&EPA
United States
Environmental Protection
Agency
Robert S. Kerr Environmental Research EPA-600/2-79-174
Laboratory August 1979
Ada OK 74820
Research and Development
Treatment of
Secondary Effluent by
Infiltration-Percolation
-------
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-600/2-79-174
August 1979
TREATMENT OF SECONDARY
EFFLUENT BY INFILTRATION-PERCOLATION
by
D.G. Smith
K.D. Linstedt
E.R. Bennett
City of Boulder, Colorado
and
University of Colorado
Boulder, Colorado 80309
Grant No. R803931
Project Officer
Lowell Leach
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. Approval does not signify that the contents necessarily
reflect the views and policies of the U. S. Environmental Protection Agency,
nor does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
11
-------
FOREWORD
The Environmental Protection Agency was established to coordinate
administration of the major Federal programs designed to protect the
quality of our environment.
An important part of the agency's effort involves the search for
information about environmental problems, management techniques and new
technologies through which optimum use of the nation's land and water
resources can be assured and the threat pollution poses to the welfare of
the American people can be minimized.
EPA's Office of Research and Development conducts this search 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 groundwater;
(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.
This report contributes to the knowledge essential if the EPA is to
meet the requirements of environmental laws that it establish and enforce
pollution control standards which are reasonable, cost effective and provide
adequate protection for the American public.
William C. Galegar \J
Director
Robert S. Kerr Environmental Research Laboratory
iii
-------
ABSTRACT
The objective of this study was to evaluate the performance of an
Infiltration-percolation system for Improving the quality of secondary
effluent from a municipal wastewater treatment system. This was done by
constructing three Infiltration-percolation basins and monitoring their
Influent and effluent quality over a two year period.
The facility consisted of three Infiltration-percolation basins of
sizes ranging between 0.24 hectares (ha) [0.6 acres(ac)] and 0.36 ha (0.9 ac).
Unchlorlnated secondary wastewater effluent was applied twice a week to each
basin at loading rates which varied between 12.2 meters/year (m/yr) [40 feet/
year (ft/yr)] and 48.8 m/yr (160 ft/yr). The wastewater percolated through
loamy and clay sands covering alluvial sand and gravel. The percolate was
collected 2.4-3.0 m (8-10 ft) below the surface by underdralns for discharge.
Analyses of the basin Influent and effluent collected from the under-
dralns Indicated that the systems were generally effective 1n reducing the
wastewater concentrations of COD, conform organisms, and ammonium nitrogen.
Phosphorus leakage occurred to some extent 1n each of the basins, with the
most heavily loaded basins yielding the highest phosphorus concentrations 1n
the discharge water. The nitrate concentration of the water Increased
significantly because of nitrification of the ammonium nitrogen retained
within the soil matrix. The concentrations of hardness, alkalinity, and
chlorides also showed significant Increases 1n the percolate water.
This report was submitted 1n fulfillment of Grant No. R803931 by the
City of Boulder, Colorado, and the University of Colorado, under the sponsor-
ship of the U. S. Environmental Protection Agency. The report covers the
period August 1, 1975, to June 30, 1978, and the work was completed as of
October 15, 1978.
