WATER POLLUTION CONTROL RESEARCH SERIES • 16080 —-10/69
NUTRIENT REMOVAL FROM ENRICHED
WASTE EFFLUENT BY THE HYDROPONIC
CULTURE OF COOL SEASON GRASSES
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER QUALITY ADMINISTRATION
-------
NUTRIENT REMOVAL FROM ENRICHED WASTE EFFLUENT
BY THE HYDROPONIC CULTURE OF COOL SEASON GRASSES
by
James P. Law, Jr., Research Soil Scientist
Water Quality Control Research Program
Robert S. Kerr Water Research Center
South Central Region
Ada, Oklahoma
for the
FEDERAL WATER QUALITY ADMINISTRATION
DEPARTMENT OF THE INTERIOR
Program #16080
October, 1969
-------
FWQA REVIEW NOTICE
This report has been reviewed by the Federal
Water Quality Control Administration and ap-
proved for publication. Approval does not
signify that the contents necessarily reflect
the views and policies of the Federal Water
Quality Control Administration, nor does
mention of trade names or commercial products
constitute endorsement or recommendation for
use.
‘or sate by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C. 20402- Price 50 cents
-------
ABSTRACT
Grasses were grown in hydroponic culture tanks to evaluate their nutrient
removal capabilities when supplied with secondary—treated sewage effluent
as the sole source of plant nutrients. Statistical methods were employed
to determine the effects of the grasses, flow rates, and seasons on nutri-
ent removal.
Two control tanks with gravel bed and no grass were maintained through-
out the study, two were planted with tall (Kentucky 31) fescue, and two
were planted with perennial ryegrass. Two flow rates were maintained In
each pair of tanks, approximately one—day and two—day detention times.
All six tanks were effective in reducing the oxygen—demanding organic
content of the effluent. Total nitrogen content was reduced appreciably
by the control tanks, but the grass tanks were significantly better at
nitrogen removal. Total phosphorus concentrations were reduced only
slightly by passage through the tanks. The fast flow rate tanks pro-
duced the greater grass yields, while the slow flow rate tanks were
more effective in nutrient removal from the sewage effluent.
From grass yield and analyses data, the amount of plant nutrient material
removed by the grasses was small compared to the total quantity supplied.
Further studies are suggested to determine economic factors and design
parameters.
Key Words: Hydroponic culture, tertiary treatment, waste water treatment,
Nutrient removal, cool season grasses, waste assimilative
capacity.
I
-------
TABLE OF CONTENTS
Page
1
1
2
5
5
11
11
11
13
15
15
19
20
20
20
21
23
28
31
. .
Introduction . . . .
Need forResearch
The Nutrient Problem .
Experimental Plan
Study Site
Experimental Procedures
Sampling Program
Grass Samples . .
Results and Discussion . . . . .
Characteristics of Effluent from Hydroponic
Treatment Efficiencies
Nitrogen removal
COD removal
BODremoval.
Total phosphorus
Grass Yields and Quality
Nutrients Removed by Harvest
Bacterial Reductions
Ac ow1ed ent . .
References
• .
•
• . • . . • . .
• . • • . • . •
•
Tanks . • • •
• S • • • S S •
S
• S S • • • S S
. S • • • • • S
. • . . . . 33
iii
-------
LIST OF FIGURES
Figure Page
1. SchematicofElevationPian. . . . . . . 7
2. Flow Diagram of the Six Tanks . 8
3. Sediment Chamber and Orifice Arrangement for
RegulatingFlov to Tanks...... 9
V
-------
LIST OF TABLES
Table Page
1. Characteristics of Sewage Treatment Plant Effluent
Supplied to the Hydroponic Tanks 14
2. Estimated Average Rates of Supply of Nacronutrients
to the Hydroponic Tanks 15
3. Mean Characteristics of Effluent from the Control
Tanks 16
4. Mean Characteristics of Effluent from the Fescue
Tanks . 17
5. Mean Characteristics of Effluent from the Perennial
RyegrassTanks 18
6. Mean Removal Percentages for Total Nitrogen, COD,
and BOD (Concentration Basis) 19
7. Mean Total Phosphorus Concentrations in Sewage
Effluent and Effluents from the Hydroponic
Tanks, mg/l 21
8. Moisture Content and Dry Weight Yield of the
Grasses . . . . 22
9. AnalysesofGrassSaiuples 24
10. Plant Nutrients Removed by the Grass Harvests,
Pounds PerAcre • • • 25
11. Plant Nutrients Leaving the Hydroponic Tanks,
Pounds Per Acre (Based on Outflow for 60 Weeks) • • • 26
12. Fate of Nitrogen, Phosphorus, and Potassium
Supplied to the jlydroponic Tanks During 60—Week
Period. . . 27
13. Mean Bacterial Counts and Percent Reductions After
Flow Through the Hydroponic Tanks . . . . • 29
vii
-------
CONCLUS IONS
Grasses were grown in hydroponic culture tanks to evaluate their nutrient
removal capabilities when supplied with secondary—treated sewage effluent.
The effluent supplied was the sole source of plant nutrients. Chemical
analyses of the sewage effluent supplied and of the outflow from each of
six hydroponic tanks were performed weekly during the one—year study.
Two control tanks with no grass were maintained throughout the study, two
tanks were planted with tall (Kentucky 31) fescue, and two were planted
with perennial ryegrass. Two flow rates were maintained in each pair of
tanks, approximating one—day and two—day detention times. Statistical
methods were employed to determine the effects of the grasses, flow rates,
and seasons on nutrient removal.
All six tanks were quite effective in reducing the concentrations of
oxygen—demanding organic materials in the effluent. Total nitrogen con-
tent was reduced appreciably by the control tanks, but the grass tanks
were significantly better at nitrogen removal. Total phosphorus concen-
trations were reduced only slightly by passage through the tanks. The
fast flow rate tanks produced the greater yields, while the slow flow
rate tanks were more effective In nutrient removals from the sewage effluent.
The fescue produced over 10 tons per acre dry weight yield and the rye—
grass yield was about 7 tons per acre for the year. From the grass anal-
yses, the amount of plant nutrient material removed by harvest was com-
puted. The quantities actually removed by the grasses were rather small
compared to the total quantity supplied. About 4 to 8 percent of the
nitrogen supplied was accounted for in the grass harvested, while about
2 to 5 percent of phosphorus and 6 to 22 percent of the potassium were
removed in the harvests.
ix
-------
RECOMMENDATIONS
The results of this study suggest several avenues that require further
investigation before the method could be recommended as feasible. Studies
on a much larger scale should be conducted to determine the economic
factors of the method. Field studies comparing the hydroponic technique
with normal spray irrigation of grassland would appear desirable.
The indications are that shallower gravel beds could be used. Optimum
bed depth and dimensions need to be determined from future studies. The
effluent from the Ada sewage treatment plant averages between four and
six acre—feet per day. If beds were constructed one—foot deep and operated
with a two—day detention time in support media with 40 percent void space,
rough calculations quickly show that 20 to 30 acres of bed area would
be required to handle the entire flow.
Additional studies should determine the most desirable grass species or
combinations of species from a nutrient utilization standpoint. These
should produce luxuriant growth in all seasons of the year for maximum
nutrient uptake. Ryegrass proved undesirable in this study because of
its dormancy during the summer months.
The results of this study showed poor performance with regard to total
phosphorus removal. This suggests that a different type of support
medium, such as crushed limestone, might be desirable. In this way, the
phosphates might be complexed with calcium and thus removed from the
effluent.
