United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada OK 74820
*
Research and Development
EPA-600/S2-81-150 Sept. 1981
Project Summary
Nitrogen and Phosphorus
Reactions in Overland
Flow of Wastewater
R. A. Khalid, I. C. R. Holford, M. N. Mixon, and W. H. Patrick, Jr.
Biochemical transformations of
labelled ammonium-nitrogen resulting
from the overland flow treatment of
simulated wastewater were studied in
small scale test models established
with vegetated soils. The results of
overland flow experiments indicated
the existence of aerobic-anaerobic
zones in the soil mass to facilitate
nitrification-denitrification processes
and enhance nitrogen losses to the
atmosphere. The incomplete nitrifica-
tion of ammonium nitrogen in the
simulated wastewater applied to
overland flow models suggests that
nitrification reactions may be limiting
the proportion of nitrate-nitrogen
available for denitrification reactions.
The loss of applied ammonium-nitrogen
attributed to denitrification reactions
in the overland flow experiments
ranged from 3 to 35%. In the growth
chamber studies where alternate
aerobic-anaerobic conditions were
maintained with controlled soil
moisture, loss was as high as 59%. The
rate of nitrogen loss in the nitrate
treatments was about twice that in the
ammonium treatments. The plant
uptake of nitrogen in the overland flow
and growth chamber studies accounted
for 23 to 62% of applied ammonium-
nitrogen. About 5% of applied am-
monium was lost through ammonia
volatilization in the studies.
The mechanisms of phosphorus
sorption and desorption were investi-
gated, under both Jaboratory and
overland flow conditions. The results
of laboratory studies indicated that
initial flooding of aerated soil for about
three weeks was accompanied by a
large increase in phosphorus sorption
capacity and decrease in phosphorus
mobility. Longer periods of flooding,
however, caused a marked decrease in
phosphorus sorption capacity and a
corresponding increase in phosphorus
mobility and leaching losses in acid
soils. Calcium phosphate precipitation
under alkaline soil conditions increased
phosphorus sorption capacity of soils.
The results of the overland flow
experiment also demonstrated that
the efficiency of phosphorus removal
from municipal wastewater was
greatly enhanced by lime addition to
the soil compared to nonlimed flooded
soil.
This Project Summary was devel-
oped by EPA's Robert S. Kerr Environ-
mental Research Laboratory. Ada,
OK, to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
The implementation of Federal Water
Pollution Control Act Amendments of
1972 (Public Law 92-500) has been a
driving force in the development of land
application as a major management
alternative for the effective treatment of
municipal wastewater. This has resulted
in a renewed interest in studying the
various processes involved in the
-------
removal of pollutants in various land
treatment systems. Most municipal
wastewaters contain significant con-
centrations of nutrients, primarily
nitrogen and phosphorus. Both nitrogen
and phosphorus can be serious pollu-
tants if Discharged to surface waters.
Overland flow, also called grass filtra-
tion, has been shown to be an effective
land treatment system for removal of
nitrogen from municipal wastewater
while phosphorus removal efficiency
has been found to vary over a wide
range. The United States Environmental
Protection Agency (EPA) had undertaken
a comprehensive research program
dealing with loading rates for different
soil types, frequency of application for
maximum nutrient removal, mecha-
nisms involved in transformations,
retention, gnd movement in the soil, and
management practices for controlling
nitrogen and phosphorus behavior in an
overland flow treatment system. The
present investigation, funded by EPA,
was focused o.n the mechanisms of
va/ious nitrogen and phosphorus re-
moval processes in an overland flow
treatment model.
The specific objectives of this research
investigation were:
1. To investigate the distribution of
applied simulated, wastewater in the
various components of overland flow
treatment system.
2. To evaluate the rote of nitrification-
denitrification reactions, ammonia
volatilization, plant uptake and immobi-
lization in nitrogen removal in small
scale overland flow treatment systems.
