent application cycle. The oxygen demand can be estimated by NH^ and organic N content of the effluent applied.
The nitrate anion can move freely through soil. Corey, Nielson and Kirkham (1967) found denitrification was increased 2196 by decreas ing the velocity of fluid movement through the soil from 1.32 to 0.11
cm/hour. At the slower velocity there was more time for microbial action on NO^. Woldendrop (1962) showed that living roots stimulated " denitrification by increasing oxygen consumption in the rhizosphere.
Roots also excrete organic substrates that can be used by denitrifing organisms as H+ donors. Several experiments have been conducted using different energy sources and measurements of redox potential to de-
I termine optimum conditions for denitrification.
Lance and Whisler (1976) found that methanol added to wastewater at the rate of 150 ppm increased N removal 12$. When dextrose was added to wastewater at the same rate, N removal increased to 90^.
Use of dextrose in the wastewater decreased infiltration due to soil clogging. This also reduced oxygen uptake in the dry periods, causing incomplete nitrification. The dextrose treatment increased NH^ in the reclaimed water. More research is needed to find out if an organic carbon source added at the beginning of the flood period would be effective in decreasing the nitrate peak that occurs in high rate in filtration systems. Mann et al. (1972) found denitrification by
Thiobacillus denitrificems was significantly increased by adding sulfur to the soil which is the energy source for this organism. Soil texture affected denitrification rates due to the ability of certain clays and silts to provide adsorption sites between bacteria and insoluble sub stances.
Whisler, Lance and Linebarger (1974) measured redox potentials in soil columns periodically flooded with sewage effluent. Denitrifi cation occurred in the range of Eh = 200 to 300 mV. Redox probes were used to indicate where denitrification was occurring in the soil column.
When soluble C was added to effluent, the redox potential dropped to
-200 mV, indicating the oxidation states of other elements besides N were becoming unstable. Meek, Grass and MacKenzie (1969) found that when the Eh was 300 mV or below there was a large loss of N by denitri fication. The soil moisture level was 32# at field capacity. At 34 to
4l# soil moisture the Eh was between 500-600 mV and NO” loss was 3*5 mg/250 gsoil. When soil moisture was increased to 48# (saturation),
Eh decreased to 300 mV and N0~ loss increased to 29.5 mg/250 g soil.
The addition of organic matter at 48# soil moisture did not change the
Eh, but increased the NO” loss to 49 mg/250 g soil. Smith, Gilbert and Miller (1976) measured redox potentials on a cropped potato pro cessing waste water disposal field, which had a fluctuating water table. In the winter months when the water table was low, redox poten tials increased (300 to 600 mV) and denitrification was slow. As the water table rose in the summer, redox potentials decreased to -400 mV and denitrification was rapid. The soil contained enough organic mat ter from the processing water to supply energy for denitrification.
Denitrification would also be slower in the winter due to lower tem peratures. Enfield (1977) designed a system to optimize nitrificationdenitrification in a soil column by using redox potential probes at different soil depths. By controlling redox potentials at 6 cm for nitrification and 30 cm for denitrification N removal was 43^.
Municipal sludge was added to the soil column at the rate of 1.3 g
C/gN denitrified. The addition of sludge into the soil increased N removal from the wastewater by 905&.
Hill (1972) tested six Connecticut soils and classified them as good or poor renovators after a two year study. Acid soils removed
" | and Mg"*"*1
Soils with the lowest permeability removed the largest amounts of nutrients. DeJong (1978) found similar results when sewage effluent was applied to a sandy loam. Chloride and SO^ both appeared before one pore volume had leached through, indicating incomplete mixing be tion was affected by exchange reactions between these cations.
Tertiary Treatment of Wastewater
Tertiary treatment of water is necessary to meet water quality standards for wastewater discharge. Stenburg, Convery and Swanson
(1968) pointed out that wastewater treatment should be tailored to the waste involved and the quality of effluent required. Pilot tertiary treatment plants are a necessity to study the most effective and eco nomical way of operating tertiary treatment systems. The District of
Columbia Water Pollution Control Plant in Washington, D.C. is testing
and evaluating different systems. Two of the most promising systems are ammonia stripping and biological nitrification-denitrification.
Ammonia stripping involves increasing the water’s pH to 11 then cir culating the water with 400 cubic feet of air per gallon to allow volatilization. This process can remove 959» of NH^ at a cost of 816 per million gallons of effluent, but is not effective on nitrified effluent. The nitrification-denitrification process involves micro organisms previously discussed. For nitrification the wastewater is aerated with activated sludge, an organic C source (methanol) added and the wastewater run through a carbon or sand filter where denitri fication can occur. This process can remove
% of N in wastewater.
Stenburg et al. (1968) estimated that tertiary treatment costs S260 to $300 per million gallons. Renovated water could be sold for reuse and thus defray part of the tertiary treatment costs.
The South Tahoe Public Utilities District has been testing several tertiary treatment methods since 1961. Preserving the quality of Lake Tahoe and the surrounding recreational areas was a major con cern. The treatments used in this pilot study were chemical coagula tion, mixed media filtration with polyelectrolyte filter, and a carbon adsorption filter (Culp 1968). One of the processes tested used alum to precipitate N and P in the wastewater. The alum treated water was subsequently filtered through two mixed media beds and passed upwards through an activated carbon column, then chlorinated. The expense of using alum in tertiary treatment led to the development of a process using lime, which can be reclaimed and reused. Lime precipitates P and increases the water's pH to 11 for ammonia stripping. Culp (1968)
13 stated that this process was more dependable than denitrification and that there were no liquid or solid wastes for disposal* Nitrogen and
P removal with lime was 98%. An activated carbon filter was effective in removing
% of the organics in the water. Water from this treat ment plant was clear, colorless, odorless, free of coliform bacteria and virus, and generally indistinguishable from drinking water (Slechta and Culp 1967). In 1963, Culp estimated that tertiary treatment would cost $30 to $90 per million gallons. In 1968, the actual operating cost of the 7.5 million gallons per day plant at Lake Tahoe was $150 per million gallons.
Eliassen and Tchobanoglous (1968) cited ammonia stripping, ion exchange, electrodialysis and electrochemical treatment of effluent as methods for removing N from wastewater. They said that nutrient re moval cost was greatly affected by the disposal method used for pollut ants removed. St. Amant and McCarty (1969) used a six-foot filter of one inch diameter gravel to reduce NO” content in wastewater. Wastewater was passed up through this filter for a one hour detention time.
Reduction of 20 ppm N0”-N was 90% when 60 ppm methanol was added to the wastewater. Water must be chlorinated after this treatment to oxidize any NO^ present in the reclaimed water due to incomplete conversion.
Filtration of the renovated water was necessary to remove biological solids and aeration was necessary to prevent any excess methanol from depleting oxygen in the discharge water. Cost of this tertiary process would be $1.55 per million gallons. This figure does not include aeration, filtration, and chlorination.
Francis and Callahan (1975) discussed three methods of treating high NO” wastewater: oodified-activated sludge units, packed bed reactor, and anaerobic columns. The optimum pH range for denitrifica tion depends on the electron donor. Below pH 7, nitrous oxide and nitric oxide are produced. Above pH 7, nitrogen gas is the predominate form. The optimum temperature range for denitrification was from 15 to
C, depending on the organism involved. Modified-activated sludge units require a long residence time to prevent loss of microbial mass.
A packed bed reactor contains an inert material, which provides surface area for microbial growth, through which effluent is passed. Biomass accumulation necessitates periodic back washing of the filter.
Anaerobic columns have a lower pore space than a packed bed reactor and can handle water containing 1000 ppm N0”-N which is low in sus pended solids. Cations in the wastewater will determine the design to be used.
Land Disposal of Wastewater
Thomas and Harlin (1972) categorized land treatment into three approaches: infiltration-percolation, cropland irrigation, and sprayrunoff. High rate infiltration techniques can handle up to 91 m of water per year. Soil physical, chemical, and biochemical reactions purify the wastewater. Crop irrigation requires more land area, using less than 3 m per year. The crop is primarily responsible for nutrient removal. Spray-runoff can be used on impermeable soils and requires a vegetative cover to improve wastewater quality as the liquid moves over the surface. Industrial plants using this method have found
15 decreases in suspended solids, oxygen demanding substances, and nitro gen. Problems limiting the use of land treatment are: heavy metals, disease, soil properties, salt accumulation, aesthetic objections, and availability of land close to the treatment plant.
Materials applied to the soil can be volatilized, retained by physical or chemical adsorption, leached, biologically degraded, or appear in runoff. Characteristics of the wastewater which affect renovation are: pH, pollutant form and concentration, complementary or accompanying ion concentration, disposal variation, and temperature.
Soil characteristics which influence renovation include: bulk density, particle density, clay mineralogy, CEC, and resident exchangeable ions
Effluent Use for Irrigation and Crons
Eastman (196?) cited the importance of land application in arid and semi-arid lands. In 1962, California, Arizona, Colorado, Nevada,
New Mexico, Texas, and Washington had
# of the sewage treatment plants that utilized land application of sewage effluent. These plants served
of the population in the 17 western states. Most of these plants use wastewater for irrigation of crops not directly consumed by humans and this irrigation practice started before World War II. Irri gation with effluent may prevent pollution of surface waters, yet con tribute to groundwater pollution.