1v
-------
CONTENTS
Foreword ill
Abstract iv
Figures vi
Tables ix
Acknowledgements x
1. Introduction 1
2. Conclusions 2
3. Recommendations 3
4. Experimental System and Procedures 4
Site Description and Physical Facility 4
Basin Modification 7
Basin Loading 7
5. Sampling and Analysis 13
Sampling 13
Analytical Methods 13
6. Hydraulic Characteristics of the Site 21
Ground Water Profile 21
Basin Hydraulics 21
Infiltrometer Studies 32
Factors Affecting Infiltration Rates 41
7. Treatment Performance of the Infiltration-Percolation
System 51
General Considerations 51
Phosphorus Behavior 51
Refractory Organics 57
Nitrogen 57
Dissolved Salt Species 65
Coliform Organisms 70
Heavy Metals 76
8. Column Studies 83
Column Operation 83
Column Performance 84
Chemical Oxygen Demand 84
Phosphorus 86
Nitrogen 86
Dissolved Solids 89
Coliforms 98
Heavy Metals 99
References 101
-------
FIGURES
Number
1 Schematic of Boulder wastewater treatment plant 5
2 Infiltration-percolation system layout 6
3 Basin underdrains 8
4 Typical section through basins 9
5 Well point locations 10
6 Modified basin configuration 11
7 Analytical procedure for heavy metals 19
8 Ground water contours and flow directions (May, 1975) 22
9 Ground water elevations surrounding basins (August, 1976) ... 23
10 Ground water mounding pattern within basins 24
11 Typical discharge profile for Basin 1 27
12 Typical discharge profile for Basin 2 prior to
modification 28
13 Typical discharge profile for Basin 3 prior to
modification 29
14 Normalized discharge profiles prior to modification 30
15 Peak discharge flows prior to basin modification 31
16 Infiltrometer test locations 33
17 Infiltrometer test results for Basin 1 36
18 Infiltrometer test data on Basin 2 prior to modification ... 37
19 Infiltrometer test data on Basin 3 prior to modification ... 39
20 Pattern of infiltration rate decline in Basin 1 43
vi
-------
FIGURES (continued)
Number Page
21 Pattern of infiltration rate decline in Basin 2 44
22 Pattern of infiltration rate decline in Basin 3 45
23 Infiltration rate as a function of suspended solids
loading in Basin 1 46
24 Infiltration rate as a function of suspended solids
loading in Basin 2 47
25 Infiltration rate as a function of suspended solids
loading in Basin 3 48
26 Infiltration rate as a function of wastewater temperature ... 50
27 Weekly precipitation and minimum temperature of renovated
water 53
28 Effluent phosphorus as a function of time 54
29 Effluent phosphorus variation within loading cycles 56
30 Effluent COD as a function of time 58
31 Effluent ammonium nitrogen as a function of time 60
32 Effluent nitrate nitrogen as a function of time 61
33 Cumulative nitrogen applied and discharged from Basin 1 .... 62
34 Suspended solids variation in applied wastewater 64
35 Nitrate variation in effluent within loading cycles 66
36 Effluent ammonium variations within loading cycles 67
37 Effluent hardness as a function of time 68
38 Change in hardness concentration through the basins 69
39 Change in calcium concentration through the basins 71
40 Change in magnesium concentration through the basins 72
41 Change in alkalinity through the basins 73
42 Change in chloride concentration through the basins 74
vii
-------
FIGURES (continued)
Number Page
43 Wastewater cadmium concentrations 77
44 Wastewater copper concentrations 78
45 Wastewater chromium concentrations 79
46 Wastewater nickel concentrations 80
47 Wastewater lead concentrations 81
48 Wastewater zinc concentrations 82
49 COD concentration variation with column depth and time .... 85
50 Wastewater phosphorus concentration as a function of
soil depth 87
51 Nitrate concentration as a function of column depth
and sampling time 90
52 Column underdrain flow as a function of time after
loading 91
53 Mass flow of nitrogen from column as a function of
time after loading 92
54 Wastewater calcium concentration as a function of
column depth 93
55 Wastewater magnesium concentration as a function of
column depth 94
56 Wastewater alkalinity as a function of column depth 95
57 Wastewater hardness as a function of column depth 96
58 Wastewater chloride concentration as a function of
column depth 97
59 Wastewater heavy metal concentration variations
with depth 100
viii
-------
TABLES
Number Page
1 Loading and Sampling Schedule (October, 1976 to January, 1977) ... 14
2 Loading and Sampling Schedule (February, 1977 to June, 1978) .... 16
3 Ground Water Elevations 25
4 Base Flows 26
5 Summary of Infiltrometer Test Results 35
6 Summary of Infiltration Rates 40
7 Seasonal Averages for Wastewater Constituents in
Boulder Secondary Effluent 52
8 Coliform Removal Data 75
9 Selected Nitrogen Removal Data 88
10 Coliform Removal with Depth 99
-------
ACKNOWLEDGEMENTS
The authors wish to thank the many people who contributed to the
completion of this research effort. The support of the City of Boulder waste-
water treatment plant staff is gratefully acknowledged. Mr. Arthur Dike,
Mr. Charles Tregay, and Mr. Alton Ragsdale were particularly helpful in the
operation and maintenance of the test facility. Similarly, Mr. Chris Rudkin
and Mr. Steve Miller provided much information and assistance in the develop-
ment of sampling procedures at the site.
Sampling, analysis, and operational coordination were provided by the
following individuals during their graduate training at the University of
Colorado: Mr. Bob White, Mr. Paul Hamilton, Mr. Dan St. John, Mr. Bill
DeOreo, Mr. Bill Earley, Mr. Joe Tamburini, and Mr. Paul King.