3. To measure the effects of controlled
oxidation-reduction conditions and pH
changes on phosphorus sorption and
mobility in a soil suspension.
4. To determine the effects of pre-
reduced soil conditoins and lime appli-
cation on the efficiency of soil-plant
system in the removal of phosphorus in
an overland flow treatment model.
Conclusions
Several controlled laboratory and
s.maH seafc overland, flow experiments
w,«f« conducted te determine the
mechanisms of nitrogen, and phosphorus
refrwwat Processes trp,rn simulated
vwastewaler during overland application.
Crowtoy, Olivier, Mhoon and Granada
soils and ly® grass, Bermuda grass and
Fiea plants \N**e used in the various
experiments.. Overland flow test models,
3Q centimeters (ctnk width, 152 cm
length and 13 cm depth, were used to
study the distribution of applied water
into various system components (Figure
1). The results were fitted into a
mathematical model to predict the
behavior of water movement under field
conditions.
Similar overland flow test models
were used to investigate the fate of
applied labelled nitrogen in an estab-
lished soil-plant system. Measurements
of gaseous nitrogen losses were made
in the sealed overland flow test models*^
(Figure 2) and growth chamber studies
(Figure 3).
The mechanisms of phosphorus
sorption and desorption were investi-
gated under both laboratory and sim-
ulated overland flow conditions. Lab-
oratory studies were conducted to
determine the effects of changes in
redox potential, pH, and the duration of
anaerobiosis and re-oxidation on phos-
l-L
Light Source
r r \
Pt. Salt
Electrodes Bridge
Plants
Pt
Electrodes
Pump
Applied
Effluent
Condensed Sub- Runoff
HZO flow
Figure 1. Longitudinal section of the overland flow wastewater treatment model.
Light Source
«\
' I i
I
' I ' \
Tx'7
12 13
1. Influent inlet 5. Sa/t Bridge 12. Hg manometer
2. Cooling H2O 6.7. Pt. electrodes 13. Pressure control
inlet 8,9. Ga$ samp/ing 14. Runoff '4
3. Cooling H20 outlet 15. Subflow
outlet 10. Thermometer 16. Condensed H20
4. Fan 11. Serum cap 17. Cu tubing
Figure 2. A schematic diagram of the sealed over/and flow treatment model used
in the nitrogen transformation's study. .
-------
1. Cooling HZ0 inlet
2. Cooling H20 outlet
3. Copper tubing
4. Thermometer
5. NH3 trap
6. COz trap
7. H2S trap
8. Plexiglass container with soil and
plants
9. Gas samp/ing and pressure
monitoring outlet
Figure 3. A schematic diagram of
the sealed dessicator
assembly used for the
determination of gaseous
nitrogen losses in a soil-
plant system.
phorus sorption and mobility in soil
suspensions. An overland flow experi-
ment was performed to determine the
influence of prereduced soil conditions
and pH amendment on the removal of
phosphorus-32 applied in simulated
wastewater.
The results of controlled laboratory
and small scale overland flow experi-
ments with simulated wastewater
containing nitrogen and phosphorus
that may be significant in the land
application of municipal wastewater are
as follows:
Water Movement
The recovery of simulated wastewater
in the runoff fraction in the small scale
overland flow experiments ranged from
50 to 60% of applied water. The fraction
of applied water collected in the subf low
was 9 to 23%. The effect of varying slope
from 1.1 to 4.4% on the flow rate of
simulated wastewater was not signif-
icant in the small scale overland flow
model with growing plants as indicated
in Figure 4.
The phenomenon of water movement
in the overland flow model as a function
of slope was computed by a set of
equations representing nonsteady flow
model. The data plotted in Figure 5
indicate that the computed rate of total
water recovered agreed well with
experimental data, but the computed
rate of subflow did not. Thisdiscrepancy
could be attributed to the difficulty in
the experimental determination of the
relative proportional of runoff and
subflow. It is possible that some
physical processes that may influence
water movment in the system were
overlooked.