In general, the results indicated a relatively minor role played by the
grasses in removing the nutrients supplied. A much closer look at the
feasibility of the method in future studies is suggested.
xi
-------
INTRODUCTION
Hydroponics is defined as the growing of plants in nutrient solutions
with or without an inert medium to provide mechanical support. It has
become well established as a culture method for many different types of
plants and is especially adapted to greenhouse operations where light,
temperature, humidity, and atmosphere can be controlled. Under normal
operations, solutions are prepared with the proper proportions of all
required nutrients and are fed to the root systems of growing plants.
The composition of these solutions may be varied to suit the nutrient
requirements of different plant species and in most commercial operations
this becomes a well—guarded trade secret.
Hydroponic culture is widely used by the floriculture industry and is
gaining much popularity in the production of high—valued vegetable crops
such as tomatoes. A recent news release in the Sunday Oklahoman reported
a commercial operation in Edmond, Oklahoma, that utilizes two 4,000—square—
foot greenhouses and expects a yield of 40,000 pounds of tomatoes during
a 10—week fruiting season; and this during the months when no field pro-
duction is possible In the area.
Another interesting report appeared in the Denver Post (1) which described
several hydroponics enterprises in that area. Kelly—Clark, Inc., of
Loveland produces tomatoes with the trade name, “Gourmato,” at the rate
of 1,000 pounds per week from three greenhouses. Three growers in Golden
market 3,000 pounds of tomatoes per week. Smaller “sideline” operations
at Sterling, Colorado Springs, and Craig each produce from 700 to 1,000
pounds of tomatoes per week. A dairy owner at Deer Trail grows 1,000
pounds of barley grass “green feed” daily for his herd of 40 cows. The
Denver Zoo produces from 700 to 1,000 pounds of barley grass daily for
feeding waterfowl; and at the Denver Federal Center, a unit of the U. S.
Fish and Wildlife Service grows 500 pounds of oat grass per day for birds
and game animals kept for research purposes.
The fact that many plants will grow and extract their nutrient requirements
from solution inspired the interest in the present study. This was viewed
as a potential method for the biological extraction of nutrient materials
from municipal waste effluents that are now being released to many of our
surface streams. The principal objectives of this preliminary study were
to find the most suitable grasses for this culture method, their yield on
a seasonal basis, and the amount of nutrients that could be removed by the
process.
Need for Research
The Water Quality Act of 1965 required all States to establish water
quality standards for their Interstate and coastal waters. These standards
are the plans established by governmental authority for water pollution
control and abatement, and they required the States to make crucial de-
cisions regarding the uses of their water resources, the quality of water
1
-------
to support these uses, and specific plans for achieving such levels of
quality. The purpose of the program is to enhance the quality and value
of the nation’s waters and to protect the quality of clean water. These
standards are the blueprints for an effective clean water program. The
Federal Water Pollution Control Administration (FWPCA) set forth guide-
lines (2) to assist the States in establishing acceptable water quality
standards and implementation plans. Those guidelines clearly established
that water quality standards should be designed to “enhance the quality
of water.” They further stated that the implementation plan “...should
include consideration of all relevant pollutional sources, such as munic-
ipal and industrial wastes, cooling water discharges, irrigation return
flows, and combined sewer overflows.” The result of these policy state-
ments is that standards and implementation plans which have been approved
call for secondary treatment of municipal wastes and equivalent levels of
treatment for industrial wastes (3).
Even though the present state of technology is such that adequate treat-
ment can be given to municipal wastes, many of the smaller municipalities
simply are not in a position that allows them to employ the most modern
treatment methods; and little imagination is required to envision the day
when secondary treatment will not be acceptable in many cases. New and
fresh approaches to waste treatment problems are called for. The hydro—
panic culture of higher plants to extract nutrient materials from waste
effluent was viewed in this light. To our knowledge, no prior investiga—
tion of this nature has been carried out. These considerations prompted
the present study as a preliminary step in meeting this need.
The Nutrient Problem
One of the most serious water pollution problems we face is the release
of nutrient compounds of nitrogen and phosphorus to our waterways. Sewage
treatment processes have been primarily designed to reduce the suspended,
organic, oxygen—demanding materials. Efficiently operated secondary treat-
ment plants remove about 90 percent of these materials, about 90 percent
of the bacteria, 50 percent of the total nitrogen, 20 to 40 percent of the
phosphorus, and only about 5 percent of the total dissolved matter. Vein—
berger et al . (4) show that only a very small fraction of river basins in
the United States have complete secondary treatment of all municipal dis-
charges, while a somewhat larger percentage of river basins have more than
90 percent of such discharges with complete treatment. They estimated the
total nitrogen and phosphorus discharges to surface streams since 1900 with
projections to the year 2000. According to their estimates, in 1900 there
were about 200 million pounds per year of total nitrogen (as N) being dis-
charged to U. S. streams. By 1965, this had grown to 1.2 billion pounds
per year and an increase to 2.2 billion pounds per year is expected by 2000.
Phosphorus discharges were estimated at about 20 million pounds per year
(as P) in 1900. [ n 1965, the estimate was 250 million pounds per year
and by 2000 the prediction is something over 500 million pounds per year.
The expected increase in secondary treatment facilities beyond 1965 accounts
for a slight decrease in the slope of these projected curves.
2
-------
Estimates (4) of biochemical oxygen demand (BOD) discharges to streams
from municipal outfalls paint another interesting picture From 1900 to
1930, BOD discharges increased from 1.5 to 3.5 billion pounds per year.
In 1930, only 15 percent of municipal sewage received primary treatment
and less than 6 percent secondary. During the 1930’s, sewage treatment
plant construction was accelerated and BOD discharges decreased slightly.
A sharp increase during World War II and a levelling—off period which
followed brought a peak of 4.5 billion pounds per year in 1958, followed
by a slight decline to 4 billion pounds per year by 1965. Assuming that
in the next 20—year period all municipal wastes will receive primary-
secondary treatment, a minimum of 2.5 billion pounds per year can be
reached by 1985. From that point on, a steady increase due to population
growth will bring us again to the 4.5 billion pound level by the year 2000.
Projections such as these strongly emphasize the fact that treatment above
and beyond our present secondary treatment capabilities will be required
in the foreseeable future.
Admittedly, the above analysis is on a gross scale, but it gives some
indication of the magnitude of the nutrient problem and the trend expected
in the years ahead. The direct result of nutrient discharges to our lakes
and streams is a greatly accelerated rate of eutrophication (5). Those
nutrient compounds that are not removed in waste treatment processes occur
in the effluent discharges in the soluble, more readily available forms,
mostly as nitrates and phosphates. Their presence, even at low concen-
trations, can stimulate prolific growths of algae and other aquatic vege-
tation. The death and decay of such vegetation then exerts an added oxygen
demand on the water which may adversely affect fish and other aquatic life.
After degradation, the nutrients are released back to the water in soluble
form and the process re—cycles. The only alternative to the problem of
induced eutrophication is to employ treatment methods and technology to
further control the release of nutrient materials to our surface waters (5).
The hypothesis of this study proposed that higher plants could be grown
either floating on lagoons or using a gravel—bed support in specially
built structures for the removal of nutrient materials remaining in sewage
waste effluents. The methods employed and the results obtained constitute
the remainder of this report.
3
-------
EXPERIMENTAL PLAN
The study was carried out in six large metal tanks constructed for the
purpose. The tanks were built of galvanized metal and were 20 feet long,
3 feet wide, and 18 inches deep. Fittings were provided at the inlet and
outlet ends and located near the bottom on each tank. Distribution mani-
folds were fitted horizontally inside the inlet and outlet ends of each
tank to aid in better distribution of the fluid flow. Clean pea gravel
was used as the inert support medium to a depth of 14 to 15 inches in each
tank. A 2—inch layer of coarse concrete sand was placed on top of the
gravel bed, and the grasses were planted in the moist sand layer. The
liquid level was adjusted to the top of the pea gravel layer and maintained
constant by a standpipe overflow at the outlet end.