Reuse of wastewater was studied by Los Angeles in 19^9 (Parkhurst
)• By 1962, a water reclamation plant at Whittier Narrows was returning 12,000 acre-feet per year to groundwater and more water
16 reclamation plants were planned. These plants reclaimed wastewater from residential areas avoiding the complications of handling indus trial wastes. Reclaimed wastewater can also be used to prevent salt water intrusion into groundwater supplies (Parkhurst 1965; White 1975)•
Livermore, California, uses reclaimed water for golf course ponds, for fire protection, and for irrigation of golf courses, airport grounds, and a junior college campus (Parness 1968). To avoid discharging wastewater into surface waters, St. Petersburg, Florida was spray irri gating parks and golf courses with secondary effluent (White 1975)•
By irrigating 2,500 acres around the city with effluent, potable water supplies were conserved. In Tucson, the Randolph Park sewage treat ment plant was designed to meet 7596 of golf course and park water re quirements during summer months (Roll 1974). Use of effluent reduces fertilizer costs. Johnson (1973) suggests that cities can be viewed as "effluent rich," when effluent available exceeds what is used on recreational areas, or "turfgrass poor." He suggests that effluent rich cities develop more parks for recreation, effluent disposal, and groundwater recharge to remedy being "turfgrass poor." In Bakersfield,
California, municipal effluent has been used for irrigating agricul tural crops since 1912 (Crites 1975)• All the effluent produced, 15 million gallons per day, is used on
acres. The crops grown had equal or larger yields compared to surrounding farms, except cotton
(Gossypium hirsutumL.) which was 20# lower. This decrease in yields illustrates the need for research into optimizing crop yields using effluent. Day has worked with effluent irrigation of crops since 1957
(Day, Tucker and Vavich 1962a) and finds more information is still
17 needed on: maximum loading rate of effluent, effect of effluent on soil properties, effect on crop yield and quality, and crop varieties adapted to wastewater (Day 1973)• After 14 years of effluent appli cation on a silt loam soil, Day, Stroehlein and Tucker (1972) found increased soluble salts, nitrate in the C horizon, P in the surface soil and modulus of rupture. Infiltration rate decreased. Sorghum
(Sorghum vulgare Pers.), barley (Hordeum vulgare L.), oats (Avena sativa L . ) , and wheat (Triticum aestivum L.) have all been grown with effluent giving equal or better yields than control plots grown with equal amounts of fertilizer (Day, Tucker and Vavich 1962a, 1962b; Day,
Dickson and Tucker 1963, Day and Kirkpatrick 1973i Day, Rahman et al.
1974; Day and Tucker 1977)• The malt quality of barley and the milling and baking properties of wheat were lowered by effluent irrigation
(Day,"Tucker and Vavich 1962a). Wheat hay had a lower relative feeding value, but higher yields when grown with effluent (Day, Rahman et al.
Renovation of Effluent by Low Rate
Kardos and Sopper (1973), working in Pennsylvania found that corn (Zea mays L . ) , irrigated with 2.5 cm of effluent per week and canary grass (Phalaris arandinacea L . ) , irrigated with 5 cm of effluent per week were not effective in keeping NO^-N concentrations in the sub soil below 10 ppm. A grass-legume hay mix and corn rotation could renovate 5 cm per week. Sopper and Kardos (1973) pointed out that vegetation is a renovating agent and yields were increased with effluent spray irrigation. Overman and Ku (1976) grew rye (Secale
18 cereale L . ) , pearl millet (Pennisetum typhoides /Burrn?, Staff and C.
E. Hubb.), and ryegrass (Lolitun multiflorum Lam.) with effluent in
Florida. Pearl millet utilized
%> of N applied, compared to 67& and
50^ for rye and ryegrass. As the amount of N applied increased, per cent utilization decreased. At the highest application rate, plots received
cm of effluent per week. The effluent was deficient in K for these crops. Hortenstine (1976) irrigated Coastal bermudagrass
(Cynodon dactylon L. Pers) and three tree species growing on a spodosol with 5 cm of effluent per week. Because of a high water table, NH^ in the effluent was not nitrified. The spodic layer effectively re moved P. Hortenstine (1976) also found the EC of the soil solution was increased by effluent applications. Parker et al. (1974) reported on lysimeter studies conducted using swine lagoon effluent. The effluent load was 2.5 cm/week. Total N removed was 99&» 20# of which was recovered in grass clippings on the lysimeters. They found that regulation of the size of aerobic and anaerobic soil reaction zones for oxidation of NH^ to N0“ , and passage of nutrients sufficient for energy requirements of denitrifying microorganisms, appear to be the critical criteria for N removal.
In 1967i Pennypacker, Sopper and Kardos applied effluent on a loam soil in a mixed hardwood forest, a red pine plantation, and an old abandoned field in Pennsylvania. Test plots were irrigated at 2.5
cm per week. They found NO^-N decreased
to 82#, organic-N decreased 75 to
#, and P decreased 99#* Renovation of the effluent was achieved by filtration and adsorption through the soil, and
19 biological activity in the root zone. Hook and Kardos (1978) found that a hardwood forest on a sandy loam was an ineffective renovator of effluent applied at 5 cm per week. White spruce forest on a clay loam irrigated at the same rate kept the NO”-N concentration below 10 ppm.
Factors to be considered in using forests for reclamation are: litter decomposition, humus mineralization and microbial population size.
Youngner, Williams and Green (197*0 spray irrigated chaparral in San
Bernardino National Forest at rates of 0, 1.9, 4.4, and 8.9 cm per week. Increasing the irrigation rate increased chaparral growth by releasing the plants from limiting environmental factors. Irrigation with effluent altered species distribution in the area. After three years of treatment no evidence of stream or groundwater degradation could be found. The combined removal capabilities of the soil and grasses accounted for nearly all the nutrients applied in the effluent.
Chadwick et al. (1974) used polluted river water in England to spray heathland at rates of 4.9 to 18.3 cm/day. At both rates, 96% of the NH^ was removed. At the low rate, 26% of the NO” was removed, but at the high rate, NO” increased 14%. The polluted river water (4 ppm
NH^-N and 7 ppm NO^-N) increased yields on the heathland, but did not appear to affect the balance of plant species. This system of disposal improved groundwater recharge while purifying river water. A camp ground in Minnesota has used a peat and sand filter with stalked blue grass to treat effluent (Osborne 1975)• This filter removed 46% of the N applied, 99% of P applied, and 99% of the coliform bacteria.
This tertiary treatment allowed use of effluent water to replenish the lake without causing eutrophication. Parizek (1973) cites several
20 factors that should be considered when selecting a site for irrigation with effluent. Application rate and precipitation must exceed evapotranspiration in order for salts to be leached. Good subsurface drainage is necessary and the hydraulic conductivity of the soil must allow a long enough residence time for renovation. Soil thickness, ion exchange capacity, and topographic setting are also important. On any land application system monitoring subsoil solution is essential.
High Rate Infiltration-Percolation
Crops and forests have been shown to be effective renovators when effluent was applied at rates of 2 to 5 cm per week. Bouwer,
Lance and Riggs (197 4 ) have shown that high rate infiltration methods can renovate 91 meters of effluent per year using 3*7 acres of infil tration basins for each million gallons per day of effluent. At this rate of application, N removal is 30#* Purification of the effluent is dependent on nitrification of ammonium and denitrification of nitrate as described previously in Lance and Whisler* s (1972) work. The flood/ dry cycle of effluent application must be experimentally determined for maximum purification. The optimum cycle will depend on the CEC of the soil, exchangeable NH^ percent, infiltration rate, and NH^ concentra tion in the effluent (Bouwer, Lance and Riggs 1974). Lance and
Whisler (1972) pointed out that the length of the flooding period should not be increased beyond the soil's capacity to adsorb NH^. The high NO^ concentration occurring at the beginning of each flood period can be collected, mixed with new sewage and reapplied, increasing N removal to 8C$ (Lance, Whisler and Rice 1976). Mixing the high nitrate
water with sewage increased the organic C content for denitrification.
Vegetation on the surface of high rate infiltration basins increased denitrification and aided in maintaining a high infiltration rate by filtering out suspended solids. Bouwer (1970) noted that aesthetically, recharge water has the advantage of being collected as groundwater and in the process loses its identity as sewer water. To avoid degrading groundwater quality, renovated wastewater can be pumped out of the aquifer at some point away from the application area (Bouwer 1976) and reused for irrigation, industry or city needs. Most quality improve ment of the wastewater takes place as it percolates through the first meter of soil (Bouwer 1970).
