United States
Environmental Protection
Agency
Robert S Kerr Environmental Research
Laboratory
Ada OK 74820
EPA >
Research and Development
xvEPA
Summary of
Long-Term Rapid
Infiltration System
Studies
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-165
July 1980
SUMMARY OF LONG-TERM RAPID INFILTRATION SYSTEM STUDIES
By
Lowell E. Leach, Carl G. Enfield, and Curtis C. Harlin, Jr.
Wastewater Management Branch
Robert S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 74820
-------
DISCLAIMER
This report has been reviewed by the Robert S. Kerr Environmental
Research Laboratory, U. S. Environmental Protection Agency, and approved
for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
ii
-------
FOREWORD
The Office of Research and Development of the Environmental Protection
Agency conducts research involving the search for information about environ-
mental problems, management techniques, and new technologies to identify,
control, and eliminate the threat pollution poses to the welfare of the
American people. This research is conducted through a nationwide network
of research facilities.
As one of these facilities, the Robert S. Kerr Environmental Research
Laboratory is responsible for the management of programs to: (a) investigate
the nature, transport, fate and management of pollutants in ground water;
(b) develop and demonstrate methods for treating wastewaters with soil and
other natural systems; (c) develop and demonstrate pollution control tech-
nologies for irrigation return flows; (d) develop and demonstrate pollution
control technologies for animal production wastes; (e) develop and demon-
strate technologies to prevent, control or abate pollution from the petro-
leum refining and petrochemical industries; and (f) develop and demonstrate
technologies to manage pollution resulting from combinations of industrial
wastewaters or industrial/municipal wasteWaters.
The purpose of this report is to summarize the results of detailed
studies of four rapid-infiltration systems. These studies were a part of
a major effort by EPA to determine the long-term effects of land treatment
of municipal wastewater. For each of the four projects summarized in this
report, a detailed final report has been prepared. These reports contain
complete data and data evaluation. The objective of this report is to pro-
vide a summary of pertinent information which may, in many cases, be
sufficient for the users. Those who are interested in the complete data
from any of these projects should refer to the individual project reports.
It is hoped that this summary will be useful to those who do not require
this much detail.
William C. Galegar
Director
Robert S. Kerr Environmental Research Laboratory
-------
ABSTRACT
This research project was initiated with the overall objective of
summarizing and comparing the data generated from individual reports of four
long-term rapid infiltration municipal wastewater systems. Evaluation of
this material gives the user community a condensed assessment of the treatment
received at each of these sites from which projected treatment of planned
systems can be made.
During the review and summary process, the treatment efficiency of the
systems was compared based on level of effluent pretreatment, hydraulic
loading rate, depth of soil profile available for treatment, and variation
in operational practice.
In addition to a summarization and evaluation of data, a hypothetical
design was made for each system based on a rationale for hydraulic loading
and effluent water quality considerations. This rationale is proposed as a
supplement to the design criteria presented in the Process Design Manual for
Land Treatment of Municipal Wastewaters (1). The rationale considers, in
addition to hydraulic acceptability of the most restrictive layer, water
quality parameters which can also be limiting. These parameters include:
sodium adsorption ration+(SAR), biochemical oxygen demand (BOD), suspended
solids (SS), ammonia (NHJ, nitrate (NCU), and phosphorus (P).
Testing of the design of each of these systems with measured water
quality data indicated three of the systems are being operated to produce
the quality of effluent predicted under present design considerations. The
fourth system should be capable of improving operation by underdraining and
changing management practices.
IV
-------
CONTENTS
Foreword iii
Abstract iy
Figures vi
Tables vii
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. Description of Four Existing Systems
General 7
Hollister, California 7
Lake George, New York 10
Vineland, New Jersey 11
Milton, Wisconsin 12
5. Comparison of Applied and Renovated Wastewater 14
6. Observed Changes in Soil Chemical Properties 24
7. Rationale for Rapid infiltration Design 29
References 49
-------
FIGURES
Number Page
1 pH comparisons in the soil profile 25
2 Zinc concentrations in the soil profile 25
3 Copper concentrations in the soil profile ...... 26
4 Total phosphate comparisons in the soil profile ... 26
5 Calcium carbonate reduction in the soil profile. ... 28
6 Organic nitrogen reduction comparisons in the
soil profile 28
7 Nitrification and denitrification versus temperature . 31
VI
-------
TABLES
Number Page
1 Initial Site Screening Criteria 8
2 Operating Factors for Existing Rapid Infiltration
Systems 9
3 Comparison of Applied Wastewater vs. Shallow Ground
Water 16
4 Comparison of Shallow Ground Water to Control
Wells 18
5 Input Data Required to Evaluate Hollister Site ... 35
6 Hollister Design Evaluation 38
7 Input Data Required to Evaluate Milton Site 39
8 Milton Design Evaluation 41
9 Input Data Required to Evaluate Lake George Site . . 42
10 Lake George Design Evaluation 44
11 Input Data Required to Evaluate Vineland Site . . . . 45
12 Vineland Design Evaluation 43
vn
-------
SECTION 1
INTRODUCTION
Many years of sewage farming, both in Europe and in the United States,
show land treatment systems using municipal sewage for irrigation and ground
water recharge can be managed so that they are viable alternatives to
conventional municipal sewage treatment systems. The number of these systems
is steadily growing, particularly in cities where land is available
and irrigated agriculture is normally practiced. But, nevertheless, only a
small percentage of the total number of municipalities use land treatment
systems.
An intense effort is being made by federal agencies to develop land
treatment technology and improve control practices. Through these efforts
and through practical experience, three basic types of land treatment
systems are now recognized. These three systems are identified as slow rate
(crop irrigation), rapid infiltration (infiltration-percolation), and overland
flow systems.
With the passage of the Federal Water Pollution Control Act Amendments
of 1972 (Public Law 92-500), there was a new impetus to utilizing land treat-
ment systems in the United States. In the Act, the Administrator of the
Environmental Protection Agency (EPA) is directed to encourage construction
of wastewater treatment systems which: (1) recycle potential pollutants
through the production of agricultural, silvicultural, and aquacultural
products; (2) reclaim wastewater; and (3) eliminate the discharge of
pollutants. Following the passage of this legislation, the EPA adopted the
policy that land treatment must be evaluated as an alternative for all waste-
water treatment systems funded under its Construction Grants Program.
With the strong committment of the Federal Government to encourage the
adoption of land treatment, where appropriate, planners and designers were
faced with the problem of inadequate and incomplete design criteria for these
systems. Furthermore, there was strong resistance, in many quarters, to the
use of land treatment. This resistance came from state regulatory agencies,
the engineering community, government officials at all levels, and often
from the general public. Much of this resistance was generated because of
questions related to health effects, lack of reliable design criteria, and
long-term effects of such systems.
In order to provide answers to these questions, the EPA, during the
latter part of 1975, awarded contracts to study the long-term effects of
eight existing land treatment systems. The sites selected for study included
-------
five slow rate and three rapid infiltration systems. The sites were selected
to represent different preapplication treatments, different geographical
locations, different soil types, and different management practices. The
contractors were to collect data on numerous chemical and biological para-
meters and compare the properties of crops, soils, and ground water at the
wastewater management sites against appropriate control sites. A basic
criterion in site selection was that all study sites had to have an uninter-
rupted history of at least ten years of operation.
In addition to the eight systems studied under these contracts, two other
systems were studied under grants, making a total of ten study sites. Of
the ten systems evaluated, four were rapid-infiltration systems and six were
slow rate systems. All projects have been completed and individual reports
have been prepared reporting, in detail, the results of the studies.
The purpose of this report is to summarize the results of the four rapid-
infiltration studies. The individual project reports contain all collected
data and are quite voluminous. It was felt that a summary, such as
presented in this report, would be helpful to many readers. Those who have
the need for complete details of the studies, however, should refer to the
individual project reports.
Subsequent to the initiation of these projects, there was published the
"Process Design Manual for Land Treatment of Municipal Wastewater" (1). This
manual was a cooperative effort of the U. S. Army Corps of Engineers, the
U. S. Department of Agriculture, and EPA. This was the first design manual
ever published for land treatment systems, and it was recognized that it was
deficient in many respects because of lack of reliable data. A major benefit
of the existing system studies is to provide new and expanded data for the
updating and revising of the design manual. Included in this report are
designs of the systems studied based on collected data with comparisons of
design criteria contained in the design manual. These comparisons show where
inadequate criteria exist and which areas of the manual need modification and
expansion.
-------
SECTION 2
CONCLUSIONS
After careful review and comparison of the treatment of the four rapid
infiltration systems, a number of conclusions can be made.
Two of the systems were applying primary pretreated wastewater to infil-
tration beds and two were applying secondary pretreated wastewater. Comparison
of the treatment efficiency of infiltration beds for these two levels of pre-
treated wastewater indicated a higher percentage of removal for selected
parameters occurred in the systems with primary pretreatment. Both levels of
pretreatment demonstrated the potential for producing a high quality effluent.
Fecal coliform reductions clearly need improvement, particularly where shallow
water tables exist.
Only one of the four systems evaluated the effects of seasonal changes on
the treatability of wastewater even though all four systems were operated
year-round. It was found that these systems can be successfully managed and
operated through severe winters, with reasonable treatment efficiency as
demonstrated in upper New York State at Lake George.
Lake George, New York, and Hollister, California, demonstrated exceptional
capability for nitrogen removal through the soil treatment systems. Both of
these systems were managed using well scheduled wetting and drying cycles.
Apparently, the total organic carbon available for denitrification and the
management utilized at Lake George and Hollister account for the effective-
ness of treatment in these two systems.
Phosphorus concentration was reduced at each of the treatment sites.
The concentration in the shallow ground water at all test sites was greater
than in the corresponding control ground water, indicating that with the
application rates and detention times used, phosphorus did migrate with the
applied wastewater.
Iron appears to be leaching from the profile at all sites, possibly due
to reducing conditions developed during flooding cycles. Generally, the
other metals reported at the sites did not indicate significant differences
in concentration between the wastewater applied and that reaching the shallow
ground water. All metal concentrations, however, were within acceptable
limits for drinking water quality (2).
-------
Copper, nickel, and zinc showed significant accumulation near the surface
of the soil profiles. This accumulation appears to be a process of precipi-
tation rather than ion exchange since there was little change in ionic ratios
of the cations. Phosphorus was also shown to accumulate in the soil profiles
throughout the 300 centimeters (cm) -[9.8 feet (ft)] monitored. Organic
nitrogen accumulated at the soil surface but accumulations did not penetrate
below 30 cm (1 ft) at any site. In the one system where calcium carbonate was
monitored in the soil profile, significant leaching was apparent. The
operation of this system was reported to have neutralized the pH of the soil
profile.
The pesticides that were detected in the wastewaters were all found in
the shallow ground water indicating these pesticides are mobile in the soil
profile. Lindane was the only pesticide that was found significantly
elevated over background ground water concentrations.