The authors wish to express appreciation to the Robert S. Kerr Environ-
mental Research Laboratory for its support of this project, and especially to
Mr. Lowell Leach, and Mr. Richard Thomas who served as Project Officers
during portions of the study. The assistance of Mr. Roger Dean, Region VIII,
EPA, in providing coordination is also acknowledged.
-------
SECTION 1
INTRODUCTION
In 1972 the United States Congress passed Public Law 92-500, a bill
which committed this country to the task of upgrading polluted waters and
preventing degradation of clean waters. Three levels of treatment were
identified as the primary goals of this act. By 1977, secondary treatment
or its equivalent would be required; by 1983, the best available treatment
would be required; and by 1985, the goal was to eliminate the discharge of
pollutants to our waterways. In order to comply with the strict standards
set forth in this law, communities and industries alike were faced with pro-
viding some form of highly efficient wastewater treatment. This treatment
would be directed at removing undesirable wastewater constituents such as
nitrogen, phosphorus, suspended solids, refractory organics, and microorgan-
isms. Their removal could be accomplished through application of either
in-plant physical, chemical, and biological processes, or by land treatment.
In addition to establishing strict discharge requirements, P.L. 92-500
also specified that land treatment be considered as an alternative to other
advanced treatment processes. With the implementation of this requirement,
it has become apparent that land treatment combines the various physical-
chemical, and biological processes into a single process which is very
effective in many wastewater treatment situations. Land treatment is espe-
cially attractive for smaller communities where the land is readily available
and the cost effectiveness of physical-chemical processes is somewhat
questionable.
Three major methods of land application have been demonstrated to pro-
vide effective wastewater treatment. They include: irrigation, overland
flow, and infiltration-percolation. Successful application of one of these
land treatment methods depends on several interrelated factors such as the
soil type, geology, topography, ground water characteristics, and climate.
As a result of these considerations, land treatment processes are very site
specific. Selection of the best land use system for a particular location
may depend on any one, or all of these considerations.
The study discussed in this report involved the operation of three small
infiltration-percolation (i-p) basins in Boulder, Colorado. These infiltra-
tion-percolation basins were installed as a demonstration project in the
spring of 1976 to treat a portion of the secondary effluent discharged by the
City of Boulder wastewater treatment facility. The treatment efficiency of
the infiltration-percolation system was monitored throughout the period of
operation by regular sampling of the applied secondary effluent and the
renovated water. The performance of this system was the subject of this
report.
-------
SECTION 2
CONCLUSIONS
The infiltration-percolation system demonstrated excellent capability
for polishing secondary effluent in the context of the Boulder, Colorado
wastewater treatment situation. Under proper conditions of hydraulic loading
and loading cycle control, the system was capable of providing very high
levels of removal for virtually all wastewater constituents of major pollu-
tion significance.
Infiltration-percolation systems were shown to be capable of operation
throughout the year in either flooded basins or ridge and furrow configura-
tions where appropriate surface maintenance was provided. The necessary
surface maintenance consisted of scarification to control plant development
in the bare basins so that ice which formed in the cold winter months would
not be anchored, but could float free with successive wastewater applications.
The infiltration rate in the ridge and furrow system appeared to be less
affected by wastewater suspended solids deposition than the flat basins which
were completely flooded. This suggests that it would be possible to operate
ridge and furrow systems longer between extended drying periods than flooded
basins loaded at comparable rates.
When a basin was loaded at a sufficiently high hydraulic rate to stress
the infiltration capabilities of the soil system, one of the indications of
the stressed condition was a deterioration of the quality of the effluent
from the basin. This quality deterioration was evidenced by increased con-
centrations of both phosphorus and ammonium nitrogen, as well as a reduction
in the amount of nitrate nitrogen in the basin discharge.
The basins demonstrated good capability for making major reductions in
the concentration of fecal coliforms in the unchlorinated secondary effluent.
Nevertheless, significant concentrations of these bacteria were still
detected after passage of the wastewater through 2.4-3.0 meters (8-10 feet)
of the composite soil mixture found in the Boulder treatment location.
Significant removal of heavy metals was observed in the infiltration-
percolation soil system. Most of this removal occurred in the tight clay
loam layer at the top surface of the soil profile.