The results of these computations
and the comparison with the experi-
mental data suggest that the water
movement in the system was primarily
Rate of total flow recovered
Rate of subflow/ratee of total flow recovered
10
Exp. Theo.
10'
700 200 300 \ 400
Time (min)
0
0.5
0
Figure 4. Comparison between the experimental data and the results of computa-
tions fa) 1.1% slope; (b) 2.2% slope; (c) 3.3% slope; (d) 4.4% slope.
-------
n 1200
5 500
10
to
CO
^600
Maximum Recovery^ 2.80% ,
300.
NzQ. not detected
O 20 40 60 80
Time, Days
Figure 5. Gaseous nitrogen-15
production as a function
of time in a sealed over-
land flow Olivier soil-
Bermuda grass system.
controlled by the application rate, the
friction slope, the slope angle, the
hydraulic characteristics of soils, and
the evapotranspiration. The computer
simulation of non-steady flow satis-
factorily predicts the rate of total water
recovered. An understanding of the
physical processes may be an important
first step to obtain quantitative infor-
mation on the mechanisms of chemical
and biological processes in the overland
flow treatment system of wastewater
containing pollutants.
Nitrogen Reactions
Tne contribution of various biochem-
ical tranformations in the removal of
wastewater nitrogen in an overland
flow treatment system were investigated
in several laboratory and overland flow
experiments. The results of these
studies which may be significant in the
land application of municipal waste
water are summarized as follows:
1. The redox potential values of
surface soil in various overland flow
experiments remained well oxidized,
indicating the existence of favorable
conditions for the nitrification of ammo-
nium-nitrogen added to simulated
wastewater. In the subsurface soil,
redox potential values ranged from well
oxidized to very reduced. These aerobic-
anerobic zones in the soil mass facilitate
nitrification-denitrification processes
and enhance nitrogen losses to the
atmosphere.
2. Low soil pH in the range of 5.1 to
6.0 and the lack of easily available
energy source resulted in negligible
losses of applied ammonium-nitrogen
through nitrification-denitrification
reactions. Unfavorably low pH is known
4
to inhibit the growth of both nitrifying
and denitrifying organisms resulting in
reduced gaseous nitrogen loss. Absence
of a readily available carbon source
results in reduced activity of denitrifying
microorganisms.
3. The loss of ammonium-nitrogen
through ammonia volatilization reaction
in the overland flow experiments and
sealed growth chamber studies ac-
counted for about 5% of the total
nitrogen added. The pH of soils used in
various ammonia volatilization studies
ranged from 6.4 to 7.6. This explains the
relatively small losses of applied.
ammonium-nitrogen through ammonia
volatilization reactions. Published lit-
erature suggests that in more alkaline
soil conditions ammonia volatilization
losses would be appreciable.
4. Incomplete nitrification of added
ammonium-nitrogen was occurring in
the overland flow experiments conducted
on Mhoon and Olivier soils and in the
growth chamber studies with Grenada
soil simulating overland flow environ-
ment. These results suggest that
nitrification reactions may be limiting
the proportion of nitrate-nitrogen
available for denitrification reactions.
5. The movement of ammonium-
nitrogen in the simulated wastewater
applied to the overland flow model was
restricted to the upper end of the slope.
Most of the residual ammonium-
nitrogen recovered at the end of
overland flow experiments was present
in the top few centimeters of soil mass.
More nitrate-nitrogen had moved
downslope and in the subsurface soil
compared to ammonium-nitrogen.
6. The loss of applied ammonium-
nitrogen attributed to denitrification
reactions in the overland flow experi-
ments ranged from 2.8 to 35.4%. In the
growth chamber studies where alternate
oxidized and reduced soil conditions
were attained with controlled soil water
content, loss of applied ammonium-
nitrogen was as high as 59%. The
recovery of nitrogen-15 gas with time as
a result of denitrification reaction in a
sealed overland flow system is given In
Figure 5.