Evaluation of the treatment achieved in the four system indicates
they are generally operating in conformance with current critera published
in the Process Design Manual for Land Treatment of Municipal Wastewaters
(1) as well as the proposed design rationale.
-------
SECTION 3
RECOMMENDATIONS
Rapid infiltration systems have been shown to effectively treat municipal
wastewater. Due to their minimal area requirements, it is recommended that
this mode of wastewater treatment be given careful consideration as a viable
treatment alternative.
Observations of all four sites indicated fecal coliforms were leached to
the ground water. Additional research is required to delineate the soil's
ability to remove fecal coliform. This research needs to determine the
effects of soil temperature, detention time, and soil texture on the soil's
ability to remove fecal coliforms.
Some treatment sites showed less than optimum nitrogen removal. Careful
consideration should be given in evaluating the drainage requirements to
assure adequate oxidation of the wastewater. Nitrogen problems are also
closely related to operational practices. Loading at more than 6 cm/hour (hr)
(2.4 inches (in)/hr) or loading cycles with less than three days of drying
tend to reduce treatment effectiveness. When plans for operation beyond
these limits are being developed, pilot scale testing may be needed to demon-
strate treatment capability at specific sites.
Metal ions showed potential mobility through the soil profile. Addition-
al study is recommended to develop a design rationale capable of addressing
this potential problem. It should be noted that municipal wastewater
generally meets drinking water standards for metals, and problems are not
anticipated except where significant industrial input is included in the
waste stream.
The total organic carbon in the applied wastewater is significantly
reduced through land treatment. It was observed, however, that some organo-
pesticides did move undegraded through the soil profile. Study is needed to
determine which toxic organics will not degrade and which of these show
a potential for mobility. Pretreatment may be required for those organic
compounds which might contaminate the ground water.
It is recommended that the revised rationale of rapid infiltration system
design presented in this report be tentatively adopted and tested for use in
updating the Process Design Manual for Land Treatment of Municipal Wastewaters
(1). This rationale considers, in addition to loading design based on the
present parameters of hydraulic acceptability of the most restrictive layer,
-------
loading design based on water quality parameters+which might be_even more
limiting. These parameters include ammonium (NH^), nitrate (NO^), and
phosphorus (P).
-------
SECTION 4
DESCRIPTION OF FOUR EXISTING SYSTEMS
GENERAL
The four existing rapid infiltration systems evaluated and compared
include Hollister, California (3); Lake George, New York (4); Vineland,
New Jersey (5); and Milton, Wisconsin (6). All of these except the Lake
George system and one slow-rate system were selected based on predetermined
screening criteria by EPA (Table 1). Grants to evaluate the Lake George
system and the San Angelo, Texas, system were already in progress when the
decision was made to conduct a study of existing land application systems by
contracts. Each of the other three sites listed above were selected from a
number of candidate sites presented by the contractor. Final project
selection was made after on-site field inspections by EPA personnel. These
sites had been operating continuously for a minimum of 30 years, and one had
been in service for as long as 50 years. Two of the sites were applying pri-
mary treated wastewater to infiltration beds, and two were applying secondary
treated wastewater. This allowed a reasonable comparison of the treatment
efficiency of two stages of pretreatment. None of the systems were chlori-
nating the treated effluent prior to application to infiltration basins.
Hollister, California
The project site is located in the San Juan Valley 35 kilometers (km)
[22 miles (mi)] inland from Monterey Bay and about 144 km (90 mi) south of
San Francisco. The City of Hollister and surrounding community collect sewage
from a population of about 10,000. This collected sewage is piped to the
treatment system which is located about 1.6 km (1 mi) west of the city, 150
meters (m) (500 ft) south of the San Benito river bed. A summary of general
operating information for the systems at all four study sites is shown in
Table 2.
Preapplication treatment at the Hollister wastewater treatment facility
consists of primary clarification. Sludge from the clarifier is regularly
drawn off and stored in a converted Imhoff tank, then periodically applied to
sludge drying beds which are operated independent of the rapid infiltration
basins.
Part of the influent wastewater flow is equalized in an excavated earthen
reservoir before entering the head works and clarifier while the remainder is
pumped directly to the clarifier. Wastewater is later pumped from the equalizer
reservoir to the clarifier during the early morning hours when the flow into
the equalizer is lowest.
-------
TABLE 1. INITIAL SITE SCREENING CRITERIA
ITEM
CRITERIA.
00
Period of operation
PreappHcatlon treatment
Wastewater sources
Wastewater flowrate
Wastewater application rate
Depth to groundwater
Control site
Availability of data
Operation practices
>10 years
Remain unmodified for at least 10 years
Domestic and commercial9
>4.3 1/s
>6 m/yr
>3 m and <30 m
Comparable geohydrologic characteristics within
1.6 km of site
Historical wastewater and groundwater quality must
be available for comparison purposes
1. Wastewater application to the spreading basins
must be intermittent^
2. Sludge must never have been applied to the
spreading basins0
3. Soil conditions in the basins should not have
been altered drasticallyd
a. Industrial wastewater, in small amounts, resembling municipal wastewater
is acceptable.
b. Systematic flooding and drying over several days with multiple
Independent basins.
c. Constituents are generally much more concentrated in sludge than
in wastewater.
d. Surface disking or scarifying to restore infiltration is normal
and acceptable.
1/s x 0.0228 = Mgal/d
m x 3.281 = ft
km x 0.621 = mi
-------
TABLE 2. OPERATING FACTORS FOR EXISTING RAPID
INFILTRATION SYSTEMS
FACTORS
Preappli cation
Treatment
Ground water
level, m below surface
Hydraulic conductivity cm/hr
Infiltration area, ha
Length of
operation, yr
Total infiltration
beds
Annual Hydraulic loading
rate m/yr
Average daily
flow, 1/s
Industrial influence
HOLLISTER
CALIFORNIA
Primary
6.8-9.2a
6.4-12.8
8.8
30
20
15.4
43.8
YesD
LAKE
GEORGE
NEW YORK
Secondary
1.3-7.0a
2.9d
2.2
38
21
44.7
30.8
No
VINELAND
NEW JERSEY
Primary
1.0-3.5a
15. 8d
36
50
32
11-21.4
21 5r
Yesc
MILTON
WISCONSIN
Secondary
2.0-3.0a
50e
0.2
40
1
110-224
14.5
No
a. Shallow water table created by infiltrating effluent.
b. Slaughterhouse = .9 1/s and paper recycle = 11.0 1/s.
c. Food processing wastewater is about 50% of Landis sewage authority flow.
d. Lowest value reported at the treatment site.
e. Estimated based on horizontal permeability.
-------
The clarified effluent flows by gravity to the 20 individually controlled
infiltration basins. The total basin area is 8.8 hectares (ha) [21.7 acres
(ac)], and the individual basins range from 0.3 to 0.7 ha (0.6 to 1.8 ac).
The basins are normally filled to depths of 30 cm (12 in); the duration of
flooding is controlled by the basin area. Two basins are usually flooded
simulatneously 8 to 10 hours each day if the daily flow exceeds the capacity
of any one basin. The interval between wastewater applications ranged from
14 to 21 days.
The interval between wastewater applications ranged from 14 to 21 days
during the study period. The length of time a basin was flooded and the
interval between applications decreased in the cooler and wet winter months
by 25 to 30 percent.
The effluent from the pretreatment facility was completely detained at
the infiltration basin to which it was applied, leaving the site only by per-
colation and evaporation. The initial percolation rate of effluent was
greater at the beginning of flooding than during the final stage because of
the initial dryness of the soil and because clogging of the soil surface by
effluent solids gradually reduces the infiltration rate. The average
infiltration rate during the study period was 10 cm/day (4 in/day) when the
basins were flooded to a depth of 30 cm (12 in).
The infiltration basins are located on alluvial deposits of the San
Benito River characterized by the Soil Conservation Service (SCS) as Metz
sandy loam. The permeability of this soil is reported to range from 6.4 to
12.8 cm/hr (2.5 to 5 in/hr) in the upper 3 m (10 ft) and 12.8 cm/hr to 25.4
cm/hr (5.0 to 10.0 in/hr) through the next 7.5 m (25 ft). The water table of
the regional ground-water aquifer occurs at a depth of 20 m (65 ft) at the
site. However, the saturated zone beneath the infiltration basins occurs at
depths of 2 to 3 m (7 to 10 ft).
Lake George, New York
The Lake George sewage treatment plant was originally constructed in
1936, put into operation in 1939, and has operated continuously since that
time. The plant is located approximately 1.6 kilometers (km) (1 mi) south-
west of Lake George in the southeast corner of the -New York State Adirondack
Park. The treatment plant receives influent from two sourcesLake George
Village and the town of Lake George. The plant was originally built to
treat flows in the summer that were three times larger than winter flows.
However, today the ratio is approximately two to one. Two Parshall flumes
record the inflow from the village and town pumping stations.
Primary treatment consists of one circular Imhoff Tank and two mechani-
cally cleaned circular settling digestion tanks. Secondary treatment is
accomplished with two high rate rotating arm trickling filters in summer and
one covered standard rate fixed nozzle sprinkling filter in winter. Two
mechanically cleaned rectangular settling tanks and two circular settling
tanks are used to complete secondary treatment. After secondary sedimentation,
the treated sewage is discharged directly to the infiltration beds. A
summary of general operating information for the system is presented in
Table 2.
10
-------
The treated sewage is carried by gravity to the 14 northern (lower)
beds while the sewage is pumped to the newer seven southern (upper) beds.
Normally, two beds are flooded simultaneously, one lower bed and one upper
bed. Dosing is changed at approximately 8:00am and 4:OGpm; thus, the entire
day's flow is discharged to four beds. The applied effluent takes from 1/2
to 3 days to infiltrate below the soil surface, depending primarily on the
size, age, and condition of the bed. The newer beds have higher infiltration
rates than the older beds. Beds are maintained by removal of surface mats
of entrapped suspended solids which inhibit infiltration. The cleaning
schedule is not fixed, but based entirely on operator experience, availability
of the bed for drying, and availability of plant personnel.
The infiltration beds in the Lake George system are located on delta
deposits left by Pleistocene glaciation. These deposits are part of the
Newland soil series which is composed of relatively fine uniform sands, low
in organic matter. The average vertical permeability is 2.8 cm/hr (1.1 in/hr).
Soil depth in the vicinity of the infiltration basins range from 2 to 9 m (7
to 30 ft) deep while the unsaturated zone ranges from 1.2 to 6 m (4 to 20 ft).
Vine!and, New Jersey
The city of Vine!and and the sewage treatment works are located in
Cumberland County in southern New Jersey about 24 km (15 mi) from the north-
eastern shore of Delaware Bay which flows into the Atlantic Ocean. The
Vineland sewage treatment works is actually composed of two separate treat-
ment facilities operated by Borough of Vineland and the Landis Sewerage
Authority. These systems are operated continuously and serve the city of
Vineland and the surrounding area. The two facilities were considered as one
test site even though historical length of operation and treatment of effluent
are different.
The Vineland plant provides primary treatment of an average daily flow of
35 liters/second (1/s) [8x10 gallons per day (gpd)]. Primary treatment
includes coarse screening of the influent and primary settling before being
discharged directly to Basin I which contains a series of 19 infiltration
beds.
The Landis Sewerage Authority's treatment plant received an average
daily flow of 180 1/s [4.1 million gallons per day (mgd)] and provides primary
treatment consisting of pre-aeration and primary settling before the treated
effluent is discharged directly to two large infiltration basins (Basins II
and III). A summary of the operation of the combined systems is presented
in table 2.
Basin I has been operated since 1928 by the Borough at Vineland. The
Landis Sewerage Authority Basin II, containing a series of eight infiltration
beds, has been in operation since 1948 and Basin III, containing a series of
six infiltration beds, has been in operation since 1974. Basins I and II
were operated by continuous flooding from the beginning of their use until
April, 1973, when operation was changed to intermittent flooding. The
11
-------
present operation is to flood the beds one at a time at each facility until
the level of a bed reaches about 45 cm (18 in). When this level is reached,
flow is diverted to the next bed. The length of time each bed is flooded
varies, depending on volume of flow, bed size, and the soil percolation rate.
Normally after each loading, when the wastewater has percolated through
the soil and the bed has dried, the entire bed is scarified to loosen the
soil and restore the permeability of the top layer. Every six months each
bed is plowed to a depth of 30 cm (12 in) to mix the soil and organic matter
that has accumulated during percolation of wastewater.
The soils at the Vineland study site have been formed from materials of
glacial origin and are characteristically deep, sandy, and well drained.
Around the infiltration basins, soils belonging to the Eavesboro, Lakewood,
and Lakehurst series are dominant. These series have relatively high
permeabilities ranging from 5 to >20 cm/hr (2 to >6 in/hr), and low soil
water holding capabilities. These soils are well drained, loose, and low in
fertility and organic matter.
Milton, Wisconsin
The project site is located in Rock County, Wisconsin, which is in the
south central part of the state about 64 km (40 mi) southeast of Madison.
The Milton wastewater treatment plant gives secondary treatment followed
by discharge to three lagoons constructed in series. The first two lagoons
provide only detention time while the third lagoon is maintained as an infil-
tration basin. The lagoons are continually flooded except as required to
maintain infiltration, and the third lagoon is drained and cleaned by removal
of the top few cm every two years. This material is replaced with fresh soil
excavated from areas near the basin. The infiltration basin covers an area
of approximately 1860 m (20,000 ft ) with an average annual inflow of 14.5 1/s
(3.3x10 gpd) which results in a loading rate of 0.67 m (2.2 ft) per day.
The secondary treatment facilities evolved from the original facilities
constructed in 1939 in an abandoned gravel pit. These original facilities
consisted of activated sludge treatment and aerobic digestion using mechanical
aerators. From 1939 to 1957 the effluent was passed through two rapid sand
filters and then discharged thru dry wells in the surrounding gravel. The
first lagoon in the series presently in use served for alternate disposal at
the time. Sand filtration was discontinued in 1957, and the single lagoon was
converted to two lagoons in series. In 1962, the original mechanical aerators
were replaced with compressed air aeration, and the mechanical bar screen was
replaced with a barminutor. Other additions included a blower building,
primary tank bypass line, and larger capacity return sludge pumps. At about
the same time, the third basin was constructed in series with the first two
lagoons and used for infiltration. Operations have continued, as described,
through the reported study period. General operating criteria are presented
in Table 2.
12
-------
The infiltration basin is located on Pleistocene glacial moraine sedi-
ments, 25 to 27 m (82-90 ft) deep at the site. These sediments generally
consist of unsorted mixtures of clay, silt, sand, and gravel with stratified
accumulations of sand and gravel. Beneath these glacial deposits are gray
to yellow-gray dolomite, sandstone, and green shale of the PIattevilie-Galena
unit of Ordovician age. The water table below the bottom of the infiltration
basin normally is about 2 to 3 m (7 to 10 ft) while the control ground water
saturated zones were a minimum of 15 to 18 m (50 to 60 ft) below ground level,
13
-------
SECTION 5
COMPARISON OF APPLIED AND RENOVATED WASTEWATER
The four rapid infiltration systems reviewed in this summary have been in
operation from 30 to 50 years, apparently without significant adverse environ-
mental effects. Comparison of the summarized data from each of these sites
along with some general interpretations of these data should give the de-
signers and planners a better insight to the treatment capability of rapid
infiltration systems where either primary or secondary treated effluent is
applied to the infiltration basins. The ability of these high rate systems
to remove most pollutants to near acceptable drinking water standards for
both levels of pretreatment lends credibility to the use of such systems
as a viable alternative for municipal wastewater treatment.
The four existing systems were historically constructed and operated
based on hydraulic acceptability of the soil at specific sites, and then the
systems were expanded as the volume of treated sewage required for infil-
tration increased. Little correlation can be made between those four sites
with respect to their comparative treatment efficiency and systems operation
due to these progressive changes. Until recent years, these four systems
were primarily operated with disposal of wastewater in mind with little
regard to treatment. Therefore, loading frequency and system management
were quite variable between the systems. As noted in Section 4, systems
operation varied from continuous flooding of primary treated wastewater
at Vineland, New Jersey, and secondary treated wastewater at Milton, Wisconsin,
to a very well scheduled cycle of flooding and drying of secondary treated
wastewater at Lake George, New York, and primary treated wastewater at
Hollister, California. All four of these systems were operated year-round,
three of them during sub-freezing winter weather. Enough heat is normally
retained in the treated wastewater (e.g. , the Lake George system) with the
pretreatment utilized to thaw the soil and ice remaining from the previous
flooding cycle. However, it was reported that during winter the hydraulic
conductivity of infiltration basins is reduced by as much as 25 to 30 percent
even at Hollister, California, which has the mildest climate of these four
sites. This change corresponds to expected changes due to temperature effects
on the viscosity of water. Data of the difference in hydraulic conductivity
between winter and summer operations at the other three sites were not
reported.
It is common knowledge to researchers conducting studies of land appli-
cation systems that nitrification and denitrification during cold winter
months is normally reduced considerably from summer operations. This is
14
-------
primarily because the organism activity is reduced by cold temperatures.
Therefore, during winter operation a greater mass of nitrogen is infiltrated
into the ground water. The data presented in this summary are averages for
the entire study period, except for Lake George which reported the range
of each parameter for seasonal extremes during the study period. The data,
therefore, do not show seasonal changes as discussed above for all four sites,
and reflections of seasonal influences cannot be verified.
The most practical way to evaluate the treatment efficiency of these
systems is to compare the chemical characteristics of the wastewater applied
to the infiltration beds to water collected from the top 2 to 20 m (6 to 60
ft) of ground water below or near these beds as shown in Table 3. These data
should then be compared to data collected from the native ground water at
remote control wells located up-gradient and out of the influence of the
infiltration beds (Table 4). Comparison of the data from these three points
of interest accurately reveals which parameters are receiving treatment and
which are reaching the ground water essentially untreated. In a review of
the reports of individual systems, it can be seen that down-gradient wells
indicate certain parameters remain elevated above those of the up-gradient
control wells for several hundred feet before differences in concentration are
insignificant.
In evaluating the data, the reviewers must keep in mind there are a
number of conditions that influence the treatment efficiency at all of the
sites such as effluent pretreatment, rate of infiltration, soil chemical
characteristics, and depth of unsaturated soil. The most obvious effect
of pretreatment is shown in Table 3 comparing the chemical oxygen demand
(COD), biochemical oxygen demand (BOD), total organic carbon (TOC), total
nitrogen (total-N), organic nitrogen (organic-N), and nitrate nitrogen
(nitrate-N) for the four systems. The Hollister and Vineland sites receive
only primary treatment before the effluent is applied to the infiltration beds
while the other two sites receive secondary treatment. There is a significant
contrast in the concentration of constituents applied to the beds. Com-
parison of these parameters in Table 3 all indicate water recovered from
shallow ground water wells beneath or near the infiltration basins is greatly
improved from that applied to the basins from both types of pretreated
effluent. However, the percent removal is significantly higher in systems
applying primary treated wastewater than systems applying secondary waste-
water.
When the quality of the shallow ground water wells is compared to that of
the control wells (Table 4), it is obvious that these constituents are not
completely removed before entering the upper levels of the ground water.
Even though complete treatment is seldom accomplished, it can be seen by
review of the entire list of parameters that drinking water quality standards
are generally met for both levels of pretreated effluent with the exception of
coliforms.
The removal of nitrogen from the wastewater in land application systems
is a multi-step operation. Oxidation of the organic and ammonia nitrogen is
the first step in the process. Comparsion of the data in Tables 3 and 4
15
-------
TABLE 3. COMPARISON OF APPLIED WASTEWATER VS. SHALLOW GROUND WATER
HOLLISTER, CA
COD
BOD
TOC
Total-N
NH--N
Kjgl-N
Organic-N
Sol. Orq.-N
Nitrate-N
Total -P
Ortho-P
PH
TDS
SS
VSS
Conductivity
(ymhos/cm)
Alk
Total-Col i
(£/100ml)
Fecal-Col i
(#/100ml)
Ma
K
Ca
Mg
B
Effluent3
706
220
248
40.2
25.3
14.5
0.43
12.4
10.5
7.3
1208
274
1790
446
27. 6x1 O6
12.4xl06
262
12.9
54
64
1.4
Shallow
Ground
Watera
50
13
11
1.7
<0.1
1.0
0.6
9.6
8.7
7.7
1275
10
1828
522
0.231xl06
f*
0.186x10°
261
14.9
107
74
1.2
LAKE GEORGE, NY
Effluentb
23-40
12.2-33.5
1.8-15.8
11.7-27.6
0.5-5.9
2.0-4.5
0.4-2.0
6.6-7.6
175-240
80-1 2*
.3-.7xlOR
.225-.525xl06
10-17
4-6
14-22
5-6.1
Shallow
Ground
Waterb
0.8-3.0
.24-7.5
.024-0.12
.10-. 49
.14-7.0
.32-1.08
.004-. 43
6.5-7.4
84-175
43.5-94
_
<1
5.7-7.4
0.87-1.2
10.2-14.6
3.7-7.2
VINELAND, NJ
Effluent0
372-287
154-149
108-82
40.4-38.8
17.2-19.0
23.1-18.5
16.8-21.1
0.1-1.3
9.3-9.0
4.8-4.8
6.5-6.8
214-304
43-41
27-29
5^8-602
115-130
TNTC
TNTC
60-75
14.5-10.4
12-11
4-4
.311-. 534
Shallow
Ground
Water0
75-99
6.5-12
23.7-28.6
11.6-17.5
10.6-11.0
1.49-0.06
1.54-3.8
6.6-6.9
266-381
468-688
119-187
TNTC
TNTC
38-102
10.6-13.3
14-9
5-4
.332-. 593
MILTON
Effluent3
84
28.0
36
30.8
19
6.5
1.3
5.3
8.2
7.0
7.7
728
29
21
1431
410
TNTC
TNTC
147
14
70
37
.681
, WI
Shallop
Ground
Water3
36
5.2
_
21.8
20.3
-
0.7
0.8
-
5.8
7.5
703
-
-
1379
427
TNTC
476
13.5
13.5
71
40
.628
(continued)
-------
TABLE 3. (continued)
HOLLISTER
Effluent9
Cl 284
Sulfide
Sulfate 213
As <0.01
Fe 0.39
Hg <0.001
Mn 0.070
Ni 0.051
Pb 0.054
Sc <0.001
Zn 0.048
Ba <0.13
Cd <0.004
Co <0.008
Cr <0.014
Cu . 034
Al
Endrin
(ng/1)
Lindane
(ng/1)
Methoxychlor
(ng/1)
Toxaphene
(ng/1)
2,4-D
(ng/1)
2,4,5-TP
silvex (ng/1)
1 = all values in ng/1 unl
, CA LAKE GEORGE, NY
Shallow Shallow
Ground Ground
Watera Effluentb WaterD
292 32-101 29.5-64
247
<0.01 <
0.51- 0.2-.85 .43-8.25
<0.001 <
0.62
0.16
0.08
<0.001
0.090
<0.15
0.050
0.012
<0.017
0.036
ess otherwise noted. b=
a = mean values, Effluent is input to land treatment c=
system.
VINELAND, NJ
Shallow
Ground
Effluent0 Water0
57-77 27-79
<.!-<. 1 5.7-16.5
14-24 22-19
.005-<.005 .01-. 008
1.5-1.2 12.6-9.0
.001-<.001 .001 -.001
<.06-<.07 .13-. 09
<.!-<.! .l-.l
<.!-<.! .l-.l
<.01-<.01 .01-. 01
.127-. 121 .024-. 043
<.02-<.02 .02-. 02
<.05-<.05 .05-. 05
<.02-<.02 .02-. 02
.058-. 077 .023-. 02
.6-. 6 .5-. 5
<.03-<.03 .03-. 03
2830-1227 453-1173
<.01-<.01 .01-. 01
<.!-<.! .l-.l
9.5-10.5 16.4-13.0
72-72 26.8-120
mean seasonal range
Vineland and Landis
respectively.
MILTON
Effluent9
200
0.1
51
.011
.50
.001
.10
.10
.2
.012
.078
.02
.05
.05
.10
0.3
.03
41
0.01
0.1
53.8
16.2
sites mean
, WI
Shallow
Ground
Water3
179
0.5
32
.033
2.0
.001
.29
.10
.2
.005
.254
.02
.05
.05
.10
1.0
.03
157.6
0.01
0.1
92.4
41.2
values,
-------
TABLE 4. COMPARISON OF SHALLOW GROUND WATER** TO CONTROL WELLS2
CO
HOLLISTER, CA
COD
BOD
TOC
Total -N
NH3-N
Kjel-N
Organic-N
Sol. Org.-N
Nitrate-N
Total-P
Ortho-P
PH
TDS
SS
vss
Conductivity
(jamhos/cm)
Alk
Total -Col i
(#/100ml)
Fecal -Coli
(#/100ml)
Control
Well a
16
3
<1
2.1
<0.1
-
<0.5
-
1.9
0.1
0.04
7.8
820
_
_
1148
425
<2
0
Shallow
Ground
Water3
50
13
n
1.7
<0.1
-
1.0
-
0.6
9.6
8.7
7.7
1275
10
-
1828
522
c
0.231xlOb
c
.186x10b
LAKE GEORGE, NY
Control
Wellb
0
1.2
-
.55-1.7
.035-. 087
.53-1.66
-
-
.01 9-. 053
.078-. 347
<.001-.001
6.6-7.8
115-145
-
-
mm
61-93
<1
<1
Shallow
Ground
Waterb
0.8-3.0
_
0.24-7.5
.024-0.12
.10-. 49
-
-
.14-7.0
.32-1.08
.004-. 43
6.5-7.4
84-175
-
-
.
43.5-94
-
<1
VINELAND, NJ
Control
Wella
7
1.1
_
3.7
0.1
_
_
1.8
1.0
0.03
5.2
40
_
_
64
7
51
0
Shallow
Ground
Waterc
75-99
6.5-12
_
23.7-28.6
11.6-17.5
_
-
10.6-11.0
1.49-6.06
_
1.54-3.8
6.6-6.9
266-381
-
_
468-688
119-187
TNTC
TNTC
MILTON, WI
Control
Well a
18.0
3.9
_
_
0.2
-
-
0.5
6.7
_
0.05
7.7
405
-
-
677
282
17
0
Shallow
Ground
Water3
36
5.2
_
-
20.3
-
-
.7
.8
-
5.8
7.5
703
-
-
1379
427
TNTC
476
(continued)
-------
TABLE 4. (continued)
HOLLISTER, CA
Na
K
Ca
Mg
B
Cl
Sulfide
Sulfate
As
Fe
Hg
Mn
Ni
Pb
Sc
Zn
Ba
Cd
Co
Cr
Cu
AT
Control
Well9
112
4.1
47
89
0.7
76
_
184
<0.01
0.03
<0.001
0.01
0.039
0.012
<0.001
0.05
<0.07
<0.003
<0.006
0.032
0.025
-
Shallow
Ground
Water9
261
14.9
107
74
1.2
292
-.
247
<0.01
<0.51
<0.001
.62
.16
.08
<0.001
0.090
<0.15
0.050
0.013
<0.017
0.036
-
LAKE GEORGE, NY
Shallow
Control Ground
Wellb Waterb
5.7-7.4
0.87-1.2
10.2-14.6
3.7-7.2
- -
32-64.4 29.5-64
_ _
.. ^
_ _
2.24 .43-8.25
-
-
_ _
_
_ _
_ _
_
_ _
_ -
_ _
_ _
-
VINELAND, NJ
Control
Well3
2
1.2
3
1
.113
4
0.1
9
.006
1.6
.001
.11
.10
.10
.01
.043
-
.02
.05
.02
.02
1.8
Shallow
Ground
Water0
38-102
10.6-13.3
14-9
5-4
.332-. 593
27-79
5.7-16.5
22-19
.01-. 008
12.6-9.0
.001
.13-. 09
<.!-.!
.l-.l
.01-. 01
.024-. 043
-
.02-. 02
.05-. 05
.02-. 02
.023-. 02
.5-. 5
MILTON,
Control
Well9
9
2.1
84
37
.248
23
0.1
43.0
.005
.5
.001
0.10
.10
.2
.05
.628
-
.02
.05
.05
.10
1.0
WI
Shallow
Ground
Water9
135
13.5
71
40
.628
179
.5
32
.033
2.0
.001
.29
.10
.2
.005
.254
-
.02
.05
.05
.10
1.0
(continued)
-------
TABLE 4. (continued)
HOLLISTER, CA
LAKE GEORGE, NY
VINELAND, NJ
MILTON, WI
ro
o
Endrin
(ng/1)
Lindane
(ng/1)
MethoxychTor
(ng/1)
Toxaphene
(ng/1)
2,4-D
(ng/1)
2,4,5-TP
silvex (ng/1)
Shallow Shallow Shallow Shallow
Control Ground Control Ground Control Ground Control Ground
Well9 Natera Well" Materb Wella Water0 Well9 Water5
.03
21.3
0.01
0.1
10.4
185
.03-.03
453-1173
.01-.01
.l-.l
16.4-13.0
26.8-120
.03
7.4
.01
.1
31.0
76.8
.03
157.6
.01
.1
92.4
41.2
1 - averages of data from well sampled in the top 2 to 20 m (6 to 60 ft) of ground water. All values
in mg/1 unless otherwise noted.
2 - averages of data from deep aquifer wells sampled up gradient out of the influence of infiltration
beds. All values in mg/1 unless otherwise noted.
a - mean values
b - mean seasonal range
c - Vineland and Landis sites mean values, respectively.
-------
indicate Hollister and Lake George were effective in this oxidation process
but Vineland and Milton did not effectively oxidize the nitrogen. The
probable causes for this ineffective treatment is insufficient oxygen in the
soil profile. A high water table where oxygen could not enter the soil
profile during drying and long flooding cycles where the available oxygen is
consumed before a new supply is made available could cause this response.
Both of these conditions existed at Vineland and Milton. This problem can be
alleviated by changing the cyclic operation allowing additional drying time,
underdraining the soil profile to allow for additional storage of oxygen
between applications, and/or reducing the application rate.
The two major fractions of phosphorus involved in the land treatment
process are soluble and insoluble phosphorus. The insoluble phosphorus is
physically filtered by the soil and degraded biologically near the surface
forming soluble phosphorus compounds. These compounds combine with soluble
phosphorus initially in the wastewater to move with the water into the soil
profile. Once in the soil profile, two primary factors control the fate
of the phosphorus. First, each soil has an ultimate capacity to retain
phosphorus. When the capacity is reached, phosphorus will move through this
portion of the profile unattenuated. The second, and probably more important
factor, is the time available for the soil to react with the phosphorus.
Table 3 indicates the total capacity for phosphorus removal by the profile
has not been reached at any of the sites evaluated. Since the concentration
in the shallow ground water is consistently lower than the applied effluent,
the removal efficiency is quite variable due to differences in detention in
the profile. This aspect is discussed in more detail in Section 7. An
example of the soil's capability for phosphorus removal was demonstrated at
Lake George, which had the lowest infiltration rate of the four systems
evaluated and 38 years of operation. Lake George, which had an unsaturated
soil depth of 1.3 to 7.0 m (4 to 20 ft), showed approximately 80 percent
removal of both total and ortho-phosphate, indicating soil depth coupled with
detention time through the profile are necessary for good phosphorus sorption
efficiency.
The total dissolved solids (TDS) were quite variable between the sites.
Evaporation from the surface can be a factor in increasing the concentration
of ions remaining in the waste stream and will increase the TDS. However, in
the operation of rapid infiltration systems, concentration changes due to
evaporation should be minimal due to the high hydraulic loading compared to
evaporation. Evaporation from basins in high rate systems seldom exceed
1 cm/day, even in hot dry climates. Dissolution of minerals is a potential
source for an increase in the TDS, particularly when the carbon dioxide (002)
partial pressure is increased due to degradation of organic material. This
phenomenon was observed in Tables 3 and 4 at Hollister where calcium (Ca)
increased and was further verified by a decrease in calcium carbonate (CaCO-J
in the profile as dicussed in Section 6. Increases due to similar mech-
anisms were observed for iron (Fe) as discussed later. On the other hand,
data from Lake George indicated a significant retention of cations (Na, Ca,
K, and Mg). The reduction in cations corresponds to a reduction in TDS and
conductivity at the- site (Tables 3 and 4).