-------
SECTION 3
RECOMMENDATIONS
From the observations developed in conjunction with this investigation,
the following recommendations seem appropriate:
The capability of the infiltration-percolation basins for
treating a wide range of wastewater pollutants suggest that these
systems would be appropriate for treating primary as well as
secondary effluent. It is recommended that a comparative study
be performed which would assess the acceptable hydraulic loading
rates and product water characteristics with these two different
levels of pre-application treatment.
From the aesthetic and operational perspectives, it would
appear that there might be advantages to the use of a ridge and
furrowed application system rather than a flat basin having no
vegetation. A study should be undertaken to identify the loading
schedule which would provide the best combination of hydraulic
loading and treatment performance in this configuration. The aim
of this effort should be to minimize the total land area require-
ments for providing this alternative application method.
In the study discussed in this report, little total nitrogen
removal was observed. It was shown in the operation of beds 1
and 2 that if the loading rate was maintained at an appropriate
level, the effluent ammonium nitrogen could be controlled at a
concentration of less than 1 mi 111gram/liter (mg/1). However, with the
loading sequence practiced 1n this Investigation, essentially all of the
nitrogen was discharged from these beds as nitrate. Other Investi-
gators have demonstrated that substantial denitriflcatlon can be
achieved In rapid infiltration systems if the loading sequence Is
appropriately managed. Modification of the loading sequence should
be attempted at the Boulder facility for the purpose of maximizing
the total nitrogen removal.
Significant phosphorus leakage was observed in the product water
from all three of the basins. It is suspected that this leakage
occurred in part because of the short hydraulic detention from the
basin surface to the underdrains. It would be of interest to
determine the impact of varying the underdrain placement to increase
the contact time between the wastewater and the soil profile. This
could be done by selectively closing off some of the existing drains
in the Boulder infiltration-percolation system.
-------
SECTION 4
EXPERIMENTAL SYSTEM AND PROCEDURES
SITE DESCRIPTION AND PHYSICAL FACILITY
The wastewater treated in the demonstration project facility was drawn
from the effluent of the City of Boulder, 75th Street Wastewater Treatment
Plant. This treatment facility processed approximately 0.53 cubic meter/
second (nr/sec) [12 million gallons per day (MGD)] of wastewater by means of
a standard rate trickling filter, as 1s shown schematically in Figure 1.
The site of the infiltration-percolation system consisted of approxi-
mately 1.0 hectares (ha) (2.5 ac) of land located adjacent to the City of Boulder
75th Street Wastewater Treatment Plant, and about 152 meters (500 feet)
south of Boulder Creek. The land, which was formerly rangeland, had a slight
eastward slope ranging between 0-1%. The soil was classified as the Niwot
series soil which has been described by the U. S. Soil Conservation Service
as a layer varying from sandy clay loam to light clay loam superimposed
over sand and gravel (1). These soils have been described to have moderate
permeability and a high seasonal water table ranging from 0.15 meters to
0.46 meters (6 to 8 1nches)(l). Underlying this soil at a depth of approxi-
mately 3.7 meters (12 feet) was the very impermeable Pierre Shale formation
(2). The depth to ground water on the site was found to vary from 0.9 to
1.5 meters (3 to 5 feet) In May, 1975(2).
The City of Boulder Wastewater Utility Department designed and
constructed the pilot plant facility. Construction began 1n December 1975
and was completed in April 1976. Three Infiltration-percolation basins were
constructed, with each basin separated by a berm approximately 0.76 meters
(2.5 feet) high. All three basins were surrounded by an impermeable clay-
core dike which was also 0.76 meters (2.5 feet) high. The 1.83 meter (6 feet)
wide clay-core dike extended from the ground surface to the impermeable
bedrock, and served to completely enclose the system, and minimize the
Interaction between the beds and the surrounding ground water. The South
basin, referred to as Basin 1, was 0.35 ha(0.87 ac); the Middle basin, Basin
2, was 0.24 ha (0.60 ac); and the North basin, Basin 3 was 0.26 ha (0.65 ac).
Figure 2 shows a plan view of the three basins and their respective sizes.
After the basin area was completely sealed, an underdrain system was
installed to lower the existing water table and to collect the applied water
during the operation of the system. The underdrain system for each basin
consisted of two 0.18 meter (7 Inch) perforated PVC drain pipes located
-------
Headworks
Flow
Diversion
Box
Primary Trickling Secondary
Clarifiers Filters Clarifiers Chlorination
City
Collection
System
\
\
\
\
\
Grit to
Land Disposal
Site
-------
Clay Dike
Distribution
Box
In-Line Flow
Meter
Secondary
Clarifier
BASIN 1 - X'ov
0.87 ac
No Scale
ump House
Clay Dike
Figure 2. Infiltration-Percolation System Layout
-------
2.43 to 3.05 meters (8 to 10 feet) under the basin surface. The collected
water flowed by gravity to a manhole in each basin, and then to a central
manhole where the monitoring and sampling took place. From this manhole the
water flowed to a wet well and was pumped to its discharge point in Boulder
Creek. A plan view showing the underdrain system is shown in Figure 3, and a
cross section of the basins is shown in Figure 4.
Well points were installed in each basin to measure the rise and fall of
the water table during the loading cycles. These well points were developed
by augering holes through the beds, and installing 0.04 meter (1.5 Inches) PVC
pipes in the holes. These pipes extended from 0.91 meters (3 feet) above the
ground surface to the bedrock. The locations and elevations of these well
points have been indicated in Figure 5. The depth of the water table was
measured by lowering a float down the pipe until the water surface was inter-
cepted, and then recording the depth.
BASIN MODIFICATION
Following six months of treatment operation on the beds which have been
described, a decision was made to modify Beds 2 and 3 in an attempt to
improve their hydraulic performance. As such, in January, 1977 the top tight
loamy layer of soil was removed from both of these beds. This necessitated
the removal of the top 0.46 meters (18 inches) of soil from Bed 3, and the
top 0.61 meters (24 inches) of soil from Bed 2. Both beds were subsequently
graded to a flat surface with a slight,eastward slope, similar to that in
their initial condition.
While Basin 3 remained in the modified form throughout the rest of the
study, Bed 2 was further altered in March, 1977 by the construction of a
ridge and furrow system. The furrows were 0.46 meters (18 inches) deep and
spaced approximately 1.98 meters (6.5 feet)on center. The furrows averaged
1.02 meters (40 inches) in width, and the ridges averaged 0.91 meters (36
inches) wide. There were 9 furrows constructed in Bed 2, each approximately
85 meters (280 feet) long. A furrow at each end was constructed to facili-
tate loading. A plan view of the three basins, and the modified areas, is
shown in Figure 6. It should be noted that Bed 1 was not altered in any way,
after the initial construction.
BASIN LOADING
Loading of the basins was accomplished by pumping the effluent from a
secondary clarifier with a centrifugal pump which was powered by a portable
gasoline engine. The unchlorinated wastewater from this point was pumped
through approximately 297 meters (975 feet) of 0.36 meter (14 inch) PVC pipe
to a distribution box which directed the water to the desired basin.
Each basin was loaded twice a week at lh day intervals. From October,
1976 to January, 1977, Bed 1 (surface of basin) was loaded at approximately
8:00am on Mondays and 7:00pm on Thursdays; Bed 2, 8:00am on Tuesdays and 7:00pm
-------
00
5115.36
Discharge to
Boulder Creek
0.40% slope ~~ "~
I No Scale
5101.49
0.40% slope
5104.81
0.40% slope """
178 mm(7 in)perforated PVC underdrain
J_ 0.40% slope
Gnd
Figure 3. Basin underdrains.
(Elevations in feet.)
-------
Standpipe Well
Clay Dike
Loam to
Clay Loam
Sand and
Gravel
.Underdrafn -~)
'. • N • .unoerarain —n
Pierre
Shale
Manhole
.X V
Figure 4. Typical section through basins.
-------
5117.66
£ N.E. Corner
Distribution
Box
5115.88
5114.78 /*
5116.43
No Scale
1114.36 |
N. Rim M.H.
^5114.02
B.M. Well C-19X
(CH2M-Hill, »
1975)
Pump House
5114.37
B.M. Well C-20~
Figure 5. Well point locations (.) and elevations (feet). CH2"-H1". 1975)
-------
Clay Dike
Distribution Box
In-Line
Flow Meter
Secondary
Clarifier
No Scale
A Pump House
Figure 6. Modified basin configuration.
-------
on Fridays; Bed 3, 8:00am on Wednesdays and 7:00pm on Saturdays. The beds
were loaded on alternating days to facilitate the ease of operation, sampling,
and analysis. In February 1977, the loading cycle was shifted one day later
to begin on Tuesday morning instead of Monday morning due to a change in
operator schedules. After six weeks of loading, all basins were dried and
then scarified. The required length of drying time varied from one week in
the summer to two or three weeks In the winter, depending on the weather
conditions during the drying period.
12
-------
SECTION 5
SAMPLING AND ANALYSIS
SAMPLING
Samples of both the influent and effluent of the three basins were taken
at regular intervals. The influent samples were taken at the inlet pipe of
each bed during loading. Following renovation treatment, the basin effluent
was sampled at the center manhole, with samples collected independently for
each basin at specified times after loading. Tables 1 and 2 summarize the
loading and sampling schedule. Only the first loading of each week was
sampled and monitored. The temperature, dissolved oxygen level, nitrogen,
hydraulic flow, and infiltration rate measurements were monitored with time
after the loadings. All other parameters were measured only once for each
loading cycle on the sample collected 24 hours after loading.
The water samples were preserved immediately after collection. Samples
collected for testing of nitrites and nitrates were preserved with 40 mgA
HgCl2, and stored at 4°C. Samples to be analyzed for organic nitrogen and
ammonia were acidified with concentrated sulfuric acid to a pH of less than
2, and refrigerated at 4°C. These procedures were in accordance with those
outlined in Methods for Chemical Analysis of Water and Wastes (3). Samples
collected for all of the tests were refrigerated at 4°C, with the exception
of those used in determining temperature and dissolved oxygen.
ANALYTICAL METHODS
All analyses, except those noted, were performed at the Sanitary
Engineering Laboratory at the University of Colorado. The samples were sub-
jected to the following series of analyses for purposes of this investiga-
tion: total solids, suspended solids, phosphorus, COD, temperature, coli-
forms, and the nitrogen series. In addition, selected samples were analyzed
for Cd, Cu, Cr, Ni, Pb, and Zn during November-December, 1977.
Temperature measurements were taken immediately following the sampling
at the pilot plant site. A mercury-filled centigrade thermometer, calibrated
to 1°C, was used for these measurements.
Total kjeldahl nitrogen (TKN) was determined by acid digestion of 250
mill niters (ml) samples and distillation Into boric add. The TKN was titrated
with standard sulfuric add to the pH of the blank carried through the same
procedures. The difference between the TKN and ammonium nitrogen was the organic
13
-------
TABLE 1. LOADING AND SAMPLING SCHEDULE
(OCTOBER 1976 TO JANUARY 1977)
Day
Time
Basin 1
Basin 2
Basin 3
Monday 8:00 a.m. Load Basin
Measure Flow
Sample Effluent
9:00 a.m. Measure Flow
Sample Influent
Sample Effluent
11:00 a.m. Measure Flow
Sample Effluent
2:00 p.m. Measure Flow
Sample Effluent
8:00 p.m. Measure Flow
Sample Effluent
Tuesday 8:00 a.m.
9:00 a.m.
Measure Flow
Sample for
Complete Analysis
11:00 a.m.
2:00 p.m.
8:00 p.m. Measure Flow
Sample Effluent
Wednesday 8:00 a.m. Measure Flow
Sample Effluent
9:00 a.m.
11:00 a.m.
2:00 p.m.
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample Influent
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample for
Complete Analysis
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample Influent
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
(continued)
14
-------
TABLE 1. (continued)
Day
Time
Basin 1
Basin 2
Basin 3
8:00 p.m. Measure Flow
Sample Effluent
Thursday 8:00 a.m.
7:00 p.m. Load Basin
8:00 p.m.
Friday 8:00 a.m.
7:00 p.m.
8:00 p.m.
Saturday 7:00 p.m.
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample for
Complete Analysis
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Load Basin
15
-------
TABLE 2. LOADING AND SAMPLING SCHEDULE
(FEBRUARY 1977 TO JUNE 1978)
Day
Time
Basin 1
Basin 2
Basin 3
Tuesday 8:00 a.m.
9:00 a.m.
12 Noon
4:00 p.m.
8:00 p.m.
Wednesday 8:00 a.m.
9:00 a.m.
12 Noon
4:00 p.m.
8:00 p.m.
Thursday 8:00 a.m.
9:00 a.m.
12 Noon
4:00 p.m.
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample Influent
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample for
Complete Analysis
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample Influent
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample for
Complete Analysis
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample Influent
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
(continued)
16
-------
TABLE 2. (continued)
Day
Time
Basin 1
Basin 2
Basin 3
8:00 p.m. Measure Flow
Sample Effluent
Friday 8:00 a.m.
8:00 p.m. Load Basin
Saturday 8:00 a.m.
8:00 p.m.
Sunday 8:00 p.m.