An overhead tank supplied a constant head pressure of about 4 feet to the
manifold and supply lines leading to each tank. A schematic drawing of
the elevation plan is shown in Figure 1. The complete flow diagram from
the overhead tank through the six tanks is represented in Figure 2. Two
tanks were operated with no grass to serve as controls. Tall (Kentucky 31)
fescue sprigs were planted in tanks 3 and 4, and perennial ryegrass seed
were planted in tanks 5 and 6. Rows were planted on 6—inch centers, and
the fescue sprigs were spaced 4 to 6 inches apart in the rows. These two
grasses were selected on the basis of a study reported by Eby (6) in which
several pasture grasses were successfully grown in animal wastes by hydro-
ponic culture in the greenhouse.
The experimental design was such that factorial analysis could be employed
to evaluate the effect of two flow rates and four seasons on the nutrient
removal efficiency of the two grasses versus the control tanks. The flow
rates were adjusted and maintained by means of a sediment chamber and ori-
fice arrangement shown schematically in Figure 3. Two flow rates were
employed to give approximately a one—day detention time (fast rate) and
a two—day detention time (slow rate) in each of the pairs of tanks—-control,
fescue, and ryegrass (Fig. 2). Thus, the principal variables of the study
were flow rate, grass species, and season. Factorial analysis of variances
was employed to detect significant effects of these variables as well as
interaction effects of the several combinations.
Study Site
The study was located at the Ada, Oklahoma, sewage treatment facility.
The effluent from the final clarifier was fed to the grass tanks and was
the sole source of plant nutrients. The treatment facility has a design
capacity of 3.0 MGD and a daily average flow of 1.5 MGD. However, at peak
flow periods during the day the flow capacity of the trickling filter is
often exceeded causing short—circuiting. The basic facility consists of
5
-------
bar racks and screening mechanism with conuninutor; grit chambers; primary
settling basin; trickling filter with rotary distributor; and final clari-
fier. Sludge—handling equipment includes digesters and open sludge drying
beds. Stormwater flow often causes dilution and wide variation in the
chemical quality of the effluent. The effluent discharges to an inter-
mittent stream of the Sandy Creek watershed, flowing eventually into the
South Canadian River to the north.
6
-------
d Supply
Tank
Constant Head
Overflow Line
Manifold Line
Valve & Orifice
Pump
Grass Cover In Tank
Outflow
Jj/ ‘5Standpipe
L4
- -- Pea Gravel — — U
FIGURE I - SCHEMATIC OF ELEVATION PLAN
-------
I 1l
Control Fast I
Flow Rote
rs3 2
Control Slow I
-n
Flow Rate
I- ____________________________
0
Fescue - Fast
Flow Rate I
C., ___________________________
1
Fescue- Slow _____
0
Flow Rate
,‘ ___________________________
I
rn -,
a 5 Ryegrass- Fast I
— Flow Rate
-i
Z 6 Ryegrass Slow
Flow Rate
-------
lnf low
Stainless Steel
Needle Valve _l/2H
Rubber Stopper
Plastic Pipe 1 ,
Coupling -2 l.D.
Eyedropper
Orifice
Rubber Stopper
Rubber
Hose - /2”
Outflow To Tank
FIGURE 3- SEDIMENT CHAMBER AND ORIFICE ARRANGEMENT
FOR REGULATING FLOW TO TANKS
9
-------
EXPERIMENTAL PROCEDURES
The tanks were installed and grasses planted in August 1967. A constant
flow of effluent from the final clarifier at the Ada sewage treatment
plant was maintained through the tanks while the grasses were being es-
tablished. During this period, the sediment chamber—orifice arrangements
were Installed and adjusted for constant control of the two flow rates.
In about two months the grasses were well established and showing lush
fall growth. The sampling program was started in November 1967 and con-
tinued for 12 months.
Sampling Program
Inflow and outflow samples from all six tanks were collected on a weekly
basis. Duplicate inf low samples were taken from the manifold line early
in the week, usually Monday or Tuesday. Outflow samples from tanks 1, 3,
and 5, the fast flow rate, were collected one day later; and from tanks
2, 4, and 6, the slow flow rate, two days later. This schedule was based
on slug flow and theoretical detention times for the two flow rates. The
following chemical analyses were performed on each sample collected: pH,
electrical conductivity, chloride, sulfate, biochemical oxygen demand (BOD),
chemical oxygen demand (COD), total organic (nonvolatile) carbon, total
and ortho phosphate, organic nitrogen, ammonia, nitrite, and nitrate
nitrogen. In addition to these weekly determinations, the metallic cations,
calcium, magnesium potassium, and sodium were determined on one set of
samples each month. Chemical analytical methods were in accordance with
APHA Standard Methods (7), ASTM Manual (8), and FWPCA Interim Methods (9).
Microbiological samples were collected periodically during the study to
determine bacterial counts of both the inflow and outflow water. Counts
of total coliform, fecal coliform, and fecal streptococci organisms were
performed by the standard plate count method (7) employing the membrane
filter technique.
Grass Samples
Harvest samples were collected and yields were calculated on each of the
grass tanks several times during the year. Fall growth was harvested
early in January 1968. The second harvest was made in late March and sub-
sequent harvests were made monthly until hot summer weather slowed the
growth rate. A total of eight harvests were made during the 12—month
study period. Dried grass samples were retained for chemical analysis at
the conclusion of the study. Crude protein, total nitrogen, phosphorus,
potassium, calcium, and sodium were determined on each of the grass samples.
11
-------
RESULTS AND DISCUSSION
Effluent from the final clarifier was supplied to the hydroponic tanks.
In order to evaluate the nutrient removal efficiencies of the hydroponic
tanks, the chemical quality of the sewage treatment plant effluent sup-
plied was determined at each weekly sampling period throughout the year.
The quality characteristics of the effluent are summarized in Table 1 for
the four seasons. The values reported are minimum, maximum, and mean for
each parameter along with the annual weighted mean values. The organic
and nutrient values appeared to be highest during the winter season and
lowest during the summer. Inorganic parameters did not exhibit consistent
seasonal variations. The principal factor affecting mineral salt concen-
tration was dilution by rainfall and snow.
At the time the tanks were installed and initially filled with a 15—inch
bed of pea gravel, the liquid capacity of the gravel bed was carefully
determined. Each of the six tanks held 220 5 gallons when filled to the
top of the gravel bed. This revealed an average value for the pore space
of the gravel bed to be 41.3 percent of the total gravel volume. Flow
rates were calculated on this basis in order to maintain theoretical de-
tention times of one and two days. The fast flow rate (one—day detention)
was calculated to be 578 mi/mm while the slow flow rate (two—day detention)
was 289 mi/mm. Flow rates were measured daily during the 3—day sampling
interval of each week. Some clogging of the flow orifice (Figure 3) oc-
curred, and it became necessary to clean them on a regular weekly basis to
ensure as nearly constant flows as possible. By this procedure, flow rates
and detention times were maintained within a range of ±10 percent.
The major fertilizer nutrients, nitrogen, phosphorus, and potassium, were
present in the sewage effluent at an approximate ratio of two parts N, one
part P, and one part K. Converting these to the standard fertilizer des-
ignations gave approximate ratio values of three parts N, four parts P 2 0 5 ,
and two parts K 2 0. The rate at which these were supplied to the hydroponic
tanks was easily estimated from the flow rates.