7. The plant uptake of nitrogen in the
overland flow and growth chamber
studies accounted for 23 to 62% of
applied ammonium-nitrogen and re-
sulted in the maximum removal of
wastewater nitrogen compared to other
chemical and biochemical processes.
Also, preferential uptake by rye grass
plants of ammonium-nitrogen over
nitrate-nitrogen was demonstrated in^j
the studies.
8. The results of overland flow
experiments and growth chamber
studies demonstrated that the rate of
nitrogen loss in the nitrate-nitrogen
treatments was about twice as much as
in the ammonium-nitrogen treatments.
The results of the experiments sug-
gest that the gaseous loss of applied
nitrogen can be maximized during land
application of wastewater if conditions
favorable for simultaneous nitrification-
denitrification reactions are attained
through careful manipulation of soil-
plant systems. Some of the important
factors that control these reactions are
redox potential, pH, readily available
carbon source, and large population of
appropriate microbes. Any overland
flow treatment facilities aimed at
maximizing nitrogen loss must optimize
these variables. Plant uptake of waste-
water nitrogen during overland flow
application accounts for a large fraction
of nitrogen removed.
The role of various physical chemical,
biochemical and biological processes in
the overall distribution of nitrogen with
the overland flow treatment of waste-
water and in the eventual reductions of
groundwater and stream contamination
is illustrated in Figure 6.
Phosphorus Reactions
Laboratory Studies
The results of these studies can be
applied, strictly, only to acid soils
containing significant quantities (more
than 100O ppm)of reductibleoroxalate-
extractable iron. The results for the pH 8
treatment may be applied, with modi-
fication to alkaline soils, taking into
account the fact that reduction of an
alkaline soil will cause a decrease in pH
whereas these results are for an acid
soil whose pH has been raised artifi-
cially to 8.0. The results of this study
may be interpreted to draw the following
conclusions:
1. At least three days of flooding are
probably required before significant
reduction and changes in phosphorus
sorption and mobility occur.
Applicable to Acid Soils
2. For about 18 days after reduction has
occurred, phosphorus sorption ca-
pacity will be significantly higher and
phosphorus mobility will be lower
than in an aerated soil.
3. After about 20 days of continuous
flooding, there will be a very large
-------
Atmosphere
Water
Flow
Aerobic
Soil Layer
Anaerobic
Soil Layer
NH3
Volatilization
Flow, Mixing
Adsorption
Downward
Flow and
Diffusion
Figure 6. A schematic diagram of various nitrogen removal processes occurring
in an overland flow treatment of wastewater.
increase in sorption capacity and
decrease in leaching losses from
wastewater additions containing
more than 5 ug phosphorus/ml.
However, from more dilute waste-
water there may be a slight increase
in leaching losses compared with
those from aerated soil.
4. Longer periods of flooding (more
than 40 days) will cause a gradual
decrease in phosphorus sorption
capacity and a corresponding in-
crease in phosphorus mobility and
leaching losses.
5. Re-oxidation for periods up to 30
days will cause a marked decrease in
phosphorus sorption capacity, and
leaching losses from previously
applied wastewater will be moder-
ately greater than from wastewater
applied to aerated soil.
Applicable to N on-Acid Soils
6. Phosphorus sorption capacity will be
increased immediately by the addi-
tion of calcium in the wastewater,
causing an increase in calcium
phosphate precipitation.
7. With increasing reduction, pH will
fall to near neutral causing a de-
crease in the negative charge of the
iron oxide surfaces and an increase
in the solubility of calcium phos-
phates. Consequently, phosphorus
sorption by calcium phosphate
precipitation will decrease but the
bonding energy of the iron oxide
surfaces will increase. The net effect
, of these changes would be some
increase in sorption capacity and a
larger increase in buffer capacity so
that leaching losses would be
smaller than from an aerated soil.
8. Re-oxidation would reverse the
changes in pH and buffer capacity so
that the phosphorus sorption capacity
and leaching losses would be similar
to those of an aerated soil.