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Load Basin
Measure Flow
Sample Effluent
Measure Flow
Sample for
Complete Analysis
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Measure Flow
Sample Effluent
Load Basin
17
-------
nitrogen. The organic nitrogen procedure has been described in detail in
Section 135 of Standard Methods (4).
Ammonium determinations were made with an Orion Research ammonium
electrode Model 95-10. The instructions provided with the electrode were
followed, with a new calibration curve prepared for each group of samples.
The nitrite determinations were initially made by following the method
outlined in Section 420 of the 14th Edition of Standard Methods (4). How-
ever, after January, 1977 nitrite analyses were modified to facilitate the
use of a Technicon Auto Analyzer II. The automated procedure was an adapta-
tion of the diazotization method outlined in Standard Methods (4).
Similarly, nitrates were initially determined by the Brucine method
outlined in Standard Methods (4). A Bausch and Lomb Spectronic 70 spectro-
photometer was used for transmittance readings with new standards run for
each set of nitrate samples. After January, 1977 an automated procedure was
adapted using the Technicon Auto Analyzer II. Nitrates were determined by
the automated copper-cadmium reduction method described in Section 605 of
Standard Methods (4).
Total phosphorus determinations were made by first acidifying the
samples, and then digesting them by the Persulfate Digestion Method outlined
in Section 223C-III of Standard Methods (4). Color development was accom-
plished by the Stannous Chloride method outlined in Section 223E of Standard
Methods (4). The transmittance was measured with a Bausch and Lomb Spec-
tronic 70 spectrophotometer at a wave length of 690 nanometers (nm).
Total and fecal coliforms were reported as coliforms per 100 ml, and
were determined by use of the membrane filter technique described in Section
408 of Standard Methods (4).
Both the suspended solids and total dissolved solids were run using some
variations to the process as described in Standard Methods (4). Suspended
solids were determined by the modified process described by Harada and his
co-workers (5). Total solids were determined by the following procedure:
first, evaporating dishes were dried at 103°C for one hour, then dessicated
for one-half hour, and weighed for tare weight. One hundred milliliters of
sample were then evaporated in the dishes on a hot plate. Following this,
they were dessicated for one-half hour and weighed.
The chemical oxygen demand (COD) analyses were determined by the method
outlined 1n Section 220 of Standard Methods (4). The soluble COD was
determined by following the Indicated procedure after filtration of the sample
through a 450 nm mllUpore filter.
During the loading cycle of November-December 1977, a special set of
heavy metal samples was taken from the secondary effluent entering the beds,
and the bed product water 24 hours after loading. These samples were stored
in new polyethylene bottles which had been acid washed three times in 1:1
HN03, 1:1 HC1, and distilled water. The samples were subsequently subjected
to the sample treatment Indicated in Figure 7.
18
-------
Secondary Eff.
[Unchlorinated)
I.P.Basin|-
Unacidified Sample
from
Basin Influent
50 ml
50 ml
i
Filtered
Within 2 Hours
0.45 y
Reagents
1 ml, 1:1 HN03
1 ml, 1:1 U3PQk
2 ml, 30% H202
Digestion
80°C for 45 min
Dissolved
Metals
Unacidified Sample
from
Basin Effluent
50 ml
50 ml
Dist. Water
Blank
Total
Metals
Zn by flame
AA
Cd, Cu, Cr, Ni, Pb
by flameless AA
Figure 7. Analytical procedure for heavy metals.
19
-------
As indicated in this figure, the concentrations of cadmium, copper,
chromium, nickel, and lead were determined by flameless atomic absorption.
Zinc was determined by flame atomic absorption since it was impossible to use
the flameless method for measurement of zinc in these samples because of the
very high sensitivity of the instrument and the excessive amounts of back-
ground zinc. The use of flameless atomic absorption allowed concentration of
the samples to be avoided, as well as the vigorous digestion necessary with
flame atomic absorption.
20
-------
SECTION 6
HYDRAULIC CHARACTERISTICS OF THE SITE
GROUND WATER PROFILE
Prior to basin construction, the ground water characteristics of the
proposed location were measured and recorded. From these measurements, the
ground water table contour of Figure 8 was developed to define the conditions
in May, 1975. From this figure it can be seen that the general pattern of
ground water flow was away from Boulder Creek and towards the north and west
sides of Basin 3. The flow was also directed towards the west end, and the
west and south sides of Basins 2 and 1, respectively. For comparison,
Figure 9 shows the ground water levels at selected sites surrounding the
basins on August 10, 1976, which was after several months of basin operation.