Of the sixteen elements essential for plant nutrition, six are required in
relatively large quantities and are referred to as the macronutrients (10).
These are nitrogen, phosphorus, and potassium, the major fertilizer elements;
and calcium, magnesium and sulfur. Using the annual weighted mean values
of Table 1, estimates of the rate at which these six nutrients were supplied
to the hydroponic tanks are shown in Table 2. Based on the tank area (57.3
sq. ft.) the values of Table 2 can be converted to pounds per acre per week
by multiplying by 760. No attempt was made In this study to analyze for the
micronutrients present in the sewage effluent supplied to the hydroponic
tanks. The essential micronutrients used by higher plants are: iron,
manganese, copper, zinc, boron, molybdenum, and chlorine (10).
13
-------
TABLE 1
CHARACTERISTICS OF SEWAGE TREATMENT PLANT EFFLUENT SUPPLIED TO ThE HYDROPONIC TANKS
Parameter
Winter
Spring
Summer
Fall
Annual
Weighted
Mean
Mm.
Max.
Mean
Mm.
Max.
Mean
Mm.
Max.
Mean
Mi
Max.
Mean
pH, units (range)
6.9
7.7
—
7.1
77
—
7.2
7.7
—
7.3
7.7
—
6.9—7.7
Electrical Conductivity (EC), micromhos/cm
825
1435
1065
985
1250
1099
925
1100
1007
950
1150
1025
1045
Chemical Oxygen Demand (COD), mg/i
49
172
89.9
60
101
73.6
42
70
55.8
43
75
56.9
Biochemical Oxygen Demand (BOD) , mg/i
32.2
69
47.6
16.4
65
28.3
18.4
49
31.1
17
43
28.7
31.8
Total Organic Carbon (TOC), mg/i
20
44.5
29.3
17.5
34.5
22.7
13.3
24
18.8
17.3
28.1
21.3
22.7
Total Phosphorus (T—P), mg/l as P
4.6
12.6
9.5
6.0
9.6
7.6
5.0
9.4
7.3
7.7
10.0
8.6
8.1
Organic Nitrogen (Org—N), mg/i as N
1.2
6.0
3.0
1.0
3.0
2.1
1.6
5.3
2.8
1.8
4.6
2.8
2.6
Ammonia Nitrogen (NR 3 —N), mg/i as N
2.2
11.7
6.7
1.0
9.1
4.9
1.2
7.5
2.8
1.9
5.2
3.4
4.4
Nitrite Nitrogen (NO 2 —N), mg/i as N
0.2
0.8
0.4
0.2
0.6
0.35
0.1
3.8
0.6
0.1
0.5
0.34
0.43
Nitrate Nitrogen (NO 3 —N), mg/i as N
5.6
10.7
8.4
1.6
10.9
7.7
4.5
10.4
7.6
5.4
12.3
9.1
Total Nitrogen (T—N), mg/i as N
14.4
22.2
18.3
10.1
19.7
15.1
9.9
19.8
13.5
10.7
20.7
15.6
15.4
Chloride (Cl), mg/i
77
106
96.6
70
121
95.5
88
96
92
82
140
98.5
Sulfate (SOt .), mg/i
40
65
51.7
53
60
57
30
69
50
33
45
41
96.8
47.3
Calcium (Ca), mg/i
69
90
78.3
90
95
92.5
70
93
81
70
81
75
Magnesium (Mg), mg/i
28
39
35
30
36
33
27
36
31.5
35
36
35.7
34.3
Potassium (K), mg/i
8.0
10.5
9.3
7.4
8.0
7.7
6.4
9.0
7.7
6.8
8.9
7.9
Sodium (Na), mg/i
73
126
96.8
68
95
81.5
85
100
92.5
79
118
92.3
8.5
92.7
Note: (1) Milligrams per liter multiplied by 8.34 x iO ” equals pounds per 1000 gallons.
(2) City water supply has electrical conductivity averaging about 625 micromhos/cm and a median pH of 7.5.
-------
TABLE 2
ESTIMATED AVERAGE RATES OF SUPPLY OF MACRONUTRIENTS
TO THE HYDROPONIC TANKS
Macronutrient
Fast F
lb/week
low Rate
(lb/ac/wk)
Slow Flow Rate
lb/week (lb/ac/wk)
Nitrogen (as N)
0.19
(144.4)
0.095 (72.2)
Phosphorus (as P)
(as P 2 0 5 )
0.10
0.23
(76.0)
(174.8)
0.05 (38.0)
0.115 (87.4)
Potassium (as K)
(as 1(20)
0.106
0.13
(80.56)
(98.8)
0.053 (40.28)
0.065 (49.4)
Calcium (as Ca)
1.00
(760.0)
0.50 (380.0)
Magnesium (as Mg)
0.44
(334.4)
0.22 (167.2)
Sulfur (as SO 4 )
(as 5)
0.59
0.195
(448.4)
(148.20)
0.295 (224.20)
0.097 (74.1)
Note: To convert pounds per week to pounds per acre per week, multiply by 760.
Characteristics of Effluent from Hydroponic Tanks
The quality of the effluent from each of the six hydroponic tanks has been
summarized in Tables 3, 4, and 5. The seasonal values are means of the weekly
data and the annual weighted mean is for the full year of the study. Comparison
of the annual mean values with those of Table 1 shows that the tanks were not
effective in reducing any of the inorganic parameters measured. Measurements
of flow rates both in and out of the tanks were made periodically. Flow re-
ductions due to evapotranspiration were found to range from not detectable to
a maximum of about five percent and estimated to average about 3 percent. It
was apparent that evaporation and transpiration from the tanks were insigni-
ficant compared to the volume flowing through, and no measurable increased
concentration of salts was detectable. The small effect of evaporative proc-
esses could not be consistently detected even on individual sets of the weekly
data.
Treatment Efficiencies
The greatest effect of passage through the hydroponic tanks was the re-
duction of the organic parameters and total nitrogen. All forms of ni—
trogen measured were appreciably reduced, but total nitrogen was taken
as the indicator of nitrogen removal efficiencies. The mean removal
15
-------
TABLE 3
MEAN CHARACTERISTICS OF EFFLUENT FROM THE CONTROL TANKS
Parameter
Tank 1. Fast Flow Rate
Tank 2. Slow Flow Rate
Winter
Spring
Summer
Fall
Annual.
Weighted
Mean
Winter
Spring
Sumriier
Fall
Annual
Weighted
Mean
pH, units (range) 6.7—7.5 6.5—7.6 7.1—7.6 7.2—7.6 6.5—7.6 6.7—7.8 7.2—7.6 7.1—7.8 7.2—7.6 6.7—7.8
Electrical Conductivity (EC), microtnhos/cm 998 1077 997 1046 1023 1043 1066 945 1010 1010
Chemical Oxygen Demand (COD), mg/i 34.2 34.0 33.0 25.8 31.6 37.8 41.6 34.1 27.2 34.9
Biochemical Oxygen Demand (BOD), mg/l 6.0 3.4 2.8 4.6 3.9 5.2 5.8 3.1 3.6 4.3
Total Organic Carbon (TOC), mg/l 13.4 13.4 11.1 10.4 11.9 12.7 13.7 10.5 10.9 11.9
Total Phosphorus (T—P), mg/l as P 8.8 7.4 7.3 8.6 8.1 8.7 7.8 6.2 8.0 7.7
Organic Nitrogen (Org—N), mg/i as N 1.2 1.0 1.3 1.1 1.2 1.2 1.3 0.8 0.9 1.0
Ammonia Nitrogen (NH 3 —N), mg/i as N 5.3 3.9 0.9 1.4 2.9 5.4 3.8 0.4 0.8 2.6
Nitrite Nitrogen (N0 2 —N), mg/i as N <0.1 <0.1 <0.05 <0.05 <0.1 <0.1 <0.05 <0.1 <0.05 <0.1
Nitrate Nitrogen (N0 3 —N), mg/i as N 3.7 3.4 1.3 3.4 2.9 3.1 2.1 0.6 3.3 2.3
Total Nitrogen (T—N), mg/i as N 10.1 8.4 3.5 5.8 7.0 9.7 7.2 1.7 5.0 5.9
Chloride (Cl), mg/i 74 109 97 114 99 98 101 88 108 98
Sulfate (SOk), mg/i 48 67 43 40 45 50 64 45 41 46
Calcium (Ca), mg/i 74 97 78 76 79 71 98 82 72 79
Magnesium (Mg), mg/i 33 29 28 38 32 33 30 30 35 32
Potassium (K), mg/i 8.7 6.9 7.6 8.3 8.1 8.0 7.0 7.6 6.8 7.4
Sodium (Na), mg/i 92 92 94 03 95 91 77 94 98 91
Note: Milligrams per liter multiplied by 8.34 x lO equals pounds per 1000 gallons.
-------
TABLE 4
MEAN CHANACTERISTICS OF EFFLUENT FROM THE FESCUE TANKS
Parameter
Tank 3. Fast Flow Rate
Tank 4. Slow Flow Rate
Winter
Spring
Summer
Fall
Annual
Weighted
Mean
Winter
Spring
Summer
Fall
Annual
Weighted
Mean
pH, units (range)
Electrical Conductivity (EC), micromhos/cm
Chemical Oxygen Demand (COD), mg/i
Biochemical Oxygen Demand (BOO), mg/i
Total Organic Carbon (TOC), mg/i
Total Phosphorus (T—P), mg/i as P
Organic Nitrogen (Org—N), mg/i as N
Ammonia Nitrogen (N11 3 —N), mg/i as N
Nitrite Nitrogen (N0 2 —N), mg/i as N
Nitrate Nitrogen (N0 3 —N), mg/i as N
Total Nitrogen (T—N), mg/i as N
Chloride (Ci), mg/i
Sulfate (SOk), mg/i
Calcium (Ca), mg/i
Magnesium (Mg), mg/i
Potassium (K), mg/i
Sodium (Na), mg/i
6.7—7.5
1020
29.9
4.2
12.6
8.3
0.9
3.7
<0.1
2.9
7.3
88
52
79
35
6.1
91
6.5—7.5
1084
31.5
3.9
16.0
7.5
1.2
3.2
<0.1
1.2
5.7
92
64
98
30
6.3
82
7.0—7.5
1016
32.7
4.2
12.2
7.2
i.O
0.7
<0.05
0.3
2.0
97
41
84
30
6.6
96
7.0—7.4
1097
26.8
4.6
10.1
8.3
0.8
0.9
<0.05
1.0
2.7
120
42
77
37
6.5
105
6.5—7.5
1047
30.1
4.3
12.4
7.8
1.0
2.1
<0.1
1.4
4.5
102
45
82
34
6.3
94
6.6—7.8
999
30.1
2.6
11.5
7.9
1.2
2.6
<0.1
2.4
6.1
91
45
78
33
5.0
91
6.9—7.4
1117
32.3
4.7
15.8
7.2
1.4
2.4
-------
TABLE 5
MEAN CHARACTERISTICS OF EFFLUENT FROM THE PERENNIAL RYECRASS TANKS
Parameter
Tank 5. Fast Flow Rate
Tank 6. Slow Flow Rate
Winter
Spring
Summer
Fall
Annual
Weighted
Mean
Winter
Spring
Summer
Fall
Annual
Weighted
Mean
pH, units (range)
Electrical Conductivity (BC), micromhoe/cm
Chemical Oxygen Demand (COD), mg/i
Biochemical Oxygen Demand (BOD), mg/l
Total Organic Carbon (TOC), mg/i
Total Phosphorus (T—P), mg/i as P
Organic Nitrogen (Org—N), mg/i as N
Ammonia Nitrogen (NH 3 —N), mg/i as N
Nitrite Nitrogen (NO 2 —N), mg/i as N
Nitrate Nitrogen (NO 3 —N), mg/i as N
Total Nitrogen (T—N), mg/i as N
chloride (Cl), mg/i
Sulfate (SO 4 ), mg/i
Calcium (Ca), mg/i
Magnesium (Mg), mg/i
Potassium (K), mg/i
Sodium (Na). ma/l
6.7—7.5
1015
31.8
4.1
14.2
8.2
1.4
3.8
<0.1
3.2
8.3
80
49
77
34
7.6
85
6.5—7.8
1096
32.6
3.9
15.2
7.7
1.2
2.9
<0.1
1.3
5.4
104
69
96
29
6.8
96
7.1—7.6
1032
33.1
2.5
11.2
7.4
1.1
0.6
<0.05
0.8
2.5
106
43
81
29
7.6
100
7.2—7.5
1103
29.5
4.8
10.9
8.5
1.2
1.1
<0.05
1.7
3.9
128
43
77
38
8.3
111
6.5—1.8
1055
31.7
3.8
12.7
8.0
1.2
2.1
<0.1
1.8
5.1
107
46
80
33
7.6
95
6.7—7.9
999
29.2
3.5
11.1
7.7
1.0
2.7
<0.1
2.4
6.1
81
44
79
33
1.0
84
6.9—7.6
1117
32.6
4.6
13.8
6.9
1.5
1.9
<0.05
0.6
4.1
104
64
100
29
6.0
87
7.2—7.7
967
33.3
4.0
12.9
6.4
1.0
0.5
<0.05
0.1
1.6
91
40
83
30
7.4
96
7.1—7.7
1042
25.5
4.0
10.7
8.2
0.7
0.6
<0.05
1.4
2.7
110
44
74
39
6.9
99
6.7—7.9
1018
30.0
4.0
11.9
7.3
1.0
2.7
<0.1
1.2
3.7
96
44
82
33
6.9
90
I-
Note: Milligrams per liter multiplied by 8.34 x equals pounds per 1000 gallons.
-------
TABLE 6
MEAN REMOVAL PERCENTAGES FOR TOTAL NITROGEN, COD, AND BOD
(CONCENTRATION BASIS)
Season
Control
Fescue
Ryegrass
Fast
RAte
Slow
Rate
Fast Slow
Rate Rate
Fast
Rate
Slow
Rate
Total
Nitrogen
Winter
47.9
52.3
65.5 72.4
58.5
72.4
Spring
34.0
42.6
50.6 65.2
53.7
64.1
Summer
72.8
85.1
82.8 84.6
78.2
89.2
Fall
64.4
69.6
83.0 89.5
75.5
83.2
COD
Winter
62.8
57.6
64.4 63.9
61.4
67.1
Spring
Summer
41.7
43.8
39.2
39.1
47.5 51.0
42.7 38.1
47.7
43.3
48.4
41.6
Fall
53.1
52.4
51.4 53.5
45.7
53.3
BOD
Winter
84.4
————
88.4 91.0
89.7
90.5
Spring
Summer
84.2
89.1
77.4
87.4
86.1 84.7
85.5 85.7
84.3
90.3
83.8
83.8
Fall
84.3
87.1
82.1 83.4
83.8
85.9
percentages for total nitrogen (T—N), chemical oxygen demand (COD), and
biochemical oxygen demand (BOD) are summarized in Table 6. The values
reported are seasonal means for each of the six tanks. These data show
that the control tanks were, in fact, quite effective in reducing the
organic and nitrogen content of the sewage effluent supplied. Statistical
techniques were employed to evaluate the effects of the grasses, flow rates,
and seasons on treatment efficiencies.
Nitrogen removal . Factorial analysis of variance was used to determine the
significance of grass species, flow rate, and season on total nitrogen re-
duction. Both fescue and ryegrass were significantly better than the con-
trol at the 5 percent level of confidence. There was no significant
19
-------
difference in the nitrogen removal by the fescue and ryegrass tanks.
Nitrogen removal in the slow flow rate tanks was significantly better
than in the fast flow rate tanks. Summer performance was significantly
better than winter and spring but not different from fall at the 5 per-
cent confidence level. Likewise, fall performance was better than spring
and winter. There was no significant difference between winter and spring
in nitrogen removal. In summary, it was shown that grasses, flow rate,
and seasons all had significant effects on total nitrogen reduction in
the hydroponic tanks.
cni rpinnw 1 . The greatest reduction of COD in the hydroponic tanks
occurred during the winter seasons, followed by fall, spring, and summer
in descending order. Factorial analysis of variance revealed winter re-
moval of COD to be significantly better than spring and summer at the 5
percent confidence level, but not significantly greater than fall removals.
There was no significant difference between the performance of spring and
summer, spring and fall, nor between summer and fall. The analysis also
showed no significant effect could be attributed to flow rate nor grasses.
The latter indicated no significant differences between the control tanks
and grass tanks in COD reduction.
BOD removal . The data of Table 6 show that there was no consistent and
predictable trend associated with ROD removals. Factorial analysis of
variance of the data verified this obvious conclusion. It did, in fact,
show that there was no significant change in BOD removal due to grasses
(versus controls), flow rate of effluent, nor season of operation. Of
all the parameters examined, BOD removal was the most consIstent. Simply
flowing the effluent through the flooded gravel bed was quite effective in
reducing the ROD content of -the sewage effluent.
Total phosphorus . One of the principal objectives of this study was to
evaluate the removal efficIencies of the hydroponic tanks for the nutrient
compounds of nitrogen and phosphorus. As previously shown, significant
nitrogen removals were obtained. This was not the case for the total
phosphorus content of the sewage effluent. Reductions were so small and
erratic that percent removals were not used to evaluate the data. In many
cases, individual weekly samples showed equal or slightly greater phos-
phorus content than that of the effluent supplied to the tanks. This also
is evident in some of the mean seasonal data summarized in Table 7.
The concentration data were analyzed using a factorial analysis of variance.
The analysis showed no significant change in phosphorus concentration due
to gr ss versus control. The most consistent effect was noted between
flow rates. Concentrations from the slow flow rate tanks were significantly
lower at the 5 percent confidence level than those from the fast flow
tanks. Likewise, significant seasonal effects were noted when comparisons
between inflow and outflow were made. The overall mean values of Table 7
show the rank of the seasons and how these compared with the mean values
for the sewage effluent supplied to the hydroponic tanks. The data in-
dicate that both biological fixation and release occurred in the tanks.
The only phosphorus actually removed from the system was that which was
utilized by the grasses. Apparently, this was a small amount in com-
parison to the quantity supplied to the tanks.
20
-------
TABLE 7
MEAN TOTAL PHOSPHORUS CONCENTRATIONS IN SEWAGE EFFLUENT
AND EFFLUENTS FROM THE HYDROPONIC TANKS, mg/i
Source
Winter
Spring
Summer
Fall
Annual
Sewage
Effluent
9.5
7.6
7.3
8.6
8.1
1. Control
Fast Flow Rate
8.8
7.4
7.3
8.6
8.1
2. Control
Slow Flow Rate
8.7
7.8
6.2
8.0
7.7
3. Fescue
Fast Flow Rate
8.3
7.5
7.2
8.3
7.8
4. Fescue
Slow Flow Rate
7.9
7.2
6.8
7.7
7.4
5. Ryegrass
Fast Flow Rate
8.2
7.7
7.4
8.5
8.0
6. Ryegrass
Slow Flow Rate
7.7
6.9
6.4
8.2
7.3
Overall Mean
(outflow)
8.4
7.4
6.7
8.2
Grass Yields and Quality
The grasses were harvested eight times during the year. Moisture content
and dry weight yield data are summarized in Table 8. For easy conversion
to pounds or tons per acre, yield samples were cut from an area of 9.6
square feet. The yield in grams from that area multiplied by 10 gave
pounds per acre. Moisture content was determined by drying the freshly
cut samples in a 65°C oven. Samples of the dried grasses were retained
for chemtcal analysis.
Considerable differences in yield were noted between the inlet and outlet
ends of the tanks at some of the harvests. The values reported in Table 8
are averages for the entire tank length. In some cases, as many as four
yield samples were taken from each tank. Never were less than two samples
taken to represent the harvest from each tank. This difference in yield
21
-------
TABLE 8
MOISTURE CONTENT AND DRY WEIGHT YIELD OF THE GRASSES
Harvest
Date
Kentucky Fescue
Perennial Ryegrass
Fast Flow Rate Slow Flow Rate
Fast Flow Rate Slow Flow Rate
Moisture Dry Weight Moisture Dry Weight
Content Yield Content Yield
Moisture Dry Weight Moisture Dry Weight
Content Yield Content Yield
percent tons/acre percent tons/acre
percent tons/acre percent tons/acre
1/5/68
3/29/68
4/30/68
6/3/68
7/1/68
8/7/68
9/27/68
11/19/68
78.2 1.69 76.1 1.70
83.8 1.32 84.9 1.78
84.7 1.87 85.3 2.40
82.4 1.39 81.4 1.16
80.8 0.79 77.7 0.52
77.2 0.91 74.6 0.70
81.6 1.04 79.4 0.70
79.8 1.58 76.8 1.09
79.3 0.90 76.9 0.55
86.5 1.52 86.2 1.24
85.8 2.62 85.0 2.67
81.2 1.06 78.2 0.70
73.2 0.31 72.7 0.17
————
80.8 0.17 76.9 0.15
74.4 0.59 72.8 0.45
Total
Yield
———— 10.59 ———— 10.05
7.17 ———— 5.93
-------
could only be attributed to the reduction of nitrogen as the effluent
flowed slowly through the tanks. This observation raises the question
of optimum tank length which was not answered by this study.
The effect of seasons on yield is evident from the data of Table 8. The
greatest 30—day yields were obtained during the month of April. The
summer months of July and August produced the least yields. After pro-
ducing seed, the ryegrass remained practically dormant until new fall
growth began. The fescue continued to grow during the summer but at a
greatly reduced rate. Other grasses should be studied to find those which
give the best growth during all seasons. Reed canarygrass is one that
looks promising and should be grown under hydroponic conditions for
evaluation.
The dried grass samples were analyzed for crude protein, phosphorus,
potassium, calcium, and sodium (Table 9). Nitrogen content was obtained
by a conversion factor from crude protein content. Nitrogen in the nitrate
form was not detectable in any of the samples. This analysis should prob-
ably have been attempted on fresh plant material rather than the dried
samples. Crude protein, nitrogen, and phosphorus content all varied ac-
cording to stage of plant growth. The most active growing periods gave
the highest values for each of these. Potassium did not show as much
variation in the fescue as it did in the ryegrass. Nitrogen content was
consistently 3 to 4 times that of phosphorus in all samples. The crude
protein content during the most active growing periods was sufficiently
high to class these grasses as having high nutritive value.’ The slow
growth period of mid—summer produced the lowest grass quality. With each
grass, this occurred after seed production and before fall growth started.
Nutrients Removed by Harvest
From yield and quality data (Tables 8 and 9) the total nutrients (N, F,
and K) removed by grass harvests are shown in Table 10. These values are
reported in pounds per acre. (To convert to pounds removed by each hydro-
ponic tank, divide by 760.) In order to establish a nutrient balance for
the hydroponic system, estimates of the major nutrients leaving the hydro-
ponic tanks in the outflow were made from the annual mean-values of Tables
3, 4, and 5. An allowance for an average of 3 percent reduction in flow
due to evapotranspiration was made and the results are shown in Table 11.
The period covered by grass harvests was about 60 weeks.
Combining data from Tables 2, 10, and 11, a balance of nutrients, N, P,
and K, was made. The fate of nutrients accounted for during the 60—week
period is shown in Table 12. These data indicate the difficulty encountered
in attempting a nitrogen balance. Much of the nitrogen that was unaccounted
for probably was lost through the processes of denitrification which
-Stidham, Neal,. Area Range Specialist, Soil Conservation Service,
Ada, Oklahoma, personal communication, February 1969.
23
-------
TABLE 9
*
ANALYSES OF GRASS SA LES
Percent, dry
weight basis
Crude
Harvest Date
Protein
Nitrogen
Phosphorus
Potassium
Calcium
Sodium
Fescue:
1/5/68
17.1
1.71
0.57
2.8
0.13
0.19
3/29/68
21.0
2.10
0.56
2.9
0.12
0.19
4/30/68
19.0
1.90
0.56
2.6
0.12
0.29
6/3/68
13.0
1.30
0.40
2.4
0.11
0.11
7/1/68
11.0
1.10
0.39
2.3
0.17
0.19
8/7/68
10.1
1.01
0.37
2.0
0.12
0.17
9/27/68
13.3
1.33
0.41
2.3
0.11
0.16
11/19/68
14.3
1.43
0.48
2.8
0.12
0.19
Ryegra8s:
1/5/68
23.3
2.33
0.54
2.1
0.12
0.45
3/29/68
20.2
2.02
0.57
3.0
0.14
0.49
4/30/68
13.2
1.32
0.46
1.8
0.12
0.72
6/3/68
8.3
0.83
0.38
1.7
0.15
0.41
7/1/68
12.5
1.25
0.39
0.93
0.20
0.52
9/27/68
14.3
1.43
0.37
1.2
0.13
0.32
11/19/68
15.6
1.56
0.46
1.9
0.12
0.32
*
The grass analyses were performed by The Samuel Roberts Noble Foundation, Inc.,
Ardniore, Oklahoma.
-------
TABLE 10
PLANT NUTRIENTS REMOVED BY THE GRASS HARVESTS, POUNDS PER ACRE
Harvest Date
Fast Flow Rate Slow Flow Rate
Nitrogen Phosphorus Potassium Nitrogen Phosphorus J’otassium
(N) (P) (K) (N) (P) (K)
Fescue:
1/5/68
3/27/68
4/30/68
6/3/68
7/1/68
8/7/68
9/27/68
11/19/68
57.78 19.26 94.64 58.14 19.38 95.20
55.44 14.78 76.56 74.76 19.92 103.24
71.06 20.94 97.24 91.20 26.88 124.80
36.14 11.12 66.72 30.16 9.28 55.68
17.38 6.16 36.34 11.44 4.04 23.92
18.38 6.72 36.40 14.14 5.18 28.00
27.66 8.52 47.84 18.62 5.74 32.20
45.18 15.16 88.48 31.16 10.46 61.04
Total
329.02 102.66 544.22 329.62 100.88 524.08.
Ryegras s:
1/5/68
3/29/68
4/30/68
6/3/68
.7/1/68
9/27/68
11/19/68
41.94 9.72 37.80 25.62 5.94 23.10
61.40 17.32 91.20 50.08 14.12 74.40
69.16 24.10 94.32 70.48 24.56 96.12
17.58 8.04 36.04 11.62 5.32 23.80
7.74 2.40 5.76 4.24 1.32 3.16
4.86 1.24 4.08 4.28 1.10 3.60
18.40 5.42 22.42 14.04 4.14 17.10
Total
221.08 68.24 291.62 180.36 56.50 241.28
Note: To convert these values to pounds removed per hydroponic tank, divide by 760.
-------
TABLE 11
PLANT NUTRIENTS LEAVING THE HYDROPONIC TANKS, POUNDS PER ACRE
(BASED ON OUTFLOW FOR 60 WEEKS )
Nutrient
Fast Flow
Rate
Slow Flow
Rate
lb/acfwk
lb/acre
lb/ac/wk
lb/acre
Control:
Nitrogen
Phosphorus
Potassium
(N)
(P)
(K)
63.84
73.72
73.72
3830.4
4423.2
4423.2
26.60
34.96
32.68
1596.0
2097.6
1960.8
‘Fescue:
Nitrogen
Phosphorus
Potassium
(N)
(P)
(K)
31.92
66.88
50.16
1915.2
4012.8
3009.6
15.20
32.68
25.08
912.0
1960.8
1504.8
Ryegrass:
Nitrogen
Phosphorus
Potassium
(N)
(P)
(K)
32.68
66.12
62.32
1960.8
3967.2
3739.2
15.96
32.68
31.16
957.6
1960.8
1869.6
Note: Pounds per acre divided by 760 equals pounds per hydroponic tank.
occurred in the anaerobic environment of the flooded gravel bed. The
oxygen—demanding nature of the waste must have kept the oxygen depleted
at all except a shallow surface layer, thus promoting denitrification
throughout most of the tank volume. Upon examination of the root systems
developed by the grasses, it was found that most of the roots were within
the top six to eight inches of the gravel bed. This further indicated
the anaerobic conditions which existed within the lower portion of the
gravel bed.
The data of Table 12 reveal that greater percentages of the nutrients
supplied were removed by grass harvests from the slow flow rate tanks of
both grasses. This is in contrast to the yield data of Table 8, which
shows that the fast flow rate tanks gave the higher yield with both grasses.
The higher yield of fescue over ryegrass is also reflected in the greater
percentages of nutrients removed by harvest.
The data of Table 12 show that the grasses were, in fact, minor contributors
to nutrient removal from the waste when compared to the total quantity of
nutrients that passed through the tanks and returned to the stream flow.
Much of the phosphorus and potassium unaccounted for was no doubt retained
in the root systems and aerial growth that was not harvested. In addition,
considerable biological slime growth developed within the gravel bed.
26
-------
TABLE 12
FATE OF NITROGEN, PHOSPHORUS, AND POTASSIUM SUPPLIED TO THE HYDROPONIC TANKS DURING 60—WEEK PERIOD
Nutrient
Fast Flow Rate
Slow Flow Rate
Supplied
to
tank
Leaving
tank in
outflow
Removed by
grass harvest
Unaccounted
for
Supplied
to
tank
Leaving
tank in
outflow
Removed by
grass harvest
Unaccounted
for
pounds
pounds
percent
pounds
percent
pounds
percent
pounds
pounds
percent
pounds
percent
pounds
percent
Control:
Nitrogen (N)
Phosphorus (P)
Potassium (K)
Fescue
Nitrogen (N)
Phosphorus (P)
Potassium (K)
Ryegrass:
Nitrogen (N)
Phosphorus (P)
Potassium (K)
11.40
6.00
6.36
11.40
6.00
6.36
11.40
6.00
6.36
5.04
5.82
5.82
2.52
5.28
3.96
2.58
5.22
4.92
44.2
97.0
91.5
22.1
87.9
62.2
22.6
87.0
77.4
————
————
————
0.43
0.14
0.72
0.29
0.09
0.38
————
————
————
3.8
2.3
11.3
2.6
1.5
6.0
6.36
0.18
0.54
8.45
0.58
1.68
8.53
0.69
1.06
55.8
3.0
8.5
74.1
9.8
26.5
74.8
11.5
16.6
5.70
3.00
3.18
5.70
3.00
3.18
5.70
3.00
3.18
2.10
2.76
2.58
1.20
2.58
1.98
1.26
2.58
2.46
42.8
92.0
81.1
21.1
86.0
62.2
22.1
86.0
77.3
————
————
————
0.43
0.15
0.69
0.24
0.07
0.32
————
————
————
7.6
4.4
21.7
4.2
2.5
10.0
3.60
0.24
0.60
4.07
0.29
0.51
4.20
0.35
0.40
63.2
8.0
18.9
71.3
9.6
16.1
73.7
11.5
12.7
N.)
-J
Note: Pounds per hydroponic tank multiplied by 760 equals pounds per acre.
-------
Nutrients thus retained are not accounted for in the data f Table 12.
This black slime growth developed around the inlet manifold to the extent
that flow was retarded at times during tFie latter months of the study.
Periodically draining the tanks to permit aeratIon assisted in restoring
normal flow through the tanks.
Bacterial Reductions
Bacterial counts were made periodically on both. inflow and outflow samples
during the study. Total counts were expressed as number of organisms per
100 ml sample. The data are summarized in Table 13. The values are means
for monthly periods and the percent reductions are means for all six tanks.
Individual data showed no consistent differences in the bacterial reductions
of any of the tanks. The only apparent effect was one of seasons as shown
in Table 13. The bacterial reductions during February were somewhat less
than those of the other months reported. No real significance can be at-
tached to this variability, since all seasons of the year were not repre-
sented in the data.
28
-------
TABLE 13
MEAN BACTERIAL COUNTS AND PERCENT REDUCTIONS AFTER FLOW THROUGH THE HYDROPONIC TANKS
Date
Total
Coliforms
(Inf low)
Percent
Reduction
Fecal
Coliforms
(Inf low)
Percent
Reduction
Fecal
Streptococci
(Inf low)
Percent
Reduction
September
1967
1.4 x 106
99.0
2.4 x lO
99.6
2.1 x lO’
99.4
November
1967
2.4 x 106
96.8
3.6 x 10
94.5
6.7 x lO
94.5
February
1968
1.9 x 106
81.9
6.2 x iO
88.0
1.7 x lO
88.3
April 1968
1.2 x 106
94.8
3.1 x 10
95.9
4.6 x lO
95.7
Note: (1) Total bacterial counts are expressed as number of organisms per 100 ml.
(2) Percent reductions are mean values for all six tanks.
-------
ACKNOWLEDGNENT
The author wishes to express appreciation for the generous cooperation
of The Samuel Roberts Noble Foundation, Ardmore, Oklahoma, for conducting
the chemical analyses of the grass samples. Also, to the city of Ada,
Oklahoma, for granting permission to conduct the study on city property
located at the Ada Sewage Treatment Plant.
Special recognition is due to Dr. W. R. Duffer, who assisted in the plan
ning of the study, and to Mr. R. E. Thomas who gave freely of his time and
talent and assisted with several phases of the actual conduct of the study.
31
-------
REFERENCES
1. Queal, Cal, “Hydroponics——tomorrow’s farms?” Empire Magazine , The
DenvérPôst . (October15, 1967).
2. U. S. Department of the Interior, Guidelines for Establishing Water
Quality Standards for Interstate Waters . Federal Water Pollution
Control Admin., Washington, D. C. (1966).
3. Hirsch, Allan, Agee, James L., and Burd, Robert S., “Water quality
standards: the Federal perspective——progress toward objectives.”
JoUrnal Water Pollution Control Federation, Vol. 40, No. 9, pp. 1601—
1606. (September 1968).
4. Weinberger, Leon W., Stephan, David G., and Middleton, Francis N.,
“Solving our water problems——water renovation and reuse.” Annals of
the New York Academy of Sciences, Vol. 136, pp. 131—154. (July 1966).
5. Martin, Edward J., and Weinberger, Leon W., “Eutrophication and water
pollution.” University of Michigan, Great Lakes Research Division,
Pub. No. 15, pp. 451—469. (1966).
6. Eby, Harry J., “Evaluating adaptability of pasture grasses to hydroponic
culture and their ability to act as chemical filters.” American Society
of Agricultural Engineering, St. Joseph, Michigan, In Management of Farm
Animal Wastes, ASAE Pub. No. SP—0366, pp. 117—120. (1966).
7. American Public Health Association, Inc., Standard Methods for the
Examination of Water and Wastewater , Twelfth Edition, New York, New
York. (1965).
8. American Society for Testing Materials, Manual on Industrial Water and
Industrial Waste Water . ASTM Special Technical Pub. No. l48—D,
Philadelphia, Pennsylvania. (1959).
9. U. S. Department of the Interior, FWPCA Official Interim Methods for
Chemical Analysis of Surface Waters . Federal Water Pollution Control
Admin., Washington, D. C. (September 1968).
10. Buckman, H. 0., and Brady, N. C., The Nature and Properties of Soils .
Sixth Edition, New York: The Macmillan Company. (1960).
33
-------
Accession Number I 2 Subject Field & Group
05
F
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
_ 1 Organization USD1, Federal Water Quality Administration
Robert S. Kerr Water Research Center
Ada, Oklahoma
A. Title
NUTRIENT REMOVAL FROM ENRICHED WASTE EFFLUENT BY THE HYDROPONIC CULTURE
OF COOL SEASON GRASSES,
10 Author(s)
Law, James P., Jr.
Project Desidnation
1608O———l0/69
Note
22 Citation
23] Descriptors (Starred First)
Hydroponic culture,* tertiary treatment, waste water treatment
25 Identifiers (Starred First)
Nutrient removal,* cool season grasses, waste assimilative capacity
27 I Abstract
iGrasses were grown in hydroponic culture tanks to evaluate their nutrient removal
capabilities when supplied with secondary—treated sewage effluent as the sole source of
plant nutrients. Statistical methods were employed to determine the effects of the
grasses, flow rates, and seasons on nutrient removal.
Two control tanks with gravel bed and no grass were maintained throughout the study,
two were planted with tall (Ky 31) fescue, and two were planted with perennial ryegrass.
Two flow rates were maintained in each pair of tanks, approximately one—day and two—
day detention times. All six tanks were effective in reducing the oxygen—demanding
organic content of the effluent. Total nitrogen content was reduced appreciably by
the control tanks, but the grass tanks were significantly better at nitrogen removal.
Total phosphorus concentrations were reduced only slightly by passage through the
tanks. The fast flow rate tanks produced the greater grass yields, while the slow flow
rate tanks were more effective in nutrient removal from the sewage effluent.
From grass yield and analyses data, the amount of plant nutrient material removed
by the grasses was small compared to the total quantity supplied. Further studies are
suggested to determine economic factors and design parameters.
Abstractor
James P.
WR: 102 (REV.
Law,
JULY
Institution Robert S. Kerr Water Research Center, SCR
Jr.
FWQA, 11 1)T
19691
WRSI C
SE D TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. 0. C. 20540
*U. S. GOVEINMENT PRDITINO OFFICE: 1 5 (0 0 - 406-907
------- |