Overland Flow Studies
The results of the overland flow
experiment conducted on Crowley silt
loam-rye grass sytem established at
1.2% slope demostrate that the effi-
ciency of phosphorus removal from
municipal wastewater would be greatly
enhanced by the addition of lime
(calcium carbonate) to the soil. The
prereduction of soil-plant system for
extended period of time may also result
in more phosphorus removal than less-
reduced, non-limed soil. A mass balance
of applied Phosphorus-32 recovered at
the end of the overland flow experiment
is given in Table 1.
The phosphorus sorption isotherms
conducted in the laboratory on the
pretreated soils and the sorption pa-
rameters computation by the Langmuir
two-surface equation demonstrated
that the efficiency of phosphorus
removal in the overland flow experiment
was related to the phosphorus sorption
capacity of various treatments. A
slightly higher desorption of phosphorus
sorbed in the overland fow as well as
laboratory studies in the limed treat-
ments was due to the lower phosphorus
bonding energy in this treatment.
Phosphorus sorbed under alkaline
conditions was more available to
growing plants than in the unampnded
treatments. The results of this study
suggest that phosphorus sorption is a
kinetic process and that the leaching
losses of phosphorus retained by the
soil mass during overland flow appli-
cation would be smaller due to the
longer reaction time as compared to the
desorption potential determined in the
laboratory.
Recommendations
Nitrogen
The results of nitrogen studies
suggest that the gaseous loss of
nitrogen should be maximized to im-
prove the efficiency of nitrogen removal
in the overland flow treatment system.
Important parameters such as pH,
easily-available energy source, and
application schedule should be carefully
manipulated to maximize simultaneous
nitrification-denitrification reactions.
Whenever economically feasible, lime
additions may be made with the waste-
water to enhance ammonia volatiliza-
tion losses of applied ammonium
nitrogen. More research should be
conducted on the selection of plant
species having affinity for greater
nitrogen accumulation and which can
be harvested often to maximize nitrogen
removal.
Table 1. Mass Balance of Applied Phosphorus-32 Recovered at the End of
Overland Experiment in Crowley Soil-Rye Grass System
System component
32P added
Runoff plus subflow
Plant uptake
Remaining in soil
untreated
100
34.65
14.37
50.98
Treatments
prereduced
32P recovered, %
100
18.25
11.32
70.43
limed
100
3.21
18.45
78.34
-------
Phosphorus
Because of the importance of soil pH
and reducible iron content in phosphorus
reactions further research of this type
should be carried out on a range of soils
varying in pH and reducible iron
content. Short-term kinetic studies of
phosphorus sorption are important
because adsorption and precipitation
are fast reactions. However, the kinetics
of anaerobic changes, such as pH and
iron chemistry, which affect subsequent
phosphorus sorption and mobility,
require further study. Long term kinetic
studies of phosphorus sorption due to
occlusion in hydrous oxide crystals and
organic incorporation also require
further research. Experimental evi-
dence suggests that re-oxidation of soils
can reverse the beneficial effects of
reduction on phosphorus sorption
capacity and mobility. Further longer
term studies should be conducted to
optimize management strategies for the
duration of wastewater application and
re-aeration intervals. Field studies
should be conducted to determine the
rates of lime applications to maximize
phosphorus precipitation during over-
land flow applications.
R. A. Khalid, I. C. R. Holford, M. N. Mixon, and W. H. Patrick, Jr., are with the
Laboratory for Wetland Soils and Sediments, Center for Wetland Resources,
Louisiana State University, Baton Rouge, LA 70803.
Bert E. Bledsoe is the EPA Project Officer (see below).
The complete report, entitled "Nitrogen and Phosphorus Reactions in Overland
Flow of Wastewater," (Order No. PB 81-239 311; Cost: $15.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, v'A 221'61
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
US GOVERNMENT PRINTING OFFICE. 1981 — 757-012/7322
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |