United States
Environmental Protection
Agency
Robert S Kerr Environmental Resg
Laboratory
Ada OK 74820
vvEPA
Research and Development
EPA-600/S2-81-093 July 1981
Project Summary
Overland Flow Treatment of
Poultry Processing
Wastewater in Cold Climates
Lloyd H Ketchum, Jr., Jack L Witherow, Arthur J Cunningham, and Robert
L Irvine
Land treatment of wastewater using
overland flow methods has been
effectively used in warm climates. A
full scale wastewater treatment facil-
ity was evaluated, including overland
flow in northern Indiana, a cold
climate. The study emphasized the
evaluation of the overland flow
system, but the treatment system
included mechanical pretreatment
facilities, a storage lagoon, a lagoon
for batch chemical treatment of the
overland flow effluent and rapid infil-
tration for effluent disposal. The facil-
ity treated wastewater from a duck
processing plant. That wastewater
was as amenable to biological treat-
ment as domestic sewage and had
similar concentrations of nutrients
and solids.
Overland flow systems typically are
limited to low permeability soils. This
system was located on a sandy loam
soil with very high permeability.
Bentonite was used to provide a
percolation barrier below the grass
covered surface.
The Project Report describes the
design and construction of the facil-
ities, presents and evaluates
monitoring data from 1-V2 years of
operation and gives construction
costs. The many problems encoun-
tered including those associated with
cold weather operation are also
described to prevent their repetition.
This Project Summary was develop-
ed 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
This project demonstrated the use of
an overland flow system in a cold
climate The system was constructed at
the Culver duck processing plant, and
was located near the Indiana-Michigan
border The major objective of the study
was to demonstrate the operation of an
inexpensive, simply operated,
inexpensive, simply operated, slaugh-
terhouse wastewater treatment facility,
which would meet the 1983 National
discharge limitations A lagoon which
collected the overland flow effluent was
added to evaluate the effectiveness of
batch chemical treatmentfor phosphate
and suspended solids reduction and dis-
infection A bentonite seal was used to
allow overland flow treatment at a site
with highly permeable soils
Treatment Plant
The wastewater treatment tram con-
sisted of five types of treatment units
-------
connected for series operation as
shown in Figure 1 The pretreatment
facilities included a fine screen, Hydro-
cyclonic Rotostramer, and a gravity
grease separator Inadequate operation
of those facilities, which were located in
the processing plant, and overloadmgs
resulting from surges of flow, caused
frequent grease and solids discharges
into subsequent treatment units These
discharges clogged pumps, pipes and
orifices, causing pump damage and
substantial operating problems
Processing Plant
Effluent
Pretreatment
Pump Manhole
Lagoon
Overland Flow
Treatment
Chemical
Treatment
Lagoon
Rapid
Infiltration
Figure 1. Wastewater treatment
plant schematic
The pretreated effluent flowed by
gravity into a manhole where two sub-
mersible pumps were located When
the pumps were off, the flow continued
by gravity into a large lagoon The
lagoon primarily served as off-line
storage during periods when the over-
land flow treatment plots were not
operating. During lagoon storage, how-
ever, suspended solids settled, bio-
chemical oxygen demand was reduced,
and the wastewater characteristics
changed from those of a raw waste-
water to those of a lagoon effluent, in-
cluding high algae concentrations
during the summer
When the pumps were on and waste-
water was also flowing from the pro-
cessing plant, the pretreated effluent
was pumped directly to the overland
flow plot During periods when the pro-
cessing plant was not in operation, the
pumps drew down the manhole liquid
level so that the lagoon water flowed
back into the manhole and was pumped
to the overland flow plots. During the
last stages of the project, the pumps
were removed from the manhole and
suspended in the lagoon, to avoid pump
clogging when the pretreatment facil-
ities failed This resulted in only lagoon
effluent being pumped to the overland
flow plots.
Two overland flow plots were opera-
ted in parallel. The influent was applied
through a low pressure pipe across the
top of the slope The effluent was col-
lected in a ditch across the foot of the
plot and piped through the V-notch weir
into the batch chemical treatment
lagoon When the lagoon became full, a
floating mechanical aerator was used to
mix the lagoon contents. Chemical solu-
tions of sodium hypochlonte for disin-
fection, followed by alum for phosphate
and suspended solids reduction, were
added at a point near the aerator for
mixing After chemical addition was
completed and large floe particles
observed, the aerator was shut off and
the contents of the lagoon were allowed
to settle
A pump connected to a floating inlet
was used to discharge the treated
supernatant into the rapid infiltration
area The rapid infiltration area provided
for final effluent polishing and water
disposal
Overland Flow Plots
The overland flow plots were unique
and warrant additional consideration
The site was located on a sandy loam
soil (U S. Soil Conservation Service -
Oshtemo Series), with a permeability of
5 to 1 5 centimeters per hour (cm/hour)
Since this high permeability would have
resulted in rapid infiltration instead of
the desired overland flow, the site was
modified to allow overland flow treat-
ment
The site was first stripped of topsoil,
then graded to provide a 4% slope down-
hill and was leveled across the direction
of the slope. Bentonite was applied at a
rate of 5 kilograms per square meter
(kg/m2) to the graded subsoil and was
mixed in to a depth of 5 centimeters
(cm) A 1 5-cm layer of topsoil then was
carefully replaced above the bentonite
seal and grass was planted A mixture of
Kentucky 31 fescue and perennial rye-
grass was used
Construction costs for the overland
flow treatment plots of 031 hectares
were about $10,000 or $32,000 per
hectare in terms of September 1978
dollars This included the cost of earth-
work, bentonite, grass, pumps, and pip-
ing but did not include the cost of the
land A detailed description of these
costs and the plot construction tech-
nique are included in the final report.
Conclusions
The discussion of conclusions has
been divided into four sections F:irst,
the overland flow plots are discussed
primarily from a consideration of the
hydraulics The performance of the
overland flow plots, or effectiveness of
treatment, is discussed in the following
two sections, divided according to warm
weather and cold weather operations
The final section is a discussion of the
batch chemical treatment lagoon
Overland Flow: Hydraulics
The plots were approximately 80
meters (m) long This is unusually long,
about two to three times longer than
most other overland flow plots The
additional length was provided because
the wastewater strength was expected
to be several times greater than actually
occurred. The mam disadvantage was'
the high hydraulic loading per unit
width of the plot For example, if two
plots were equal in area, but one was
one-third as long, the longer plot would
have three times as much water flowing
along a unit width of the surface As a
consequence of the high hydraulic load-
ing, the grass at the top of the plots died,
and that area served ma inly to distribute
the flow
Shortcircuitmg was most prevalent in
the area without grass cover, but also
occurred in the downhill grass-covered
areas due to surface irregularities An
effective method of control was devel-
oped Small earthen dikes, about 1 5 cm
high and 30 cm at the base, were con-
structed at 45° or less to the desired
downhill direction of flow. Most dikes
were 2 m to 5 m long and were con-
structed during wastewater application
to redirect the flow in the desired direc-
tion
Another unique feature of the plots
was the void space m the topsoil above
the bentonite seal, which needed to be
filled with water prior to any overland
flow effluent. The flow rate downhill
through the topsoil layer was very low
Typically, it took about two days of con-
tinuous application to fill the void spacjB
-------
and for effluent to begin flowing from
the plot After that, wastewater flowed
along the ground surface with detention
times of about 3 hours to 4 hours, much
like sites with impermeable soils
Early in the study, the plots were
loaded, similar to other overland flow
sites, 8 to 10 hours per day 5 to 7 days
per week This cycle resulted in even
more serious problems at the top of the
plot The area was almost always satur-
ated and little effluent was generated
The lower portions of the plots were
underloaded and often too dry
Later in the study, a different cycle
was adopted This cycle provided con-
tinuous application for three to five days
followed by two or three days of drying,
and resulted in hydraulic loadings of 14
to 34 cm/week The continuous loading
for several days followed by extended
drying periods proved to be an effective
method of operation
Effluent flow rates, measured during
periods of wastewater application,
varied from zero, when the plots were
dry and evapotranspiration was high, to
rates greater than influent flow rates
The high effluent flow rates occurred
during periods of rainfall and snowmelt
Table 1 was prepared for one of the plots
which had total flows available for all
four seasons Averaged over those
1 longer time periods, total effluent flow
was about 40% of the total influent,
Table 1. Overland Flow Water
Balances
Total
Seasonal
Water
Volume,
m3 Fall Win Spr Sum
Waste-
water
applied 900 580 1700 1080
Precipita-
tion 90 50 270 50
Total
Influent 990 630 1970 1130
Effluent 410 470 780 500
except during the winter The high
winter effluent flows resulted from low
evapotranspiration rates, because the
fields were snow covered, and low infil-
tration rates, because the ground was
frozen
Overland Flow: Warm
Weather Treatment
Waste production at the Culver
processing plant was determined on
several different days with the average
rates shown in Table 2. The USEPA ef-
fluent limitations proposed for 1983
duck processors, in terms of averages of
daily values for 30 consecutive days, are
shown in Table 3 Also included in Table
3, are estimates of maximum concen-
trations in the Culver wastewater if it is
to meet those proposed limitations
Table 2. Wastewater Characteristics
Item
Waste
Concentration production
mg/l per duck
Wastewater
Volume
5-Day Biochem-
ical Oxygen
Demand
Chemical
Oxygen
70.4 liters
256 18 grams
Demand
Suspended
So/ids
Total Kjeldahl
Nitrogen
398
170
28
28
12
2
grams
grams
grams
Effluent
as percent
of total
influent
40
75
40
45
During all warm-weather periods,
and under widely varying wastewater
application schemes (i e , both 8-16
hours/day at 9 to 25 cm/week, and 3 to
5 days continuous application followed
by 2 to 3 days drying at 14 to 34
cm/week), the overland flow plots alone
met the limitations shown in Table 3,
except when algae problems were en-
countered Effluent concentrations of 5-
day biochemical oxygen demand
(BODs), suspended solids and oil and
grease during warm-weather operation
were all typically about one-half of the
limitations shown m Table 3 Table 4
shows average effluent values for the
fall and summer periods The summer
period has been further divided to show
the effect of high influent algae concen-
trations During the early summer
period there was no substantial influ-
ent algae However, algae produced in
the lagoon resulted m high influent
algae concentrations during the mid-
summer periods, and a reduction in
effluent quality The poor quality efflu-
ent was almost a direct result of the
presence of influent algae Effluent
samples which were filtered to remove
algae, exhibited characteristics very
nearly the same as occurred during
periods when algae was not present m
the influent
Effluent analyses were made during
several short intense rainfall periods
when wastewater was being applied
and effluent was being produced Al-
though flow rates increased, no sub-
stantial change in effluent concentra-
tions was observed
Overland Flow: Cold
Weather Treatment
Cold weather treatment refers to the
winter and early spring periods when
temperatures averaged -6°C and 5°C,
respectively Soil temperature appeared
to be the most important variable affec-
ting effluent quality The presence or
absence of snow cover was very impor-
tant m determining the effect of air
temperatures on soil temperatures.
Operation was begun during the
coldest month of the year, January,
after two months of no operation At the
beginning of the January operation, the
overflow plot was covered with snow at
a depth of about 1 m The wastewater
applied was either raw wastewater at
about 20°C or lagoon effluent at about
5°C All wastewater was applied during
the day (8 to 16 hours) and the plot was
allowed to rest during the night. As a
result of the daily loading, the soil was
saturated all the time because the
nighttime drying periods were insuffic-
ient to remove the water Also, hydrau-
lic loading rates were high, about 15
cm/week, which probably contributed
to reduced effluent quality In retro-
spect, the cyclic operation adopted
during the following summer, continu-
ous application for several days fol-
lowed by several days of rest, would
have been a better choice
Air temperatures did not exceed 0°C
during the first three weeks after start
up However, the snow cover across the
top 1 5 m of the plot was melted by the
influent wastewater, and temperatures
-------
Table 3.
Effluent Limitations
Item
5-Day Biochemical
Oxygen Demand
Suspended Solids
Oil and Grease
Ammonia Nitrogen
Proposed 1983
Limitation
0.39 kg/kkg LWK*
046 kg/kkg LWK
026 kg/kkg LWK
4 0 mg/l
Estimated Concentrations
for Culver Duck Faring
33 mg/l
39 mg/l
22 mg/l
4 mg/l
*L/ve weight killed kg/kkg LWK is equivalent to lb/1000 Ib LWK.