21
-------
Fecal coliform counts were generally reduced by two orders of magnitude
except in the Vineland system which applied primary effluent containing coli-
forms that were too numerous to count (TNTC) through soils only 1.0 to 3.5 m
(3 to 10 ft) thick (Table 3). Both total and fecal coliforms at this site
were also TNTC in the shallow ground water beneath the infiltration beds
indicating very poor treatment (Table 4). The Milton system with a similar
shallow water table also experienced high total coliform counts in the ground
water. However, total coliform counts in soil is not necessarily indicative
of pollution but more of biological activity.
Even though 6.8 to 9.2 m (22 to 30 ft) of unsaturated soil profile exist
at Hollister, significantly higher total and fecal coliform counts were found
in the shallow ground water as compared to the control wells (Table 4). These
data indicate that renovated water should be allowed to move laterally a
sufficient distance through the aquifer to obtain coliform-free renovated
water.
Thirteen metals were measured on the applied wastewater and ground water
beneath the basins at all sites and control wells with the exception of Lake
George where only Fe was measured. All these metals were in very low concen-
trations with the exception of Fe, manganese (Mn), and zinc (Zn). The Fe
data indicated significant leaching from the soil profile at all four sites
as shown in Tables 3 and 4. The behavior of Mn at the three sites where it
was measured was very similar to that of Fe. Data indicate Zn was leached
from the soil 'at Hollister and Milton, but was accumulated in the upper soil
profile at Vineland. Soil leaching of arsenic (As) was evident at Milton when
the data of the shallow ground water wells are compared to the control wells.
The mechanism for leaching of these specific metals is reported to be the
reduction of oxides and hydroxides of the complex in reducing environment to
more soluble forms and/or chelation by the organic constituents remaining in
the water (2).
Six pesticides were measured in the applied effluent and ground water at
the Vineland and Milton sites. These include endrin, lindane, methoxychlor,
toxaphene, 2,4-D, and 2,4,5-TP silvex. Endrin, methoxychlor, and toxaphene
were not found above the respective detectable limits of <0.03, <0.01, and
<0.1 nanogrmas/liter (ng/1) at either site.
The herbicides 2,4-D and 2,4,5-TP silvex were found at concentrations
in the same range as encountered in the control ground water. The National
Interim Primary Drinking Water Regulations allow maximum concentrations of
100,000 ng/1 and 10,000 ng/1 for 2,4-D and 2,4,5-TP silvex, respectively, in
domestic water supplies (2). The measured values of these herbicides were
insignificant compared to the standard and should be of no concern.
Lindane concentrations in the ground water beneath the infiltration beds
ranged from maximums of 6480 ng/1 to 1360 ng/1 for the Vineland and Milton
sites, respectively. The average concentration of lindane in the ground water
at these sites was 453 to 1173 ng/1 beneath the Vineland basins and 157.6 ng/1
beneath the Milton basin. The National Interim Primary Drinking Water
22
-------
Regulations standard maximum contaminated level for lindane is 4000 ng/1 (2).
The only system in which there appeared to be any treatment of the pesticides
or herbicides was at Vineland Basin I in which about 85 percent of the lin-
dane was removed. This system had been in operation for 50 years but had
no record of how long wastewater, high in lindane, was applied. Basin II,
which had been in service for about 30 years, had a reduction of only about
5 percent of the lindane as a result of the treatment process (Table 3).
Lindane reaching the ground water remained significantly higher than the
control wells.
23
-------
SECTION 6
OBSERVED CHANGES IN SOIL CHEMICAL PROPERTIES
Describing the response of the pH at a rapid infiltration system is
extremely complex. There are several factors which tend to increase the
acidity and several factors which tend to decrease the acidity. In the pro-
cess of organic matter decomposition, both organic and inorganic acids are
formed. The simplest and, perhaps the most widely found, is carbonic acid
(HpCOo) which results from the reaction of C02 and water. Soils with high
water contents and ample supply or organic carbon tend to have elevated CO^
partial pressure and greater potential to form acids. Since organic matter
decomposition is a biological process, temperature of the system indirectly
affects the pH. The organic carbon is also attacked by fungi which have
among their metabolic end products strong complex organic acids. On the
other hand, supplying dissolved solids in the wastewater changes the ionic
balance supplying metals to the exchange sites replacing hydrogen. This has
a tendency to increase the pH of the soil.
The actual pH of the operating systems was not measured. Slight changes
in the C02 partial pressure will result in significant changes in the measured
pH. The reported results indicate changes in base saturation of the soil
profile rather than hydrogen ion activity" of the operating system. Figure 1
shows the changes that were observed at the Hollister, California, and
Vineland, New Jersey, sites. The results indicate the systems are reequili-
brating near neutrality throughout the 500 cm (17 ft) depth of monitoring.
Metals are known to be adsorbed by the soil matrix and to precipitate
forming relatively insoluble compounds. In general, the more acidic the
system, the more soluble the compounds. For this reason, toxicity to metals
tends to occur more frequently in acidic conditions, and the mobility of the
metal is greater under acidic conditions.
Two metals were monitored which showed significant accumulations in the
soil profile, zinc (Zn) in Figure 2 and copper (Cu) in Figure 3. In both cases
there is significantly more accumulation at the surface than there is with
increasing depth. Also, the more acidic sites show more accumulation with
depth. These observations are consistent with theoretical expectations,
and one would not anticipate en-countering difficulties with metal toxicity
unless acidic conditions develop or occur during system operation.
In some respects, phosphates behave similarly to metals in that sorption
and precipitation reactions are involved. The major difference is that
maximum solubility of phosphate compounds occur near neutral conditions. One
24
-------
Q_
hJ
Q
200-
300L
LEGEND
HOLLISTER VINELAND
__._- CONTROL SITE*
------ TREATMENT SITE
Figure 1. pH comparisons in the soil profile.
100'
X
O-
UJ
0
200
300
ZINC (pg/g)
5 10 15 20 25
/ LEGEND
'HOLLISTER VINELAND
---- CONTROL SITE
- - - - TREATMENT SITE
Figure 2. Zinc concentrations in the soil profile,
25
-------
COPPER (pg/g)
0 5 10 15
100
£L
Ill
°aoo
300
LEGEND
HOLLISTER VINELAND
. ---- CONTROL SITE
, - -«- - TREATMENT SITE
Figure 3. Copper concentrations in the soil profile.
Copper concentrations at the Vineland control
site were below detectable limits (<2yg/g).
50
100
150
200-
250
300
TOTAL P HOLLISTER (yg/g)
2000
(' LEGEND
/ HOLLISTER VINELAND
1 __.-- CONTROL SITE
- - - - TREATMENT SITE
I
50 100 60 2OO 250
TOTAL P VINELAND (pg/g)
Figure 4. Total phosphate comparisons in the soil profile.
26
-------
would, therefore, expect, under similar loading conditions, greater pene-
tration in neutral soils. Figure 4 shows the total phosphorus sorbed or
precipitated in the solid phase as a function of depth for the two sites
monitored. Each show significantly more accumulation at the surface than at
greater depths. Both also show significant accumulation throughout the depth
monitored. Since phosphorus is not a toxic compound, significant accumu-
lations should not be considered detrimental to future system operation.
Since the soil does have a finite capacity to react with phosphorus, the
depth of penetration can be expected to increase with time and the amount
entering the receiving body to increase once the capacity of the surface soil
is reached.
In those soils where CaCCL is present, the increased C02 partial pressure,
caused by microbiological activity and high water contents, tends to dissolve
the CaCOo minerals. Major reductions throughout the profile were observed in
CaCCU at the Hollister site as shown in Figure 5. Hollister was the only site
where this mineral was reported. Losing the CaC03 would tend to increase the
IDS and hardness in the reclaimed water and reduce the capacity of the profile
to react with phosphorus where dicalcium phosphate dihydrate (CaHPO.«2H20) is
a dominant reaction product under calcareous conditions.
Organic nitrogen build-up was all in the surface layers of the soil as
shown in Figure 6. The only significant differences were in the surface most
sampling of the soil profile. This observation is most likely due to the
physical filtering of the suspended organic material which is high in organic
nitrogen. Once a system is abandoned one would anticipate this observed
increase in organic nitrogen to revert to background conditions due to
biological degradation of the organic materials.
SUMMARY
Long-term operation of rapid infiltration wastewater treatment systems
have significant impact on chemical properties of the soils where the waste-
water is applied. But the changes observed are not necessarily detrimental
to man's existence or to his environment. The following general observations
can be made:
1) Operation of the treatment system tends to drive the pH of the
soil toward neutrality.
2) Elements which form relatively insoluble compounds tend to
accumulate near the soil surface.
3) Organic materialswhich can be filtered from the solution accumulate
until they are degraded as evidenced by accumulation of organic
nitrogen.
4) Relatively soluble minerals such as CaC03 may be leached from the
profile..
27
-------
100
I
I-
a
LJ
a
200
300
CaCO, equivalent (%)
05 1.0 1.5 20 25 30
LEGEND
HOLLISTER
----- CONTROL SITE
- TREATMENT SITE
Figure 5. Calcium carbonate reduction in the soil profile.
ORGANIC-N HOLLISTER (ug/g)
O 400 800 1200 1600
100 f
O-
LU
200
300
LEGEND
HOLLISTER VINELANP
-- CONTKuu alTE
---- TREATMENT SITE
_1_
0 100 200 300 400 5OO
ORGANIC-N VINELAND
Figure 6. Organic nitrogen reduction comparisons in the soil
profile.
28
-------
SECTION 7
RATIONALE FOR RAPID INFILTRATION DESIGN
The design of a rapid infiltration wastewater system is based on deter-
mining the most limiting criteria. The design can be based on either
hydraulic loading limitations or based on water quality leaving the treatment
system. A preliminary rationale is presented for selected parameters. The
rationale is then tested at the four existing systems and design results
compared with actual practice.
HYDRAULIC LOADING
The Process Design Manual for Land Treatment of Municipal Wastewater (1)
bases rapid infiltration (RI) design on hydraulic acceptability. The four
parameters required to evaluate the suitability of the wastewater and deter-
mine the maximum hydraulic loading for hydraulic acceptability are:
/
1) the SAR of the applied wastewater
2) the saturated vertical hyraulic conductivity of the most
hydraulically restrictive soil layer
3) the BOD of the applied wastewater and
4) the total suspended solids (TSS) of the applied wastewater
The sodium adsorption ratio is defined by the equation:
Na+
SAR = (1)
. I Ca++ * Mg++ "
\1 2
where the concentrations are expressed in milliequivalents per liter. This
parameter is important because of its tendency to promote conditions favorable
to flocculation or deflocculation of the soil particles and thereby affects
the soil's capability to accept and transmit water. This property is more
important in fine textured soils than those normally associated with rapid
infiltration systems. In the design of rapid infiltration systems, the SAR
will be a severely limiting factor if the value is greater than 9 (1). A
value of 15 should be considered as a practical upper limit for the SAR.
29
-------
The most difficult parameter to accurately evaluate is the hydraulic con-
ductivity of the soil profile. The Process Design Manual for Land Treatment
of Municipal Wastewater (1) specifies design based on the hydraulic conductiv-
ity (sometimes called permeability or transmissivity) of the most restrictive
layer. A more rational approach would be to consider the geometric mean of
individual measurements of the hydraulic conductivity from the entire profile.
This would permit considering the effect of a small perched water table in-
creasing the hydraulic gradient across the restricting layer. Infiltrometers
are often used to estimate the hydraulic conductivity under the assumption
that when the infiltration rate reaches a constant value, gravity will be the
dominate force causing flow, thereby creating a unit gradient. This would
make the infiltration rate approximately equal to the saturated hydraulic
conductivity. Using the double ring cylindrical infi1trometer [as described
in Section C.3.1.4,3 (1)] or the U.S. Public Health Service percolation test
for septic tanks (7) can give highly erroneous results. The U.S. Public
Health Service percolation test does not consider the correct geometry for
large systems while the infiltrometer may lead to an inaccurate evaluation due
to spatial variability. Field studies have shown as much as an order of mag-
nitude variation within a single mapping unit on a single field using the
infiltrometer. The best currently available technique to evaluate the hy-
draulic conductivity is the sprinkler infiltromter [described in C.3.1.4.2
(!}] which averages a relatively large area. In this technique, water is not
ponded on the surface which eliminates the flow through worm holes and cracks.
These channels should not be considered when evaluating the hydraulic conduc-
tivity. Where a sprinkler infiltrometer is not available, the basin flooding
method [described in C.3.1.4.1 (1)] should be used since it will give the
average for a reasonably large area. A maximum hydraulic loading of 10 percent
of the saturated hydraulic conductivity of the most restrictive layer has been
suggested [Figure 3-5 (1)]. This limit-can also be modified by the BOD and SS
of the applied wastewater. A maximum of 2.24 milligrams per square centimeter/
day (mg/cm/day) [200 Ib/ac/day] for BOD and SS is rarely exceeded [pg. 5-8 (1)]
With loadings in this range one would anticipate after passing thru 4.5 m (15
ft) of soil, the BOD and SS would be less than or equal to 5 mg/1 [pg. 2-4
(1)1.
This analysis procedure assures the designer that the soil will accept
the hydraulic load. However, no assurance is made from the analysis that a
ground water mound will not develop, thereby impairing the operation of the
system. Maintaining a minimum of 1.5 m (5 ft) of unsaturated soil profile is
recommended (1). Although higher water tables are permitted during flooding,
the designer must further evaluate the system's geohydrology and be assured of
adequate drainage, either by natural or man-made systems.
WATER QUALITY LIMITATIONS
The Process Design Manual (1) does not give the consulting engineer a pro-
cedure to estimate the percolate water quality. The manual only indicates 15
to 25 percent of the nitrogen will be lost. There are two ways to consider
nitrogen transformation. One is the form of nitrogen (nitrate, ammonia, or
organic-N) reaching the ground water or being discharged from the soil treat-
ment system. The alternate approach is to consider the total nitrogen dis-
charged to either the ground water or another receiving body of water.
30
-------
Loading to Obtain Nitrification
When the receiving body of water is ammonia limited, the maximum loading
rate is limited by the nitrification rate. Nitrification is a process of
enzymatic oxidation brought about by a specialized group of bacteria. Nit-
rifying bacteria are present in almost all soils and can remain active over a
wide range of moisture and temperature conditions. These autotrophic bacteria
obtain their energy by the oxidation of NH, to NO,. The bacteria are obligate
aerobes but are capable of functioning at oxygen concentrations significantly
lower than atmospheric levels. The complete reaction converting NH4 to NCL
takes place in two steps. Nitrosomonas and Nitrosoccoccus species are con-
cerned with conversion of ammonium to nitrite following the reaction:
2NHJ + 302 > 2N02 + 2H20 + 4H+ + Energy
The oxidation of nitrite to nitrate is controlled by Nitrobacteria species
following the reaction:
2N0
0
+ Energy
This latter reaction generally proceeds rapidly.
are, therefore, rarely found in soil solution.
Significant levels of N02
Nitrifying bacteria are extremely sensitive to their environment. Where
nitrification is the objective of the treatment process, activities promoting
nitrification should be encouraged. Since nitrification is a process of
oxidation, activities increasing soil oxygen should, up to a limiting reaction
rate, encourage bacterial activity. Drying or resting and cultivation are
recognized methods of promoting soil nitrification. The rate of nitrification
is directly proportional to the temperature. Increasing the temperature
increases the rate of nitrification throughout the temperature range found
at rapid infiltration treatment systems. Figure 7 shows the limiting rate of
nitrification, as a function of temperature assuming a neutral pH.
IZOO- DENITRIFICATION
10 20 30 40
TEMPERATURE (C°)
Figure 7. Nitrification and denitrification versus temperature.
31
-------
An approximate relationship to estimate maximum hydraulic loading and main-
tain the desired level of nitrification can be described by the equation:
LRNn - 30 (T-3)
N03 X (2)
where LRNQ = maximum daily loading expressed in cm/day to obtain
3 the desired level of nitrification
T = temperature of the applied wastewater in C°. This may
be approximated as mean air when the system is operated
continuously and lagoon pretreatment is not employed.
X = total kjeldahl nitrogen applied minus the design TKN
reaching the ground water or discharge point (measured
in mg/1).
Equation 2 is believed to be valid as long as the TKN reaching the ground
water is greater than or equal to 1 mg/1. Maintaining a concentration of
less than 1 mg/1 is extremely difficult and should not be considered practical
through rapid infiltration technology. The rate of nitrification is also
affected by pH. Below a pH of 4.5, nitrification ceases: However, since
most RI systems operate near neutral conditions, this rarely is a problem.
Loading to Obtain Denitrification
Nitrogen removal is more complex than nitrification alone, since both
nitrification and devitrification processes are involved. The denitrification
process is a biochemical, process that reduces nitrate and nitrite to ammonia,
nitrous oxide (O) and molecular nitrogen (N2). Denitrifying bacteria use
the oxygen to nitrate and nitrite in their oxidation of organic matter.
Several bacteria are capable of accomplishing this process. From a land
treatment point of view, the most important transformation involves the con-
version of nitrate to the gaseous forms N,,0 and N2« The process can be
described by the reaction:
CCH1<30C + 4NOI * 2N0 t + 6H00 + 6CO,
b ic. 0 6 f. c. c.
Denitrifying bacteria, like nitrifying bacteria, are sensitive to their
environment. Denitrification is inhibited by oxygen. Processes reducing soil
oxygen, up to a limiting rate, encourage denitrification. The more reducing
the soil environment is the greater the denitrification rate. In wastewater
studies, denitrification rates exceeding 20 kg/ha/day (110 Ib/ac/day) have
been observed.
Denitrification requires organic carbon. The potential carbon
limitation at a land treatment site using municipal wastewater has been
described by the equation:
AN = 1° (3)
32
-------
where AN = change in total nitrogen (mg/1)
TOC = total organic carbon in the applied wastewater (mg/1)
K = TOC remaining after wastewater treatment which is
typically 5 for municipal wastewater after passing
through 1.5 m (5 ft) of soil during summer operation.
This value increases during cold weather (8).
The coefficient two (2) in the equation was based on experimental data
where two grams of wastewater carbon were required to denitrify one gram of
wastewater nitrogen (9). For nitrate nitrogen, 1 mg/1 should be considered
the lowest concentration consistently obtainable through RI.
In addition to requiring sufficient carbon for the reaction, the denitri-
fying process may be rate limiting. Thus, even if sufficient carbon is available,
adequate treatment may not be achieved if sufficient time is not allowed for
the reactions to take place. Figure 7 graphically shows the effect of tem-
perature on the reaction. In terms of maximum loading rate to achieve the
desired denitrification, an approximate relationship can be given as:
LR - 40 (T-5)
LKDN Y (4)
where LRDN = maximum daily loading expressed in cm/day to obtain the
desired denitrification
Y = nitrate applied plus the nitrate converted from Equation 2
minus the nitrate entering ground water (all in mg/1).
Ammonia Volatilization Considerations
The ammonia reported in water quality standards is primarily in the form
of an ammonium ion. As the pH rises, there is a higher percentage of ammonia.
Management which promotes high pH followed by conditions conducive to volatiliza-
tion of the ammonia to the atmosphere such as sprinkler operation help to lower
the ammonia applied at the land treatment site. Volatilization plus denitri-
fication is estimated to be responsible for the removal of at least 15 to 25
percent of the applied nitrogen (1). A portion of this loss is due to the
volatilization process. But most researchers believe this to be of minor
importance when considering land application of municipal wastewaters following
conventional pretreatment.
Transport of Phosphate
Predicting changes in phosphorus concentration as wastewater passes
through the soil is quite complex. In RI systems the potential exists for a
significant level of phosphorus to remain in solution after passing through
the soil at the treatment site. The reaction in the soil system is primarily
one of chemical sorption and precipitation. A mathematical model has been
developed which permits predicting phosphate transport and transformation for
an undirectional flow path. This model reported elsewhere (10, 11, 12), takes
into consideration which compounds are formed and how fast they form. In
addition, the amount of adsorption is estimated.
33
-------
The design engineer will not ordinarily need to precisely estimate the
phosphorus accumulation in the soil profile or the concentration in the soil
solution. It is possible to make a "worst-case" estimate by assuming a
neutral pH and estimating the detention in the soil profile using the
equation:
x 1
where k = instantaneous rate coefficient (0.002 hr )
t = detention time = - :
where X = distance along flow path (cm)
0 = volumetric water content (cc/cc)
tf = infiltration rate during system operation (cm/hr)
C = total phosphorus concentration in applied wastewater (mg/1)
C = total phosphorus concentration at point X along the flow path
X
Using this abbreviated approach does not allow one to predict the ulti-
mate capacity of a soil. As the chemical composition of the soil changes, its
capacity to react with phosphorus changes. At this time, it is not possible
to obtain a definitive answer as to total capacity of the soil to react with
phosphorus. Since phosphorus is not considered toxic and systems have been
in operation for several decades without significant problems, this short-
coming is not considered serious.
RATIONALE EVALUATION
Many of the important parameters in the design of a rapid infiltration
system are included in this design rationale. The design yields maximum
loadings to achieve a predetermined level of treatment. The management of a
system will drastically affect its treatment efficiency. Therefore, adequate
design is no guarantee of adequate performance.. Several additional factors
must be included when making a complete design. The designer should also
consider:
1) preappli cation treatment
2) storage requirements
3) metal limitations
Several procedures can be used to evaluate the adequacy of a design
rationale. The qualitative "how well" a system works depends on who is
evaluating the system and the objectives of the treatment process. In each
of the following examples, a system was designed based on conditions at the
existing system. The design was then compared to system operation for an
evaluation of the rationale. The first consideration was to determine, based
on a design for hydraulic acceptability, whether the system should be capable
34
-------
of accepting the loading reported. After establishing the hydraulic accepta-
bility, a best estimate for the design loading to achieve required nitrogen
treatment was developed based on the concentration of nitrogen reaching the
water table. This is the same as determining the design hydraulic loading
assuming the discharge permit for nitrogen is the same as the concentration
observed at the ground water surface. In most cases where the treated
effluent reaching the ground water does not meet drinking water standards for
nitrogen, the system was reevaluated projecting modifications in system
operation required to achieve adequate treatment.