Amramy (1968) cited an example in Israel where high rate in filtration was used. The optimum cycle was found to be 2 to 3 days wet and 7 to 8 days dry. Total N was decreased 64 to 84#. The water was of potable quality except for the NO™ peak which occurred when
wastewater was applied after the dry period. Infiltration rates on these basins decreased from an average of
cm/day to 50 cm/day after
920 days. After 18 months of using these basins the hydraulic conduc tivity of the subsoil decreased; this may have been due to salts in the effluent affecting the clay subsoil. Reclaimed water was mixed with fresh water (2:1 ratio) to reduce NO^ and Cl” content prior to reuse. By excluding industries from the region that might contribute harmful wastes, this reclaimed water was available for reuse, rather than being disposed of in the sea.
The high rate infiltration method of wastewater reclamation can be complicated by accumulation of suspended solids on the soil
22 surface forming a thin layer of high impedance (Bouwer, Rice and
Escarcega 197*0 • Periodic clearing of the basin surface relieved this problem. Thomas (1973) cited soil clogging as the main disadvantage of high rate infiltration systems. Clogging usually occurs under anaerobic conditions on the soil surface and can be remedied by drying the soil surface. Rice (197*0 determined that for a given suspended solids load, impedance of the clogged layer increased with an increas ing hydraulic gradient. Amount of water infiltrated increased with an increased hydraulic gradient. Increasing suspended solids in the wastewater increased hydraulic impedance. Suspended solids moved further into the coarse sand creating a thicker layer of clogged soil.
Three years of infiltration decreased hydraulic conductivity 50 to
Rice (197*0 proposes this decrease in conductivity may be the result of trapped gas, unable to move through the clogged soil layer.
Spray-runoff Land Treatment
Law, Thomas and Meyers (1970) evaluated cannery wastewater treatment in which
cm per week was applied to land with red canary
(Pholaris acundinacea L.), tall fescue (Festuca arundinacea Schreb.) and red top (Agrostis stolonifera L.)grasses growing. Runoff accounted for 6l3» of the water applied, 21% of the wastewater entered the soil and 18% of the water was lost to evaporation. In this treatment 85% of the N applied was removed. The sandy loam soil had a greater treatment efficiency than the clay loam. This spray system had the capacity of
3*6 million gallons per day. Bendixen et al. (1968) compared spray, flood, and ridge-furrow irrigation as methods of wastewater renovation.
On a long term basis, spray irrigation removed 30# of N applied, flood irrigation removed
and ridge-furrow irrigation 14#. The conclusion from these tests was that the ridge-furrow system would have the longest life and was best adapted for use in northern latitudes.
Adriano et al. (1975) reported on long term land disposal of food processing wastes. Nitrate and PO^ exceeded public health stan dards in the groundwater at sites used 10 and 20 years for waste dis posal. Vegetation on sites did improve nutrient removal. Lehman
(1968) concluded that overland flow using Bermuda grass for filtration did not reduce the N content of effluent. In experiments using inter mittent flooding of soil columns seeded to bermudagrass, Lehman con cluded that the cycle of one day flood with 10 cm water and three days dry would be the most effective for water renovation.
This study attempted renovation by secondary effluent applied at high rates to a soil-turf filter. Turfgrass was the crop chosen for this study because it can be grown year round and the grass uti lizes large amounts of N. Maintenance of the turf would be less costly if effluent was used as fertilizer and water. The use of effluent to water recreational areas would leave more groundwater available for household use
MATERIALS AND METHODS
This study was conducted at The University of Arizona* Rincon
Vista Turfgrass Research Center in Tucson. The experiment was con ducted in order to determine the maximum rate of effluent that could be applied to a soil-turf filter and yield recharge water meeting USPHS standards. For this study, effluent, recharge water, and grass clip pings were all analyzed for N to obtain the N balance for the system.
Twenty lysimeters were used in this study. Ten units were filled with 99% sand, 1% silt, and 4# clay, referred to as sand. The sand soil had a CEO of 2.1 meq/100 g soil. The other 10 units con tained
% sand, 9% silt, 4$ clay and 2% organic matter, referred to as mix. The mix soil had a CEC of 4.8 meq/100 g soil. The organic matter added to the mix soil was Loamite, a sulfuric acid treated red wood bark. Both soils had a pH of 8.3 and were texturally sands.
Units were arranged in a split plot design, with t w o .different soils, five application rates, and two replicates of each rate.
The lysimeters used in the study were one meter square by 60 cm deep. These units were constructed with wooden walls and lined with sheets of 6 mil polyethylene plastic. The bottom of the lysimeter was sloped towards the center where a polyethylene tube was sealed into the plastic lining. Tubes leading out of each lysimeter were placed on a slope so water would flow to an underground service area.
Barrels were placed in the service area, below the bottom of
25 the lysimeters, to collect leachate from the plots. A glass tube was inserted into the barrel and calibrated to indicate the amount of water contained in each barrel. Plastic tubes at the bottom of each barrel were inserted to allow sample collection and drainage of the barrels between irrigations.
A stand was built beside the plots to elevate barrels. One barrel was assigned to each plot. The barrels were calibrated to over flow when they had been filled with appropriate amounts of effluent.
Effluent was delivered by gravity flow from the barrel to the plot, using 15 mm drip irrigation tubing, with emitters that could deliver four gallons per hour at 15 pounds of pressure.
The secondary effluent used in this study was obtained twice a week from the Randolph Park Sewage Treatment Plant. Effluent from this plant is stored in a lagoon and used to water the surrounding golf course and park. Lagoon storage of effluent and seasonal changes caused the nitrogen composition of the effluent to vary, due to microbial transformations (Fig. 1). In the summer the effluent con tained 17 to 23 ppm NH^-N, 0.5 ppm N0”-N, and 2 to 5 ppm organic-N.
In winter, values were 3 to 6 ppm NH^-N, 4 to 16 ppm N0~-N, and 1 to
5 ppm organic-N.
Two types of grass were used in this study. A cool season annual ryegrass (Lolium multiforum Lam.) was used from 30 April to 3
J u n e .1977 and from 4 November 1977 to 28 April 1978. During the inter vening summer months plots were seeded to warm season bermudagrass
•--- - N O j - N
II 25 9 23 6 20 3 17
I 15 29
12 26 10 24 7 21 4
18 4 18 I
Figure 1. Effluent N composition during the year long study,
(Cynodon doctylon L. Pers). To insure a good stand of grass, addi tional tap water was applied between effluent applications during reseeding.
Plots were irrigated twice a week. A 300 gallon tank trailer with a pump welded to the side was used to transfer effluent from
Randolph Park to Rincon Vista. Effluent was pumped into each barrel and barrels drained in
hour depending on irrigation levels.
High infiltration rates for both soils eliminated ponding of effluent on the soil surface. Immediately after irrigation of all plots, drip irrigation lines were removed. In the service area, a hand-operated vacuum pump or aspirator was used to evacuate air from lysimeter tubes and initiate the flow of gravitational water. Plots were allowed to drain four hours before
ml leachate samples were collected from each barrel in the service area. Final leachate volumes, indicating the amount of water available for recharge were noted the next day.
Collection barrels in the service area were emptied between irrigation.
Samples of leachate were collected after four hours. Samples were frozen at 4 C after collection from the service area. For each two week period four 250 ml samples were combined to form a composite sample of 1 liter of leachate from each lysimeter. These combined samples were stored at room temperature for 1 to 2 weeks, until
28 analysis. A study of samples stored at room temperature showed no N transformations occurred if the bottles were kept free of algae.
Effluent samples were collected at each irrigation, frozen immediately, and later combined to form a composite sample before analysis. Nitrogen transformations were found to occur if effluent samples were left at room temperature; therefore these samples were always refrigerated.
Grass clippings were collected once a week when plots were mowed to a height of 1.5 inches. Clippings were dried at
C, then weighed to obtain total yields. Samples from both mowings for a two week period were combined and ground for subsequent chemical analysis.
Leachate and effluent samples were analyzed for N content using micro-Kjeldahl techniques (Bremner 1965). A 25 ml aliquot of leachate or effluent was used in steam distillation for NH^-N and NO~-N. Mag nesium oxide was added to the sample first for distillation of NH^-N.
Then Devarda1s Alloy was added to the same sample for distillation of
NO~-N. Distillate was collected in boric-acid indicator solution and titrated with
normal potassium biiodate.
A 50 ml aliquot was used in micro-Kjeldahl digestion to de termine organic-N. A potassium sulfate-catalyst mixture was added to the sample with 3 ml of concentrated sulfuric acid. Samples were heated for two hours to evaporate water and digest organic-N. These samples were subsequently distilled after the addition of
# sodium hydroxide. The same indicator and titrant were used as pre viously stated.
Grass clippings were analyzed for organic-N using 100 mg samples. Catalyst and acid were added as previously described. Grass samples were heated four hours to allow for complete breakdown of organic-N. Samples were distilled after sodium hydroxide was added, then titrated with
normal potassium biiodate.