\Based on 70 liters/duck wastewater production, 50% water loss during overland flow
treatment and a live weight of 3 kilograms per duck
Table 4. Average Warm Weather
Overland Flow Effluent
Wastewater Summer
Characteristic Fall Early Mid
5-day Biochemical
Oxygen Demand,
mg/l 7 16 38b
Suspended So/ids,
mg/l n.a. 12 45C
Total Nitrogen3, mg/l
as N 4 6 11
Total Phosphate,
mg/l as P 2.2 2.3
Oil and Grease, mg/l n. a. 7
aSum of organic, nitrite, nitrate, and
ammonia nitrogen.
bWhen filtered to remove algae, the
average was 17 mg/l
c These suspended so/ids were primarily
algal cells
were observed as high as 4°C m the top
5 cm of soil. Below that warmer soil, the
ground remained frozen Shortly after
each period of wastewater application
ended, ice formed across the top 15-m
area However, during wastewater
application no freezing occurred even
when air temperatures dropped to
-15°C, except occasionally when the
colder lagoon effluent was being
applied
Below the snow cover, a thin layer of
soil remained at temperatures between
0°C and 2°C Below that warmer soil
layer, the soil remained frozen. Overthe
lower one-third of the plot, there was a
layer of slush about 2 cm to 5 cm deep
The effluent temperatures were never
observed to be greater than 1°C and
typically were less than 0 5°C
During that period of snow cover, the
start up period, the BOD5 in the effluent
gradually decreased from about 240
mg/hoabout 95 mg/l Cold weather did
not prevent microbial activity Clusters
of what appeared to be white fungi were
observed The fungi probably dominated
other microbes due to their ability to
adapt to the cold, and their ability to
extend hyphae to the water surface to
obtain oxygen Psychrophilic fungi are
known to grow and oxidize carbon in
temperatures near and lower than 0°C
The fairly rapid improvement of BOD5
observed after start up, was probably
due to increases in microbial population
and not increases in oxidation rates of
individual microbes It is unclear how
much greater the improvement would
have been if the snow covered condi-
tions had continued for a longer period
of time However, warmer weather,
including ram, occurred and the snow
cover was lost. Effluent BODs finally
reached effluent limitations shortly
after the snow melted
After the snow melt period, the
average air temperature was 4°C for
about one month During that period, air
temperatures fluctuated from above
freezing to below freezing. Effluent
BOD5 values were very sensitive to air
temperatures A fairly linear relation-
ship between air temperatures and
effluent BOD5 values was observed In
general, effluent BOD5 values did not
meet effluent limitations atairtempera-
tures below 0°C, just met them atO°Cto
5°C, and were well within the limita-
tions above 5°C One interesting
observation was made as average air
temperatures continued to increase
Following a week when daily mean air
temperatures averaged 11 °C, mean
daily temperatures dropped to -2°C on
one day and -6°C on the next Effluent
BOD5 increased from the warmer
weather values of about 10 mg/l to over
50 mg/l on the second day of the cold
period That increased sensitivitytocold
temperatures after a period of warm
temperature probably was due to a
change in the soil microbial population.
Mesophilic bacteria and other moderate
temperature microbes probably gam the
ecological advantage during the warm
spell The ability of the newly dominant
mesophilic microbes to oxidize orgamcs
was seriously hindered in sub-freezing
temperatures
Effluent suspended solids concentra-
tions generally met" effluent limitations
during the cold weather period
Nitrogen removals were generally poor
during that same period. A major spill of
anhydrous ammonia from the plant
refrigeration system into the waste-
water also seriously interfered with the
nitrogen study Some nitrogen removal
did occur during the cold weather The
mechanisms of nitrogen removal were
uncertain.
During snow melting periods, effluent
BOD5, suspended solids and ammonia
concentrations were all relatively high
However, effluent from the adjacent
overland flow plot, which had received
no wastewater during the winter, was
essentially the same as the effluent
from the plot which had operated during
the winter. The total mass of effluent
pollutants associated with the melted
snow was less than the total mass
which would have been applied in about
3 crn of a mixture of raw wastewater
and lagoon effluent.
The average effluent characteristics
are shown in Table 5 for the winter
period, when snow cover existed and
spring, a time after snowmelt had
occurred
Chemical Batch Treatment
Lagoon
Overland flow treatment resulted in
very little phosphate reduction during
cold weather and warm operation How-
ever, effective phosphate reduction was
provided during chemical batch treat-
ment in the lagoon located at the foot of
the overland flow plots Disinfection
was also provided in that lagoon.
-------
Table 5. Average Cold Weather
Overland Flow Effluent
Wastewater
Characteristic Winter Spring
5-day Biochemical
Oxygen Demand,
mg/l 135 22
Suspended Solids,
mg/l 40 15
Total Nitrogen", mg/l
as N 54 11
Total Phosphate,
mg/l as P 2826
Oil and Grease, mg/l 7 n.a
*Sum of organic, nitrite, nitrate, and
ammonia nitrogen.
A complete ana lysis was made on one
batch treatment The treatment in-
cluded addition of a sodium hypochlo-
rite solution resulting in a 5 mg/l
chlorine dosage. Mixing was provided,
using the floating mechanical aerator.,
; during chlorine addition and was con-
tinued for 30 minutes Chlorine resid-
uals both total and free, were greater
than 3 mg/l after the 30-mmute mixing
period An alum solution was then
added to the mixing lagoon over a period
of one hour, which resulted in an
aluminum dosage of 50 mg/l After 10
minutes more of mixing, the aerator
was shut off, and the lagoon contents
allowed to settle for 1 5 hours Table 6
compares the lagoon contents before
and after treatment.
Recommendations
Overland flow should be considered
as a wastewater treatment alternative
at sites with*permeable as well as im-
permeable soils Application of more
than 5 kg/m2 of bentonite is recom-
mended to ensure a better percolation
barrier at future sites.
In systems requiring increased phos-
phate removal disinfection, or a greater
assurance of suspended solids reduc-
tion, a small lagoon should be construc-
ted to collect the overland flow effluent.
This provides opportunity, using simple
batch chemical treatment and
controlled discharge, to improve efflu-
quality. This relatively inexpensive
treatment is well suited for use in con-
junction with overland flow.
Research should be undertaken to
determine optimum wastewater appli-
cation cycles Experience at the site
suggests that 24 hours per day applica-
tion for several days, followed by two or
three days of drying, is a cycle which
deserves future investigation.
Winter storage is needed However,
winter operation needs to be investi-
gated further This investigation should
concentrate on overland flow treatment
on snow covered plots, because sub-
stantially higher quality effluents are
likely with the snow cover to protect the
soil from very low and highly fluctuating
temperatures The study indicated that
some degree of treatment is possible in
the winter, but that loading rate must be
reduced, and unsatisfactory effluent
returned for further treatment. Winter
wastewater application is recommend-
ed to develop a microbial population
capable of treating the wastewater
immediately following the spring thaw.
Further investigations are necessary
to determine the effect of sudden
decreases in temperature in late fall and
early spring. It may be necessary to stop
Table 6. Lagoon Chemical Batch
Treatment
Wastewater
Characteristic
Treatment
Before After
5-day Biochemical
Oxygen Demand,
mg/l 38
Suspended Solids,
mg/l
Ammonia Nitrogen,
mg/l
Nitrate plus Nitrite
Nitrogen, mg/l
Total Kjeldahl
Nitrogen, mg/l
Total Phosphate,
mg/l as P
29 20
72 70
0.3 0.1
12.9 8.5
2.8 0.4
Inorganic Phosphate,
mg/l as P 2.1 0.1
wastewater application during these
short cold spells
Application of lagoon wastewater to
an overland flow plot should be avoided
if possible. Operators of systems using
storage lagoons prior to overland flow
treatment are recommended to by-pass
these lagoons during peak algae grow-
ing periods. Removal of sufficient grit
and solids to prevent pump and nozzle
plugging is recommended when
lagoons are by-passed or raw waste-
water is applied directly to overland flow
plots
Coliforms
pH
n.a. 0
n. a. 6.5
-------
Lloyd H Ketchum, Jr. and Robert L. Irvine are with the University of Notre Dame,
Notre Dame, IN 46556; Arthur J. Cunningham is with the New Hampshire
Water Supply and Pollution Control, Concord, NH 0330 1; Jack L. Witherowis
the EPA author as we/1 as the EPA Project Officer (see below).
The complete report, entitled "Overland Flow Treatment of Poultry Processing
Wastewater in Cold Climates," (Order No. PB 81-213 225; Cost: $12.50,
subject to change) will be available only from.
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
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 1W1 -757-012/7273
-------
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
RETURN POSTAGE GUARANTEED
i) POO.}?1*
c *. 'J I H t KjI"r,Ci luM A*.if?>C
« 5 M.M
------- |