Since phosphorus is not a critical element, the approximation procedure
presented projects "worst-case" operation. Thus, we would anticipate observed
phosphorus concentrations to be less than or equal to the projections.
Hollister, California
The data required for a complete design are presented in Table 5.
Determination of Maximum Acceptable Hydraulic Load--
a) determine SAR
SAR = ^
\
Ca + Mg
SAR =
\
1.35 + 2.63
2
SAR = 8.1, therefore, not limiting.
b) determine limitation based on worst case hydraulic conductivity
6.4 _ x 24 x 0.1 = 15.36 cm/day
c) BOD limitation for hydraulic loading
2.25 mg/cm2-day = 1Q cm/day
220 mg/1000 cc 1U'" cm/aay
d) SS limitation for hydraulic loading
2.25 mg/cm2-day_ g 2« /d
274 mg/1000 cc *'*L cm/aay
Based on these calculations one would project the maximum hydraulic
load of 8 cm/ day. This is approximately twice the loading applied at the
Hollister site.
35
-------
TABLE 5. INPUT DATA REQUIRED TO EVALUATE HOLLISTER SITE (2)
Ca applied
Mg applied
Na applied
Hydraulic Conductivi
BOD applied
SS applied
Temperature mean air
TKN applied
TKN at ground water
NOl applied
NOg at ground water
TOC applied
54 mg/1 =
64 mg/1
262 mg/1
ty of Restrictive Layer
(design TKN)
(design NOj)
Total Phosphorus applied
Total Phosphorus at
7.7 m (design P)
1.35 meq/1
2.6 meq/1
11 .4 meq/1
6.4 cm/hr
220 mg/1
274 mg/1
18°C
39.8 mg/1
1 mg/1
0.4 mg/1
0.6 mg/1
248 mg/1
12.4 mg/1
8 mg/1
36
-------
Estimate Hydraulic Loading to Achieve Water Quality Arriving at the Ground
Water Interface 7-10 m (23-33 ft) Below the Surface for the Hollister Site--
a) Determine, based on carbon to nitrogen ratio, the maximum
nitrogen removal from Equation 3
AM TOC-5 _ 248-5 _ 243 _ ,91 ,,
AN = = ~~ * * 9' '
Based on this determination, there is sufficient carbon to reduce all nitrogen
forms to 1 mg/1 or less.
b) Determine loading rate to obtain required level of nitrifi-
cation from Equation 2
Max daily loading = 3Q T"3^
Since the TKN remaining after treatment is 1.1 mg/1,
&a = 11.6 «/day
the max loading to achieve nitrification is 11.6 cm/day.
c) Determine loading rate to obtain required level of denitrifi-
cation from Equation 4
Max daily loading = 40 !~^
Since the nitrate remaining after treatment is 0.6 mg/1,
38.$" M- 0.6 ' 13'
The maximum design loading to achieve denitrification is 13.5 cm/day
versus the 4 cm/day actually applied to the system. As stated earlier,
concentrations of less than 1 mg/1 are difficult to maintain.
Estimate Phosphorus Arriving at Ground Water Under Actual Application
Conditions--
From a previous study (12), the water content and infiltration rate
during system operation were estimates at 0.275 and 2.64 cm/hr, respectively.
To calculate the maximum estimated phosphorus at the ground water surface
735 cm (24 ft) below the surface, one uses Equation 5
n 002 hr'1 735 cm (0.275) = -, 12.4 mg/1
u.uu^ nr 2>64 cm/hr m c^
where C is the solution concentration at 735 cm (24 ft). C in then equal
to 10 mi/1. This is slightly greater than the 8 mg/1 reported. For more pre-
cise estimates using a computer model, see Enfield et al , (11, 12).
37
-------
Table 6 presents design loadings and concentrations versus reported
values. For Hoi lister, the design compares favorably with practice.
TABLE 6. DESIGN EVALUATION FOR HOLLISTER, CA (2)
Parameter
Hydraulic Acceptance cm/day
Max Loading for N removal * cm/ day
BOD mg/1
SS mg/1
Phosphorus mg/1
Design
8
11.6
1 5
1 5
10
Reported
4
4
6
N.R. **
8
* Concentrations less than 1 mg/1 not consistently obtainable thru RI.
Limitation is based on ability to nitrify.
** N.R. is not reported.
Mi 1 ton, Wi_sconsj_n
The data required for a complete design are presented in Table 7.
Determination of Maximum Acceptable Hydraulic Load--
a) Determine SAR
6.4
|Ca + Na jl.8 + 1.!
'M 2 N 2
SAR is, therefore, not limiting.
b) Determine limitation based on worst case hydraulic conductivity.
The vertical hydraulic conductivity was not measured at this
site. It was estimated as 25 percent of the horizontal
permeability or approximately 50 cm/hr. Then
50 cm/hr x 24 hr/day x 0.1 = 120 cm/day
38
-------
TABLE 7. INPUT DATA REQUIRED TO EVALUATE MILTON SITE (5, pg. 47.48.57)
Ca applied 70 mg/1 =
Mg applied 37 mg/1 =
Na applied 147 mg/1 =
Hydraulic Conductivity of restrictive layer*
BOD applied
SS applied
Temperature mean air
TKN applied
TKN at ground water (design TKN)
NOg applied
NO, at ground water (design NO^)
TOC applied
Total Phosphorus applied
Total Phosphorus at 8 m (design P)
1.8 meq/1
1 .5 meq/1
6.4 meq/1
50 cm/hr
28 mg/1
29 mg/1
11 °C
25.5 mg/1
21.0 mg/1
5.3 mg/1
0.8 mg/1
36 mg/1
8.2 mg/1
5.92 mg/1
* Estimated from horizontal permeability.
39
-------
c) BOD limitation for hydraulic loading
2
2.25 mg/cm -day _ Q« ^m/j-.,
28 mg/1000 cc ~ 80 cm/day
d) Suspended solids limitation for hydraulic loading
2
2.25 mg/cm -day _ 7ft rm/f4av
29 mg/1000 cc " 78 cm/day
Based on these calculations, one would project the maximum hydraulic
load of 78 cm/day. This is approximately 20 percent greater than actual
practice at the Milton site.
Estimate Hydraulic Loading to Achieve Water Quality Arriving at the Ground
Water Interface 8 m (26 ft) Below the Surface for the Milton Site
a) Determine, based on carbon to nitrogen ratio, the maximum
nitrogen removal possible from Equation 3
AM _ TOC-5 _ 36-5 _ 15.5 mg/1
AN n ~ n ~
Based on this calculation, there is not sufficient carbon to reduce the total
nitrogen below 15.3 mg/1 (the total nitrogen reported in the vicinity of site
is 21.8 mg/1). Regardless of management practice, this system should not be
capable of meeting drinking water standards for nitrogen, since ammonia and
organic nitrogen can be rapidly converted to nitrate in the presence of
oxygen.
b) Determine loading rate to obtain required level of nitrifi-
cation from Equation 2. Since the TKN remaining after treat-
ment is 21.0 mg/1,
Max daily loading = 30 j^T"3) = 25 S^-'zi.O = 53 cm/day
The design maximum loading rate to achieve nitrification is 53 cm/day.
c) Determine loading rate to obtain required level of denitrifi-
cation from Equation 4. Since (the nitrate remaining after
treatment is 0.8 mg/1,
«ax loading = «f^ - *°5'"'^ - 0.8 ' " <***
The design maximum loading to achieve denitrification is 27 cm/day. The
design rationale is conservative particularly in the area of denitrification.
The rate of denitrification appears to be significantly greater than pro-
jected by Equation 4.
40
-------
Estimate Phosphorus Arriving at Ground Water Under Actual Application Con
di tions--
The water content and application rate are estimated at 0.3 and
2.8 cm/hr, respectively (2.8 cm/hr was selected which is in the range re-
ported with continuous operation). Then estimating the phosphorus concen
tration at a depth of 800 cm (26 ft) using Equation 5
0.002 hr-. ln
X
one sees the estimated phosphorus concentration is 7 mg/1.
Table 8 presents design loadings and concentrations versus reported
values. The observed nutrients are consistently lower in concentration than
those projected from the design rationale.
TABLE 8. DESIGN EVALUATION FOR MILTON, WISCONSIN
Parameter
Hydraulic acceptance cm/day
Max loading
BOD mg/1
SS mg/1
Phosphorus
for N removal cm/ day
at 8 m (26 ft.) mg/1
Design
78
27*, 53**
i 5
1 5
7
Reported
67
67
5.2
N. R.
5.92
* Based on limitation for denitrification
** Based 9n limitation for nitrification
*** N. R. is not reported
Lake George, New York
The data required for a complete design is presented in Table 9.
Determination of Maximum Acceptable Hydraulic Load--
a) Determine SAR
41
-------
TABLE 9. INPUT DATA REQUIRED TO EVALUATE LAKE GEORGE SITE (3)
Ca applied
Mg applied
Na applied
18 mg/1
5 mg/1
13 mg/1
Hydraulic Conductivity of restrictive layer
BOD applied
SS applied
Temperature as monitored in
TKN applied
TKN at ground water (design
NOZ applied
NOZ at ground water (design
TOC applied
Total Phosphorus applied
ground water
TKN)
NO")
Total Phosphorus at 22 m (72 ft.)
0.4 meq/1
0.2 meq/1
0.6 meq/1
2.9 cm/hr
28 mg/1
N. R.**
11 °C
19.6 mg/1
0.3 mg/1
3.2 mg/1
3.6 mg/1
42 mg/1*
4 mg/1
0.3-1.1
* Estimated TOC « 1.5 x BOD
** N. R. is not reported
42
-------
SAR = Na = °'6 = 1
[Ca + Mg JO.4 + 0.2*
'Vj 2 ^2
SAR is, therefore, not limiting.
b) Determine limitation based on the worst case of hydraulic
conductivity.
2.9 cm/hr x 24 hr/day x 0.1 = 7 cm/day
c) BOD limitation for hydraulic loading is
2
2.25 mg/cm -day _ pn
28 mg/1000 cc " ou
d) Suspended solids limitation for hydraulic loading.
Suspended solids were not measured at the Lake George site. One would pro-
ject a maximum hydraulic loading less than or equal to 7 cm/day based on the
most restrictive layer approach. Apparently there is sufficient soil of a
greater permeability to permit a higher hydraulic loading than estimated.
Estimated Hydraulic Loading to Achieve Water Quality Arriving at the Ground
Water Interface 22 m (72 ft) Below the Surface for the Lake George Site--
a) Determine, based on carbon to nitrogen ratio, the maximum
nitrogen removal possible from Equation 3.