Effluent Application Rates
Irrigation rates were based on consumptive water use rates determined by Krans and Johnson (197*0. Maximum consumptive water use of creeping bentgrass (Agrostis palustris Huds.) occurred in August and was 5 mm/day. This maximum consumptive use served as a basis for de termining rates of effluent application. In a preliminary study 30
April to 3 June rates of 1 to 4 times consumptive use were used. On a daily basis this was 5 to 22 mm per day. Leachate from this study indicated 95& of the N applied was removed. On 3 June, the annual rye grass was removed and warm season bermudagrass reseeded. Also, at this time rates were amended to include two higher rates of application.
These rates were approximately seven and eight times consumptive use or
34 and 43 mm/day respectively. Table 1 shows the rates used for the
' rates for five weeks. Plots receiving 10, 17, and 22 mm/day were not changed for the entire year.
Interpretation of Data
The nitrogen balance of this system was based on a N input in the form of effluent. Nitrogen output was in the form of grass clip pings or N in leachate. Additional losses of N occurred via
Table 1. Effluent application rates used 3 June 1977 to 28 April 1978.
Rate in Multiples of Consumptive Use
Twice a Week
31 adsorption, denitrification, and volatilization processes in the soil.
Reactions occurring in the soil were determined indirectly as the dif ference between effluent-N input and clipping-leachate-N output. The efficiency of the soil-turf filter in removing N was evaluated by N removed.
N removed of leachate (liters) x total ppm N of effluent (liters) x total ppm N
The ability of the grass to utilize the applied N was determined by percent utilization.
N utilization =
l V U
of «UPPi°8». <g) * organic-H iVol. of effluent (liters) x total ppm N
The ability of the soil turf filter to supply water for reuse or groundwater recharge was evaluated by percent water recharge.
* water recharge . (% £ g
% £ £
These are the three main criteria that were used to evaluate the effec tiveness of the soil-turf filter. All data were statistically analyzed for significant differences at p =
RESULTS AND DISCUSSION
This study was conducted from 50 April 1977 to
Initially the treatments used were: 5i 10, 17, and 22 mm/day; with an additional
mm/day treatment in which a water table was maintained in the lysimeter. After five weeks, the data collected indicated a consistent 933» N removal. On 3 June the 5 mm/day rate was changed to
34 mm/day and the water table treatment was changed to 43 mm/day with out a water table. Statistical analysis of all five rates started with the sample period beginning 11 July. The five week interim allowed the plots to equilibrate with the new effluent application rates.
Percent water recharge indicates the quantity of water that would be available for reuse with this tertiary treatment system.
Statistical analyses of the study from 11 July 1977 to 28 April 1978, showed that recharge, as affected by rates and soils, was significantly different at p =
level and the main effect of time was significant at p = 0.01 (Table 2). Figure 2 illustrates recharge values for all five rates, averaging replicates, on the sand soil for the year long study. Figure 3 illustrates the same data for the mix soil. When recharge for each rate was averaged for both soils over
time periods, only the
mm/day rate was significantly different from the other four rates, as shown in Table 3» Rates were not broken down by
Table 2. Analysis of variance for percent water recharge 11 July 1977 to
R x S
R x T
S x T
R x S x T
•Significant at the 0.05 level of probability.
••Significant at the 0.01 level of probability.
"Effluent Appllcolton Rate (mm /day)
—— ■ |o
* • 17 o • 22 o - 5 4
--------- 4 3
H 2 8 I I
. AUGUST SEPTEMBER
29 12 K> 2 4
NOVEMBER DECEMBER JANUARY FEBRUARY MARCH
Figure 2* Percentage water recharge for all effluent application rates on sand.
Effluent Application Rate (mm /day)
10 o . 1 7 o ■ 22 a ■ 34
----- -- 43
JULY AUGUST SEPTEMBER
NOVEMBER DECEMBER. JANUARY FEBRUARY
1977 I 1978
Figure 3» Percentage water recharge for all effluent application rates on mix.
Table 3» Percent water recharge means.
Average # Water Recharge
Overall Sand Mix
LSD = 11.4
Any two means with a letter in common are not significantly different
(p = 0.05).
37 soil because the analysis of variance showed that the rate by soil interaction was not significant. The 10 mm/day rate is generally the lowest on the recharge graph. Figure 2 does show an increase in re charge at the
mm/day rate during October and January. Rainfall was high during these months and this probably contributed to the increased percentage of recharge. The 10 mm/day rate was only twice peak con sumptive use, so soil moisture had a greater effect at this rate than at the higher rates.
Table 4 illustrates the effect of evapotranspiration on re charge values. As the effluent application rate increased, grass yields increased. By comparing yields and evapotranspiration, it is clear that at the higher rates plants were consuming more water per gram of clippings than at the low rates. These data indicate a luxury consumption of water. In areas where recharge is a primary concern, effluent application could be regulated to avoid excessive water loss due to vegetative transpiration. Table 4 also shows that as applica tion rates increased, volume of recharge water increased, even though the percent water recharge showed no significant difference.
Growth rates would also influence the amount of water available for recharge. Table 4 shows a similar value of evapotranspiration per gram of dry clippings for effluent application rates of
22 mm/day. At these rates, clipping yields increased with increasing effluent application; while evapotranspiration per gram dry matter averaged 2.4 liters/gram. At effluent application rates of 34 and 43 mm/day, the difference between yields was not significant, but evapo transpiration per gram of dry clippings increased to am average of
Table 4. Total recharge volumes, total eyapotranspiration volumes, and total grass yields for
30 April 1977 to 28 April 1978.
Total Yields (g)
liters/gram. These data indicate at the high effluent application
39 rates, the turf reached a plateau in growth response.
A seasonal trend in recharge can be seen in Figures 2 and 3«
Recharge volumes in the summer months were lower than those in the winter months. This trend was due to seasonal variations affecting the soil-turf filter. Figure 4 shows temperature and precipitation data collected at Rincon Vista during the study. In the summer, moisture loss increased due to evaporation from the soil surface, and transpirational water loss by bermudagrass. Relative humidity would play an important role in determining transpirational water loss. Be cause the soil surface is protected by turf, surface evaporation would have a small effect on soil moisture loss. In winter, cooler tempera tures and increased precipitation decreased moisture loss from the lysimeters, increasing the volume of water for recharge.
In June and November all growing turf was removed (scalped) and reseeded to the appropriate warm or cool season grass. During these periods of reseeding, plots were watered several times a day to aid seedling establishment between effluent applications. The increase in percentage of water recharge during these months could be due to this excess tap water maintaining soil moisture. Percentages would also be higher during these periods due to less transpirational water loss because a turf stand was not established. The
nan/day rate on the sand soil showed a greater fluctuation during reseeding than the same rate on the mix. The 10 mm/day rate on the sand also appeared to be more influenced by precipitation. Peaks for
mm/day on sand occurred in October and January, corresponding to periods of rainfall.
II 25 9 23 6 20 3 17 I 15 29 12 26 10 24 7 21 4
JUNE JULY AUGUST SEPT. OCT. NOV.
18 4 18 I 15 29
DEC. , JAN. FEB. MARCH APRIL
Maximum and minimum air temperatures and rainfall measured at Rincon Vista
Turfgrass Research Center during the year long study.
The larger water holding capacity of the mix soil probably buffered these changes, so less fluctuation was seen, even at
At the 10 mm/day rate recharge from the sand soil averaged 18# more than recharge from the mix. At 43 mm/day rate the average re charge from the sand was 1.5# more than from the mix. As rate of effluent application increased, there was less difference in percent recharge values between rates and less difference in recharge between soils (Table 3). Figure 5 illustrates the recharge trend for both soils, averaging all five rates for each soil. The sand soil, with a lower water holding capacity, averaged
# higher recharge than the mix.
Differences between soils over time were consistent, with no signifi cant interaction (Table 2). This consistent difference between re charge values for the two soils, shown in Figure 5, indicated a physi cal property, rather than evaporation or transpiration which would vary with season. The difference in water holding capacity between these two sand soils would be due to the presence of 4# more silt and
# organic matter in the mix soil, which increased the water holding capacity of the mix.
The effect of evaporation and transpiration on recharge is shown by the seasonal trend in Figure 5* Lower evaporation and trans piration in the winter months contributed to larger recharge volumes.
Lower yields on the sand (Table 4) were associated with lower transpirational loss and consequently higher recharge rates (Fig. 5). The evapotranspiration per gram of dried clippings at the two highest application rates indicated that the turf on the sand soil used water less efficiently than the turf on the mix. This increased loss could
Figure 5» Percentage water recharge, showing all five effluent application rates averaged for each soil.
43 be due to increased transpiration or evaporation on the sand-soil plots. Increased yields on the mix soil indicated more leaf surface area to transpire and actual ground evaporation was probably minimal.
The mix soil had higher grass yields and higher transpiration losses
(Table 4) decreasing the total volume of recharge water available for reuse. Seasons, soil water holding capacity, and turf growth all affect the amount of recharge water available from this soil-turf system.
Percentage N utilization indicates the amount of N utilized by turf for growth. There were significant (p = 0.01) rate, soil, and time effects on N utilization and these interacted with each other
(Table 5). Figure 6 illustrates utilization values for all five rates, averaging replicates, on the sand soil for the year long study. Figure
7 shows the same data for the mix soil. Dashed lines over time periods in June and November indicate periods of scalping and seedling estab lishment. During these times no clippings were collected. Clippings collected during scalping were not included on the graph or in sta tistical .analysis (except Table 4, and Table 11, p. $8). Scalping removed most of the organic matter from the soil surface and created utilization values ranging from 39 to 117% on the sand and 55 to 154% on the mix. Figures 6 and 7 demonstrate that the lower application rates had a higher percent utilization values than the higher applica tion rates. As application rates increased, amount of N applied
Analysis of variance for N utilization, 11 July 1977 to
28 April 1978.
Bate 4 3597.72
R x S
R x T
S x T
R x S x T
•Significant at the 0.10 level of probability.
••Significant at the 0.01 level of probability.
E flh w it AppKeotion Rote (mm/doy)
-- - 10 e ■ 5 4
JULY AUGUST SEPTEMBER
NOVEMBER DECEMBER JANUARY FEBRUARY
1977 | 1978
Figure 6. Percentage N utilization for all effluent application rates on sand.
47 increased, but N utilization was limited by N uptake and plant growth.
At the 10 ram/day rate, utilization accounted for 3296 of N applied when averaged over soils and time (Table 6). At 43 mm/day only 1596 of the N applied was utilized. Table 6 shows the means for utiliza tion at each rate are significantly different at p = 0.05. Rates were not ranked on each soil because the analysis of variance showed rate by soil interaction was not highly significant (Table 5)»
A seasonal trend can be seen in Figures 6 and 7, and Table 5 shows significant rate by time interaction. A good turf stand and favorable climate for the grass species used was necessary to maximize
N utilization. August through September appear to be most favorable for bermudagrass growth. Decreasing utilization in October marked a change in weather as shown by Figure 4. Plots were reseeded to rye grass for the cooler months. February through April were most favor able for ryegrass growth. From these graphs there is evidence that ryegrass utilized more nitrogen during its growing season than bermuda grass, regardless of rate.
Table 7 shows yield and percent N data for bermudagrass and
Table 8 shows the same data for ryegrass. These tables show that as effluent application rate was increased, yields and percentage of N in clippings both increased. The mix soil had higher yields on rye grass at all rates, and on bermudagrass at the 34 and 43 mm/day rates.
Percentage of N in clippings was higher on the mix soil at the 43 mm/day rate for both grasses. The higher yields on the mix soil were possibly due to ttye higher CEO of the soil and a higher water holding capacity. Since the soils responded differently, for all three
Table 6. Nitrogen utilization means.
Overall Sand Mix
LSD = 2 . 8
Any two means with a letter in common are not significantly different
Table ?• Total yields and percent N (organic -N and
in bermudagrass clippings collected for 16 weeks, 11 July to 29 October 1977-
Total Clippings (g)
LSD for #N = .206 - -
Any two means with a" letter in common are not significantly different
(p = 0.05).
Table 8, Total yields and percent N (organic -N and NH^-N) in rye grass clippings collected for 22 weeks, 28 November 1977 to
Total Clippings (g)
Sand Mix .
LSD for 95 N = .260.
Any two means with a letter in common are not significantly different
50 parameters tested with statistics, this illustrates the influence of
CEO and organic matter on soil-plant relations. These tables show that ryegrass did utilize more nitrogen as shown by the increased percent N in clippings from both soils. It is unclear, from these tables, if a growth plateau has been reached at effluent rates of 54 and 45 mm/day.
The decreased yields are not significantly different. These figures may be an indication that higher effluent application rates would have a detrimental effect on turf growth.
The turf plots showed yellowing in the last month of bermudagrass growth (October 1977) and during the first months of ryegrass growth (December 1977 through February 1978). During these months, the turf growing was generally a yellowish green. It is possible this chlorosis was due to Fe deficiency. The soil's pH of 8.5 could have caused micronutrient deficiencies. Chlorosis on the high effluent application rate plots was more severe and present throughout the turf.
On the low rate plots, chlorosis was spotty. A yellow strip effect occurred in the first month of the experiment (May 1977)• At this time the effluent being applied was predominantly NH^" -N. Using the drip irrigation lines, N H ^ was held on the cation exchange sites nearest the irrigation line. Yellow areas developed in between irri gation lines due to lack of N. By rotating the drip irrigation lines, an even distribution of N was achieved and the yellow strips eliminated.
Periods of decrease in utilization and subsequent recovery; as seen in September, December, and January, could have been due to removal of too much leaf area by mowing. Plots were mowed to 1.5
51 inches height every week, regardless of how much growth had occurred.
In weeks of rapid growth, this amounted to a proportionately larger decrease in leaf area. Although the remaining leaf area was nearly the same, whatever the amount of growth removed, the decrease in leaf area while the root system remained the same size would slow turf growth. In this situation, photosynthate would probably be used to maintain root respiration, decreasing photosynthate available for leaf growth. Nitrogen uptake would be decreased as the turf would have less leaf area for photosynthesis, therefore less energy for growth and assimilation of N. This decreased utilization illustrates a disad vantage of mowing by time intervals, rather than by turf growth.
Figure 8 summarizes N utilization with all five application rates averaged for each soil. Comparing the utilization means for all time and rates, ttirf on the mix soil utilized 8# more N on the average than turf on the sand soil. Seasonal trends, decrease in utilization, and subsequent recovery can also be seen. Figure 8 illustrates the soils responded differently over the year long study and Table 5 shows significant soil by time interaction. The difference in soil response is illustrated by the fluctuating distance between the two soil lines.
Turf on the mix soil recovered faster from periods of stress than did turf on sand. Higher N utilization during the period of ryegrass growth is evident from this graph. High N utilization on the mix soil demonstrates the fact that grass yields and $ N in clippings from the mix soil were higher. Higher yields on mix could be explained by the higher CEC of the soil and higher water holding capacity. The higher
Figure 8. Percentage N utilization showing all five effluent application rates averaged for each soil.
CEC would allow for more N H^ adsorption, making more K available for plant growth.
The part of this study concerned with N utilization brought up several questions that require further research. The effect of cation/anion concentrations in the effluent on plant uptake of nutri ents and toxicity of elements in the effluent accumulated by the soil are important questions that must be answered in order to maximize turf removal of nutrients from effluent. Management practices; such as mowing, reseeding, and method of effluent application, also need to be studied. The difference in N utilization between bermudagrass and rye grass indicates that further research into turfgrass cultivars might lead to selection of turf grasses more effective in renovation.
Recent N removal reflects the quality of recharge water col lected. The USPHS standard for nitrogen in potable water is 10 ppm
N O ^ - N (McKee and Wolf 1963) • This was the criterion used to deter mine water quality. The average N content of the effluent for this study was 20 ppm -N so a minimum of $0# N removal was necessary to meet USPHS standards. The organic -N content of the leachate was less than 1.0 ppm N at all rates. The ammonium concentration in the leach ate was less than 0.5 ppm N for all rates. The nitrate concentration varied with time of year, as shown in Figures 9 and 10, ranging from less than 1 ppm N0”-N to 20 ppm N0~-N. Leachate samples were spot checked for N0~, but none was found. Statistical analyses for this
Effluent Application Rote (mm/doy)
-- - 10 a o
" 2 2
o ■ 34
■ 4 3
AUBU8T SEPTEMBER OCTOBER
NOVEMBER DECEMBER. JANUARY FEBRUARY
1977 I 1978
Figure 9» Percentage N removed for all effluent application rates on sand*
18 8# i
Effluent Application Rote (mm/doy)
-- ■ 10
* • 17 o s 22 o b
------■ 4 3
NOVEMBER DECEMBER JANUARY FEBRUARY
1977 | 1978
Figure 10. Percentage N removed for all effluent application rates on mix
56 parameter showed the main effects of rates, soils, and time were all significantly different at the p = 0.01 as shown in Table 9»
Figure 9 illustrates purification efficiencies for all five rates, averaging replicates, on the sand for the year long study.
Figure 10 illustrates the same data for the mix soil. Table 10 shows the mean percent N removal. The 10 mm/day rate was significantly dif ferent from the other rates. The 17 and 22 mm/day and, 34 and 43 mm/day were also significantly different from each other. Rates were not ranked on each soil because the analysis of variance showed that the rate by soil interaction was not significant (Table 9). Figures
9 and 10 illustrate this relationship of rates and percent N removal.
The lowest effluent application rate consistently had the highest percent N removal. At 10 mm/day percent N removed was 8lS£ compared to 44# N removal at 43 mm/day, when averaged over both soils and time.
Percentage of N removed was larger at the lower application rates due
partly to the fact that less N was applied.
Table 11 clarifies this relationship of percentage N removed and amounts of N actually removed. Even though percentage N removal decreased with increasing effluent application rates, the amount of N being removed increased with increasing rates. This table also indi cates the important role of soil processes in effluent renovation.
As indicated by percent N removal (Table 10), the mix soil was more effective in removing N than the sand soil. Figures 9 and 10 show that percent N removal was lowest during times of scalping in June and November. With no turf growing on the plots, the soil filter was responsible for N removal. This illustrates the role of utilization
Table 9- Analysis of variance for percent N removed, 11 July 1977 to 28 April 1978.
R x S
R x T
S x T
R x S x T
••Significant at the 0.01 level of probability.
Percent N removed means and average ppm -N in leachate.*
N Removed ppm -N
Leachate ppm -N
LSD = 7.2
Any two means with a letter in common are not significantly different
(p = O.
•Assuming 20 ppm -N in effluent applied.
Table 11. Total N removed by clippings and by soil processes from
30 April 1977 to 28 April 1978.
Total N Applied in Effluent (g)
Total N in
Total N Removed by Soil (g)
59 in maintaining a high percentage N removed. Table 10 also shows which rates of effluent met USPHS standards for N0~-N in the recharge water.
On the sand soil, up to 1? mm/day could be applied. On the mix soil, up to 34 mm/day of effluent still yielded potable water, based on nitrogen content only.
Tables 10 and 11 clearly show the mix soil was more effective at removing N. Although both soils were texturally sands, the mix soil contained 496 more silt and 2)6 organic matter. These factors con tributed to a higher CEO in the mix soil. It is also possible the hydraulic conductivity of the mix soil was less than the sand, allowing more time for nutrient absorption by plants, and adsorption of nutri ents by roots and soil. Table 11 shows that N in clippings and N removed by soil processes were both higher on the mix soil. The higher yields on the mix (Table 4) indicate the importance of CEC to
Increased N removal on the mix soil could also indicate greater root adsorption or denitrification.
Rate by time interactions were significant and this could be caused by many factors. Figures 9 and 10 show initially high values of percentage N removal at the beginning of the study. This was probably due to NH^ adsorption in the soil as plots had not been fer tilized for three months prior to the study. During the summer months, until October, the N in the effluent was mainly present as NH^-N (Fig.
1). The form of N in the effluent would affect N removal processes.
The initially high values of percentage N removal in this study, indi cate the importance of cation exchange reactions in removing N from the effluent. When NH^-N was the predominate form of N being applied,
60 this cation could be held on CEO sites for plant uptake, or adsorbed by plant roots. Ammonium held on cation exchange sites would also be more available for microbial immobilization.
High N removal from August to October corresponds with high N utilization during bermudagrass growth. The adsorbed NH^ was being used by plants. Warmer summer temperatures could have increased micro bial activity contributing to increased N removal and increased turf growth rates. By the end of October the main form of N in the effluent was N0”-N which is not adsorbed on cation exchange sites, but tends to move with the soil water. Figures 9 and 10 both show decreased per centage N removed during the months when the effluent was high in
N0~-N. Nitrogen removal on the sand was decreased more than removal on the mix. Nitrogen removal could also have decreased in winter due to lower temperatures decreasing turf growth and microbial activity.
Organic matter in the mix soil may have contributed to a larger microi bial population by supplying adsorption sites for nutrients and microorganisms, as well as increasing moisture available to micro organisms and plants. Figures 6 and 7 both show that N utilization increased in the winter when ryegrass was growing. Table 8 indicates increased percent N in clippings of ryegrass. These factors indicate that root absorption of N0~-N may have been important, during winter months, for removing N from the effluent.
Rainfall and lower temperatures also contributed to decreased
N removal in October and January. Figures 9 and 10 show large de creases in percentage N removed during these months which corresponded to high rainfall (Fig. 4) and high N0”-N in effluent (Fig. 1). High
61 rainfall increased soil moisture, increasing the velocity of effluent movement through the soil. Because soil particles were already water coated, effluent moved through soil pores with less interaction with the soil and root surfaces. This would decrease plant absorption of
NOj and microbial uptake. Microbial denitrification could be decreased in the winter due to lower temperatures slowing microbial action. All these factors could contribute to the significant rate by time inter action as well as soil by time interaction.
The decrease in N removal at the end of January illustrates a time when N utilization was increasing but percentage N removed de creased. This showed, despite N utilization by turf, high rainfall decreased the efficiency of root and soil processes to remove N.
These same events would contribute to decreased purification during reseeding, when tap water was applied daily. Effluent flow may have been faster, decreasing cation exchange reactions and plant roots in the soil would be acting as adsorption sites only, with ho active uptake, due to removal of the growing turf. Nitrogen removal in creased as a new turf stand grew.
Figure 11 shows percentage N removed with the five rates on each soil averaged. Percentage N removed for all times and rates by the mix soil was 129» greater than removal by the sand. Greater per centage N removed was observed during periods without rainfall, and when a good turf stand was established. This graph shows a relation ship similar to Figure 8 for utilization. Figure 11 shows that the soils responded differently over the year long study and this inter action was significant (Table 9). The difference in soil response is
II 25 9 23 6 20 3 17 I
15 29 12 26 10 24 7 21 4
18 4 18 I
Figure 11. Percentage N removed showing all five rates of effluent application averaged for each soil.
63 illustrated by the fluctuating distance between the two soil lines.
Sand-soil N removal was more drastically reduced in unfavorable periods, and did not recover efficiency as rapidly as the mix-soil. This inter action between soils over time, found in utilization and N removal indicates the important role of CEO, root surface area, and organic matter in these processes. Variations in moisture, temperature, and effluent composition affected the chemical reaction of cation exchange as well as biological uptake and use of nutrients from the effluent.
These soil characteristics, in turn, affect turf growth. The mix soil appeared to be better buffered against fluctuations in the environment and turf recovered more rapidly from adverse conditions. These re sults suggest soils containing more clay and organic matter would be able to remove more N from effluent. Good soil conditions for turf growth contribute to higher N removal by turf and by soil processes.
Nitrogen Transformations in Leachate
Organic carbon was not directly measured in this experiment, but a time study of N transformations in leachate samples was conducted in September, when NH^-N in the effluent was high. Leachate samples were taken from the 43 mm/day rate at 0, 1, 2, 3i 4, 3, and l8 hours.
No change in NH^ or NO” content was found between 5 to 18 hours. In the first five hours of the time study, NH^ content decreased and N0”-N increased by 5 ppm in the mix leachate and 1 ppm in sand leachate.
This change from NH^ to NO™ indicated that autotrophic bacteria were present in the leachate and had sufficient carbon dioxide or carbonate to affect an increase in N0”-N. The fact that the N0™-N content
64 remained constant indicated either a lack of denitrifying organisms or a lack of organic carbon for denitrification. Since denitrifying bacteria are usually present in the soil, it is probable that lack of denitrification was due to the absence of an energy source. Nitrite levels were spot checked in leachate samples and no NO" was found.
This indicated nitrification of the effluent was complete.
As mentioned earlier, leachate samples stored at room tempera ture did not show any changes in NH^, NO", or organic-N content. If leachate samples were stored in the light, algae growing in the bottles decreased the nitrogen content of the leachate. Effluent samples stored at room temperature did show NH^ to NO" transformations.
The Soil's Role
The soil's role in this soil-turf filter system was determined indirectly. The role of the soil system in renovation was considered to be the difference between percentage N removed and percentage N utilization. Nitrogen removal statistics showed the soils were dif ferent at the p = 0.01 level. The mix-soil removed more N than the sand. This could be explained by the higher CEC and organic matter in the mix soil. Because the mix-soil could adsorb more NH^, there was more N available to soil microorganisms and plants. There are several soil processes that could contribute to purification: adsorp tion, denitrification, and volatilization. The role of adsorption is clear from purification values for the first month. Initial adsorption of NH^ yielded purification efficiencies over 9096 even though utili zation was only 109$. The CEC of the mix soil is 5 meq/lOOg soil.
This is more than twice the CEO of the sand. This higher CEC con tributed to higher purification and utilization on the mix. CEC played a more important role in purification when NH^ was the predominate ion in the sewage effluent. Higher grass yields on the mix soil (Table
4) could have resulted from increased adsorption of the NH^, increasing N available for growth. The higher yields on the mix soil could indi cate more root surface area for adsorption and absorption of N con tributing to greater N removal.
Denitrification is a biological process occurring in the soil which would improve purification. It is evident from this study that nitrification was occurring in the soil because effluent applied (May through September, as shown in Fig. 1) was predominately NH^ and the leachate collected contained only NO^. Conditions necessary for deni trification limited purification. The sand soils used have hi$i in filtration rates and permeability, decreasing the possibility of creating micro-anaerobic zones for denitrification. The organic carbon content of the effluent was another factor limiting denitrification.
Although organic carbon was not directly measured in this experiment, the stability of NO~-N concentration in leachate samples indicated that no denitrification was occurring, and hence the lack of an organic carbon source. This indicated any organic carbon originally present in the effluent was consumed or filtered as it passed through the soil.
These results suggest purification of effluent could be improved by maintaining a water table, for an anaerobic zone, and adding addi tional energy sources for denitrifying anaerobes. The turf rhizosphere could contribute to denitrification in the soil. Roots use oxygen in
66 the soil, creating an anaerobic zone and contribute substrates that could be used by denitrifying organisms for energy.
Volatilization could <have contributed to purification of effluent. In the summer months, when the effluent contained mainly
NH^ and was applied to plots with a pH of 8.3, conditions are favor able for volatilization. Surface ponding of applied effluent could increase volatilization. Ponding did not occur on these soils due to their high infiltration rates, so volatilization was not maximized.
Nitrogen would also be removed by microbial immobilization as well as plant uptake. Fluctuation in purification and utilization over time was a result of environmental influences on these processes. Comparing
Figure 8 for utilization to Figure 11 for purification showed that although N utilization was increased in the winter months during rye grass growth, purification efficiency in the winter was not as high as purification in the summer. This indicated the importance of the soil i filter in purification. Summer temperatures increased biological activity and volatilization of NH^. During the cooler winter months biological activity was decreased because the temperatures"decreased below the
to 30 C optimum f o r "nitrification and denitrification; this subsequently decreased N losses due to soil processes. The higher utilization of N by ryegrass did not fully compensate for decreased biological activity in the soil. The importance of NH^ volatilization was not quantified.
Table 12 shows purification and utilization averages for the
21 time periods. The differences between these values indicate the percent N applied that was lost due to soil processes. Increasing
Table 12. Percent N removed by soil.
68 the amount of N applied, decreased the percentage of N being removed by the soil, although absolute values increased. Table 11 shows that even though percentage N removed by the soil decreased, the total amount of
N removed by the soil increased with increased effluent application rates. Table 11 indicates that the plateau for nitrogen removed by soil processes has not been reached, since total quantities of N re moved increased with each rate. Tables 7 and 8 show N removal by turf increased as effluent application rate increased. The yield plateau for the grasses appears to occur between the 3^ or 43 mm/day rates.
By optimizing conditions in the soil for nitrification and denitrifi cation, N removal by the soil could be increased.
SUMMARY AND CONCLUSIONS
This experiment investigated the potential of a soil-turf filter as a tertiary treatment for N removal from secondary effluent applied in excess of consumptive water use. Two soils were used.
Both soils were texturally sands. The sand soil had a CEC of 2 meq/lOOg soil, the mix soil contained 2# organic matter, 4# more silt and had a CEC of 5 meq/lOOg soil. Lysimeter plots were drip irrigated twice a week with effluent at rates of 10, 17, 22, 34 and 43 mm/day.
Leachate and effluent were analyzed for NH^-N, N0~-N, and organic -N.
Grass clippings were collected once a week. Clippings were oven dried, weighed, and analyzed for organic -N. From these analyses a nitrogen balance for the year long study was obtained.
The ability of the soil-turf filter to remove N from effluent was evaluated by percentage N removed. This parameter indicated the percent of N applied that was removed by the soil-turf filter.
Nitrogen utilization indicated the percent of N applied collected as grass clippings. Water recharge indicated the percent of water applied that would be available for reuse or groundwater recharge.
Results were evaluated using these three parameters.
This study showed that as effluent application rates increased, percentage N removed and N utilization decreased. Table 11 shows that even though percentage N removed decreased, total amounts of N in clippings and removed by soil processes increased with effluent
70 application rates. The plateau for maximum N removal by turfgrass occurred around the 3*+ to 43 mm/day rate. Soil removal of N increased with rate. On the mix soil 34 mm/day of effluent could be applied and yield leachate containing less than 10 ppm NO^-N. On the sand soil
17 mm/day would meet USPHS standards. Because N removal varied with season and turf conditions, leachate water from these rates may not always be less them 10 ppm N0“-N. These rates represent an average
N removal for the whole year. In seasons of low N removal less effluent could be applied. Conversely rates could be increased during more favorable times. During the year of this study, the total amount of N applied at the 34 mm/day rate amounted to 2182 lbs. N/acre. The amount of N utilized by turf at the same rate was 338 lbs. N/acre. At this rate, 74# of the water applied was available for reuse, amounting to 9 million gallons per acre per year. To purify the 35 million gal lon per day of effluent that Tucson produces would require 968 acres.
Table 4 indicated that increased effluent application rates could increase water consumption. Table 11 indicates increasing water consumption did not increase the amount of N removed by the turf. The sand soil averaged 75# water recharge, compared to 67# recharge from the mix. The mix soil averaged 27# N utilization compared to 19# on the sand. Nitrogen removal by the mix soil averaged 64# compared to
52# on the sand.
The results indicated the important role of cation exchange capacity, organic matter, water holding capacity, soil microorganisms, and vigorously growing turf in removing N from effluent applied to a soil-turf filter. Selecting soils with a greater CEC could maximize
the effectiveness of the soil filter. Selecting turfgrass genotypes
71 for maximum nutrient uptake would improve water purification. Turf grass, as a filter, offers the advantage of growing year round and providing recreational uses. Use of effluent for watering parks and golf courses would increase water for domestic use and could contribute to groundwater recharge.
LIST OF REFERENCES
Adriano, D. C., L. T. Novak, A. E. Erickson, A. R. Wolcott, and B. G.
Ellis. 1975* Effect of long term land disposal by Apray irrigation of food processing wastes on some chemical proper ties of soil and subsurface water. J. Environ. Qual. 4:242-
Amramy, A. 1968. Reuse of municipal wastewater. Civ. Eng. 38(5):
Bendixen, T. V/., R. Hill, W. A. Schwartz, and G. G. Roberts. 1968.
Ridge and furrow liquid waste disposal in a northern latitude.
J. Sanit. Eng. Div., Proc. Am. Soc. Civ. Eng. 9^(SA1):147-157•
Bouwer, H. 1970. Ground water recharge design for renovating waste water. J. Sanit. Eng. Div., Proc. Am. Soc. Civ. Eng.
. 1976. Zoning aquifers for tertiary treatment of wastewater. Groundwater 14(6):386-395*
Bouwer, H . , J. C. Lance and M. S. Riggs. 1974. High rate land treat ment: II Water quality and economic aspects of the Flushing
Meadows project. J. Water Pollut. Control Fed. 46:844-859*
Bouwer, H., R. C. Rice and E. D. Escarcega. 1974. High-rate land treatment: Infiltration and hydraulic aspects of the Flushing
Meadows project. J. Water Pollut. Control Fed. 46(5):834-843.
Bremner, J. M. 19&5. Total nitrogen; Inorganic forms of nitrogen;
Organic forms of nitrogen. In C. A. Black et al. (eds.)
Methods of soil analysis (part 2), Agronomy 9, pp* 1149-1254,
Am. Soc. Agron., Madison, Wis.
Chadwick, M. J., K. J. Edworthy, D. Rush, and P. J. Williams. 1974.
Ecosystem irrigation as a means of groundwater recharge and water quality improvement. J. Appl. Ecol. 11(1):231-247.
Cluff, C. B . , K. J. DeCook and W. G. Matlock. 1972. Technical, eco nomic, and legal aspects involved in the exchange of sewage effluent for irrigation water for municipal use. Case study,
City of Tucson. Technical completion report, Office of Water
Resources Research, project A-022, Arizona.
Corey, J. C . , D. R. Nielson and D. Kirkham. 1967* Miscible displace ment of nitrate through soil columns. Soil Sci. Soc. Am. Proc.
Crites, R. V/. 1973* Wastewater irrigation: how. Water Wastes Eng. 12(7):49-30*
This city can show you
Culp, R. L. 1963* Wastewater reclamation by tertiary treatment.
J. Water Pollut. Control Fed. 35(6):799-806.
__________ . 1968. Wastewater reclamation at South Tahoe Public
Utilities District. J. Am. Water Works Assoc. 60:84-94.
Day, A. D. 1973* Recycling urban effluents on land using annual crops. Proceedings of joint conference on recycling municipal sludges and effluents on land. July, 1973, Champaign, 111.
Day, A. D., A. D. Dickson and T. C. Tucker. 1963* Effects of city sewage effluent on grain yield and grain and malt quality of fall-sown, irrigated barley. Agron. J. 55:317-318.
Day, A. D. and R. M. Kirkpatrick. 1973* Effects of treated municipal wastewater on oat forage and grain. J. Environ. Qual. 2:
Day, A. D., Abdul Rahman, F. R. H. Katterman and V. Jenson. 1974.
Effects of treated municipal wastewater and commercial fer tilizer on growth, fiber, acid-soluble nucleotides, protein, and amino acid content in wheat hay. J. Environ. Qual.
Day, A. D., J. L. Stroehlein and T. C. Tucker. 1972. Effects of treatment plant effluent on soil properties. J. Water Pollut.
Control Fed. 44:372-375*
Day, A. D. and T. C. Tucker. 1977* Effects of treated municipal wastewater on growth, fiber, protein, and amino acid content of sorghum grain. J. Environ. Qual. 6:325-328.
Day, A. D . , T. C. Tucker and M. G. Vavich. 1962a. City sewage for irrigation and plant nutrients. Crops Soils l4(8):7-9*
. 1962b. Effect of city sewage effluent on the yield and quality of grain from barley, oats, and wheat. Agron. J.