AN =
Based on this calculation, it is possible to reduce the total nitrogen from
25.8 mg/1 to 7.3 mg/1 with optimum management. Since the TOC is estimated
from the BOD, the actual optimum level is subject to the reliability of
estimates of TOC.
b) Determine loading rate, using Equation 2, to obtain required
level of nitrification. Since the TKN remaining after treat
ment is 0.3 mg/1 and the TKN applied was 19.6 mg/1, then
Max daily loading = 30 j*T"3) = 3° g1*"^ = 12 cm/day
The design maximum loading to achieve nitrification is 12 cm/day which
is slightly conservative based on the actual practice.
c) Determine loading rate, using Equation 4, to obtain required
level of denitrification. Since the nitrate remaining after
treatment is 3.6 mg/1,
43
-------
Max
loading = 4Q JT"5) = 19 s^'^z - 3 6 = l2'7 cm/day
Estimate Phosphorus Arriving at Ground Water Under Actual Application
Condi tions--
The water content and application rate are estimated at 0.5 and 2 cm/hr,
respectively. Then estimating the phosphorus concentration at a depth of
2200 cm (72 ft) using Equation 5,
kt = in §
0.002 hr-ff -,.("*)). in 4-
Cx = 1.3 mg/1
one sees the phosphorus is estimated to be 1.3 mg/1 compared to the 0.3 to
1.1 which is the range of values observed at well 11s. This well was sampled
22 m (72 ft) below the surface.
Table 10 presents design loadings and concentrations versus reported
observations. The observed nutrients are essentially the same as those
projected from the design rationale.
TABLE 10. DESIGN EVALUATION FOR LAKE GEORGE
Parameters
Hydraulic acceptance * cm/day
Max loading for N removal ** cm/day
BOD mg/1
SS mg/1
Phosphorus at 22 m mg/1
Design
7
12
£5
<_ 5
1.3
Reported
13
13
< 5
N.R.***
0.3-1.1
* Based on hydraulic conductivity of most restrictive layer
** Based on limitations for nitrification
*** N.R. is not reported
44
-------
Vine!and, New Jersey
The data required for a complete design are presented in Table 11.
Determination of Maximum Acceptable Hydraulic Load
a) Determine SAR
SAR = -N§ = 2'16 = 5.4
|Ca + Mg JO.3 + 0.1?
SAR is, therefore, not limiting since 9 is the value where the designer
should become concerned.
b) Determine limitation based on worst case hydraulic conductivity.
15.8 cm/hr x 24 hr/day x 0.1 = 38 cm/day
c) BOD limitation for hydraulic loading
2
2.25 mg/cm -day _ 1(- ,.
154 mg/1000 cc " ib cm/aay
d) Suspended solids limitation for hydraulic loading
The limiting criteria for hydraulic loading is BOD with a maximum design
loading of 15 cm/day.
Estimate Hydraulic Loading to Achieve Water Quality of the Vineland Site
a) Determine, based on carbon to nitrogen ratio, the maximum
nitrogen removal possible from Equation 3.
AM TOC-5 _ 108-5 _ /!
AN = * 2 ^'
Based on this calculation, it should be possible to reduce the total nitrogen
to 1 mg/1 with proper design and management.
b) Determine loading rate, using Equation 2, to obtain required
level of nitrification. Since the TKN remaining after treat-
ment is 22.2 mg/1 and the TKN applied is 40 mg/1
Max daily loading = 30 ^T"3) = ^ [^l = 17 cm/day
45
-------
TABLE 11. INPUT DATA REQUIRED TO EVALUATE VINELAND SITE (4)
Ca applied 12 mg/1 =
Mg applied 4 mg/1 =
Na applied 60 mg/1
Hydraulic Conductivity of restrictive layer
BOD applied
SS applied
Temperature mean annual air
TKN applied
TKN at ground water (design TKN)
NO, applied
NOZ at ground water (design NCQ
TOC applied
Total soluble Phosphorus applied*
Total Phosphorus at 150 cm*
0.3 meq/1
0.16 meq/1
2.61 meq/1
15.8 cm/hr
154 mg/1
43 mg/1
13°C
40 mg/1
22.2 mg/1
-
1.5 mg/1
108 mg/1
4.5-7.2 mg/1
4.3 mg/1
* (4, pg. 84)
46
-------
The design maximum loading to achieve nitrification is 17 cm/day. This
is significantly greater than the actual practice. There are two apparent
problems at this site. First, the water table is mounding to within 90 cm
(3 ft) of the soil surface. Thus, some sort of artificial drainage is re-
quired to achieve a minimum of 150 cm (5 ft) unsaturated profile at this
location. The second problem appears to be with the continuous flooding.
The operating schedule does not allow sufficient time for the system to
oxidize and nitrify the wastewater. These are apparent management reasons
for not achieving the design water quality.
c) Determine loading rate to obtain required level of denitrifi-
cation. Since the nitrate remaining after treatment is 1.5
mg/1
Max loading is -"lS = 20
The design maximum loading to achieve denitrification is 20 cm/day.
Again, there are management problems which apparently are limiting system
operation.
Estimate Phosphorus Arriving at Ground Water Under Actual Application
Condi tions--
The water content and application are estimated at 0.5 and 0.5 cm/hr,
respectively. The total soluble phosphorus applied at the basins ranged
from 4.5 to 7.2 mg/1 during the study period (4, pg. 84). But the mean
total phosphorus was reported as 9.3 mg/1. At one well cased and perforated
in the depth range of 120 to 180 cm (4 to 6 ft), the phosphorus ranged from
2.9 to 5.7 mg/1 with a mean of 4.3 mg/1. Estimating the phosphorus concen-
tration at a depth of 150 cm (5 ft) using Equation 5 gives
0.002 hr- in
C = 4.3 mg/1
A
which is the same as observed at the site.
Table 12 presents design loadings and concentrations versus reported
values. If the design rationale is adequate, changing the operation of the
system should improve the water quality reaching the ground water with
respect to nitrogen. Phosphorus is being treated as expected.
CAPABILITY FOR IMPROVEMENT
The Vineland site should be capable of meeting primary drinking water
standards by increasing the size of the plant, adding required drainage,
and changing the scheduling of flooding cycles. Since sufficient carbon is
available to denitrify the applied wastewater, it should be possible to design
the system such that not more than 10 mg/1 total nitrogen will reach the
ground water. On the other hand, the Milton system could not be significantly
47
-------
improved with present level of pretreatment, due to the lack of available
carbon. Hollister and Lake George are operating their systems efficiently
and no changes would be recommended based on this study.
TABLE 12. DESIGN EVALUATION FOR VINELAND
Parameters
Hydraulic acceptance cm/ day
Max loading for N removal* cm/day
BOD mg/1
SS mg/1
Phosphorus at 150 cm (5 ft.) mg/1
Design
15
17
±5
1 5
4.3
Reported
3-6
3-6
10
N. R.**
4.3
* Based on limitation due to nitrification.
** N. R. is not reported
48
-------
REFERENCES
1. U. S. Environmental Protection Agency. Process Design Manual for Land
Treatment of Municipal Wastewater. EPA-625/1-77-008. Washington, DC.
October 1977.
2. EPA National Interim Primary Drinking Water Regulations. Federal
Register. Dec. 24, 1975.
3. Pound, C. E.} R. W. Crites, J. V. Olson. Long-Term Effects of Land
Application of Domestic Wastewater, Hollister, California, Rapid Infil-
tration. EPA-600/2-78-084. U. S. Environmental Protection Agency, Ada,
Oklahoma. April 1978. 150 pp.
4. Aulenbach, D. B. Long-Term Recharge of Trickling Filter Effluent into
Sand. EPA-600/2-79-068. U. S. Environmental Protection Agency, Ada,
Oklahoma. March 1979. 146 pp.
5. Koerner, E. L., and D. A. Haws. Long-Term Effects of Land Application
of Domestic Wastewater, Vineland, New Jersey, Rapid Infiltration Site.
EPA-600/2-79-072. U. S. Environmental Protection Agency, Ada, Oklahoma.
March 1979. 167 pp.
6. Benham-Blair and Affiliates, Inc. and Engineering Enterprises, Inc.
Long-Term Effects of Land Application of Domestic Wastewater, Milton,
Wisconsin, Rapid Infiltration Site. EPA-600/2-79-145. U. S. Environ-
mental Protection Agency, Ada, Oklahoma. August 1979. 128 pp.
7. U.S. Public Health Service. Manual of Septic-Tank Practice. Publ. No.
526. 1969.
8. Leach, Lowell. Management of Rapid Infiltration Systems. U. S.
Environmental Protection Agency, Robert S. Kerr Environmental Research
Laboratory, Ada, Oklahoma. Unpublished data.
9. Enfield, Carl G. Servo-Controlled Optimization of Nitrification-
Denitrification of Wastewater in Soil. J Environ. Qual 6:456-458.
10. Enfield, Carl G. Evaulation of Phosphorus Models for Prediction of
Percolate Water Quality in Land Treatment. ln_ State of Knowledge in
Land Treatment of Wastewater. Vol. 1. Intnl. Sym., Hanover, New
Hampshire. August 20-25, 1978. pp. 153-162.
49
-------
11. Enfield, Carl G. et al. Kinetic Model for Phosphate Transport and
Transformation in Calcareous Soil. I. Kinetics of Transformation.
To be published in Soil Sci. Soc. Amer. J.
12. Enfield, Carl G. et al. Kinetic Model for Phosphate Transport and
Transformation in Calcareous Soils. II. Laboratory and Field
Transport. To be published in Soil Sci. Soc. Amer. J.
50
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-RO-165
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Summary of Long-Term Rapid Infiltration
System Studies
5. REPORT DATE
JULY 1930 ISSUING DATE.
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Lowell E. Leach, Carl G. "Enfield, and
Curtis C. Harlin, Jr.
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Robert S. Kerr Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Ada, Oklahoma 74820
10. PROGRAM ELEMENT NO.
A35B1C
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
In-House
Same as above.
14. SPONSORING AGENCY CODE
EPA/600/15
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This project was initiated with the objective of summarizing and
comparing the data published in individual reports of four long-term
rapid infiltration municipal wastewater systems. Evaluation of this
data provides the user community with a condensed assessment of the
treatment received at each of these sites from which projected treat-
ment of planned systems can be estimated.
In addition to a summarization and evaluation of data, a hypothetical
design was made for each system based on a rationale for both hydraulic
loading and effluent water quality considerations. This rationale is
proposed as a supplement to the design criteria presented in the Process
Design Manual for Land Treatment of Municipal Wastewater, EPA-625/1-77-008,
1977.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Wastewater Treatment -(Municipal)
Ground Water Quality
Design Criteria - Rapid Infiltration
Systems
Rapid Infiltration
Land Treatment
Wastewater Treatment
Systems
68D
91A
94B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21 ..NO. OF PAGES
59
20 SECURITY CLASS (TMspagc)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
51
it U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0049
------- |