DeJong, E. 1978. The movement of sewage effluent through soil columns: The major ions (Na, Ca, Mg, Cl, and SCh )• J. Environ.
Eastman, P. W.
. Municipal wastewater reuse for irrigation.
J. Irrig. Drain Div., Proc. Am. Soc. Civ. Eng. 93(123):25-31*
Eliassen, R. and G. Tchobanoglous. 1968. Chemical processing of wastewater for nutrient removsil. J. Water Pollute Control
Enfield, C. G. 1977. Servo controlled optimization of nitrificationdenitrification of wastewater in soil. J. Environ. Qual.
Evans, J. 0. 1970. The soil as a resource renovator. Environ. Sci.
Francis, C. W. and M. W. Callahan. 1975• Biological denitrification and its application in treatment of high-nitrate wastewater.
J. Environ. Qual. 4(2):153-163.
Hajek, B. F. 1969* Chemical interactions of wastewater in a soil environment. J. Water Pollut. Control Fed. 41:1775-1786.
Hill, D. E. 1972. Wastewater renovation in Connecticut soils. J.
Environ. Qual. 1(2):163-167.
Hook, J. E. and L. T. Kardos. 1978. Nitrate leaching during long term spray irrigation for treatment of secondary sewage effluent on woodland sites. J. Environ. Qual. 7:30-35.
Hortenstine, C. C. 1976. Chemical changes in the soil solution from a spodosol irrigated with secondary treated sewage effluent.
J. Environ. Qual. 5(3):335-338.
Johnson, G. V. 1973. Irrigating recreational turfgrass with sewage effluent. Prog. Agric. Ariz. 25:8-9, 15.
Kardos, L. T. and W. E. Sopper. 1973. Renovation of municipal wastewater through land disposal by spray irrigation.
In W. E. Sopper and L. T. Kardos (eds.), Recycling treated municipal wastewater and sludge through forest and cropland, pp*, 148-163.' Penn. State Univ. ,Press, Univ. Park, Pa.
Klausner, S. D. and L. T. Kardos. 1975. Oxygen relationships in a
Krans, J. V. and G. V. Johnson. 1974. Subirrigation and fertilization of bentgrass during prolonged heat stress. Second Intern' 1
Turfgrass Proc., pp. 527-533*
Lance, J. C. 1972. Nitrogen removal by soil mechanisms. J. Water
Pollut. Control Fed. 44:1352-1361.
Lance, J. C. and F. D. Whisler. 1972. Nitrogen balance in soil columns intermittently flooded with secondary sewage effluent.
J. Environ. Qual. 1:180-186.
__________ . 1976. Stimulation of denitrification in soil columns by adding organic carbon to sewage water. J. Water Pollut.
Control Fed. 48:346-356.
Lance, J. C . , F. D. Whisler and H. Bouwer. 1973• Oxygen utilization in soils flooded with sewage water. J. Environ. Qual. 2:
Lance, J. C., F. D. Whisler and R. C. Rice. 1976. Maximizing denitrification during soil filtration of sewage water.
J. Environ. Qual. 5:102-108.
Law, J. P., Jr., R. E. Thomas and L. H. Meyers. 1970. Cannery wastewater treatment by high-rate spray on grassland. J. Water
Pollut. Control Fed. 42:1621-1631.
Lehman, G. S. 1968. Soil and grass filtration of domestic sewage effluent for the removal of trace elements. Unpublished doctoral dissertation, University of Arizona, Tucson.
Mann, L. D . , D. D. Focht, H. A. Joseph and L. H. Stolzy. 1972.
Increased denitrification in soils by additions of sulfur as an energy source. J. Environ. Qual. 1(3):329-332.
McKee, J. E. and H. W. Wolf.
. Water quality criteria. Publi cation 3-A, California State Water Resources Control Board.
Meek, B. D . , L. B. Grass and A. J. MacKenzie. 1969. Applied nitrogen losses in relation to oxygen status of soils. Soil Sci. Soc.
Am. Proc. 33:575-578.
Osborne, J. M. 1975. Tertiary treatment of campground wastes using a native Minnesota peat. J. Soil Water Conserv. 30(5)$235-236.
Overman, A. R. and Hsiao-Ching Ku. 1976. Effluent irrigation of rye and ryegrass. J. Environ.Eng. Div., Proc. Am. Soc. Civ. Eng.
Parizek, R. R. 1973• Site selection criteria for wastewater disposal
— Soils and hydrogeologic considerations. In W. E. Sopper and L. T. Kardos (eds.) Recycling treated municipal wastewater and sludge through forest and cropland, pp. 95-147. Penn.
State Univ. Press, Univ. Park, Pa.
Parker, R. L . , J. Wang, M. R. Overcash and F. J. Humenik. 1974.
Lysimeter studies with long term application of swine lagoon effluent. Am. Soc. Agric. Eng. paper No. 74-4036.
Parkhurst, J . D.
. Progress in wastewater reuse in southern
California. J. Irrig. Drain. Div., Proc. Am. Soc. Civ. Eng.
Parness, W. H. 1968. Golf course and airport irrigation with sewage effluent. Am. City
Pennypacker, S. P.$ W. E. Sopper and L. T. Kardos. 1967* Renovation of wastewater effluent by irrigation of forest land. J. Water
Pollut. Control Fed. 39(2):
Pratt, P. F . , M. D. Thorne and F. Wiersma. 1977. Future direction of waste utilization. I n , L. F. Elliott and F. J. Stevenson (eds.)
Soils for management of organic wastes and waste waters, pp.
621-635. Soil Sci. Soc. Am., Madison, Wis.
Preul, H. C. and G. J. Schroepfer. 1968. Travel of nitrogen in soils.
J. Water Pollut. Control Fed. 40:30-48.
Rice, R. C. 1974. Soil clogging during infiltration of secondary effluent. J. Water Pollut. Control Fed. 46:708-716.
Roll, Ed. 1974. New Arizona Wastewater plant to alleviate problems.
Water Sewage Works 121(4):60.
St. Araant, P. P. and P. L. McCarty.
. Treatment of high nitrate waters. J. Am. Water Works Assoc. 61:659-662.
Sidle, R. C. and G. V. Johnson. 1972. Evaluation of a turf-grasssoil system to utilize and purify municipal wastewater.
Hydrology and Water Resources in Arizona and the Southwest
Slechta, A. F. and G. L. Culp.
. Water reclamation. J. Water
Pollut. Control Fed. 39:787-814.
Smith, J. H., R. G. Gilbert and J. B. Miller. 1976. Redox potentials and denitrification in a cropped potatoe processing wastewater disposal field. J. Environ. Qual. 5(4):397-400.
Sopper, W. E. and L. T. Kardos. 1973. Vegetation responses to irri gation with treated municipal wastewater. In W. E. Sopper and L. T. Kardos (eds.) Recycling treated municipal wastewater and sludge through forest and cropland, pp. 271-294.
Penn. State Univ. Press, Univ. Park, Pa.
Stenburg, R. L., J. J. Convery and C. L. Swanson. 1968. New approaches to wastewater treatment. J. Sanit. Eng. Div., Proc. Am. Soc.
Civ. Eng. 94:1121-1136.
Stephan, D. G. and L. V. Weinberger. 1968. Wastewater reuse — Has it arrived? J. Water Pollut. Control Fed. 40:529-539.
Thomas, R. E. 1973• The soil as a physical filter. In W. E. Sopper and L. T. Kardos (eds.) Recycling treated municipal wastewater through forest and cropland, pp. 38-4$. Penn. State Univ#
Press, Univ. Park, Pa.
Thomas, R. E. and C. C. Harlin, Jr. 1972. Experiences with land spreading of municipal.effluents. First Annual Institute
Food Agric. Sci. Workshop on land renovation of wastewater,
31 May-1 June 1972, Tampa, Florida.
Tofflemine, T. J. and R. A. Farnan. 1975• Land disposal of wastewater. J. Water Pollut. Control Fed. 47:1344-1352.
Whisler, F. D., L. C. Lance and R. S. Linebarger. 1974. Redox poten tial in soil columns intermittently flooded with sewage water.
J. Environ. Qual. 3(1)i68-74.
White, W. T. 1975. Reusing waste in St. Petersburg goal. Water
Wastes Eng. 12(3):64-69.
Wilson, L. G. and G. S. Lehman. 1967* Reclaiming sewage effluent.
Prog. Agric. Ariz. 19(4):22-24.
Woldendrop, J. W. 1962. The quantitative influence of the rhizosphere on denitrification. Plant Soil 17:267-270.
Young, E. C. and G. A. Carlson. 1975. Land treatment versus con ventional advanced treatment of municipal wastewater. J.
Water Pollut. Control Fed. 47:2565-2573.
Youngner, V. B., T. E. Williams and L. R. Green. 1974. Ecological and physiological implications of greenbelt irrigation.
Progress report Univ. Calif. Water Resources Center contribu tion No. 148.
7 4 9
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project