The water levels at that time were actually somewhat.1ower than those prior
to initiation of basin operation, reflecting the change in ground water table
with the time of the year.
Within the basins, the ground water elevations were lowered by the
underdrain svstem to aoproximately 2.4 meters (8 feet) from the surface
throughout the area of the infiltration-percolation system. These ground
water elevations were monitored on a time profile basis during a typical week
of loading, with the elevations reported in Table 3. Mounding occurred to a
small degree, but serious ground water mounding was prevented by the under-
drain system. The maximum mounding condition is indicated in Figure 10,
which shows a cross-section of the area from Boulder Creek to the
infiltration-percolation system.
With the steep gradient of ground water elevations across the
peripheral clay dike, it was expected that some flow would occur. As a
result, prior to application of any wastewater to the beds, the base ground
water flow discharged from each basin was measured. These flows are indi-
cated in Table 4, and show that the two outside basins had significant flow
across the dike, while the middle basin received essentially no ground water.
Under the minimum loading condition on Bed 3, the ground water constituted
about 2Q% of the average underdrain flow following basin loading. However,
with the higher loading following modification of the basins, the ground
water contribution to the underdrain flow was reduced to less than 5% of the
total underdrain flow in all the basins.
BASIN HYDRAULICS
Following flooding of each basin, the wastewater percolated downward
21
-------
PO
PO
Figure 8. Ground water contours and flow directions (feet) (May, 1975)
-------
ro
to
Well C-25
Water - 5108.13
Ground - 5112.58
Well C-19
- 5106.27
Ground - 5111.73
Well C-20
Water - 5106.43
Ground.- 5112.54
0)
£
1/1
Figure 9. Ground water elevations (feet) surrounding basins
(August, 1976)
-------
ro
ELEV. 5115 ft.
(15.59.m)
Clay
ELEV. 5110 ft.
(1557.4m)
Ground
Water
Table-.
ELEV. 5105 ft.
(1555.9m)•
n-
Access Road
to Treatment
PI ant-7 BASIN 1
BASIN 2
BASIN 3
Bed Surface
Boulder Creek
Bottom of Furrow
Ground Water.Table
•Ground Water
• • Table
Bedrock
Figure 10. Ground water mounding pattern within basins.
-------
TABLE 3. GROUND WATER ELEVATIONS (feet)
Bed and
Time After
Loading
Bed 1 - IP 0 hrs
19 8
1012
1024
1932
1948
1072
Bed 2 - 20 0 hrs
29 8
2924
2032
2948
Bed 3 - 39 0 hrs
30 8
3024
3972
Hell Point Location
1-1 1-2 1-3 2-1 2-2 2-3 2-4 3-1 3-2 3-3 3-4 3-5 3-7 3-8 A B C
Well Point Elevation
5116.27 5116.44 5116.16 5111.54 5115.02 5115.34 5115.14 5114.94 5115.30 5116.06 5116.43 5116.06 5115.78 5115.88 5114.78 5114.02 5114.37
5104.60 5105.69 5104.99 5103.54 5104.19 5103.52 5104.31 5103.19 5103.97 5104.39 5107.43 5104.14 5106.70 5108.21 5107.78 5106.77 5107,95
5104.52 510S.86 5104.99 5103.29 M04.10 5103.44 5104.14 5103.44 5104.02 5104.64 5107.43 5104.06 5105.89 5108.13 5107.78 5106.85 5107.70
5104.69 5106.44 5105.24 5103.37 5103.94 5103.44 5103.97 5103.44 ---------
5104.69 5106.44 5105.24 5103.37 5103.94 5103.44 5103.97 5103.44 5103.52 5104.56 5107.26 5104.14 5105.48 5107.88 5108.03 5107.10 b!08.20
5104.52 5106.44 5105.24 5105.96 5104.85 5104.35 5104.72 5103.44 5103.86 5104.39 5107.43 5104.06 5105.89 5108.38 5108.03 5107.02 5108.37
5104.77 5106.44 5105.24 5104.46 5104.27 5103.77 5104.22 5103.44 5103.97 5104.56 5107.51 5104.23 5105.73 5108.21 5107.86 5107.10 5108.04
b!04.81 5106.44 5105.33 5104.04 5104.35 5103.69 5104.31 5103.86 5103.94 510b.39 5107.76 5104.48 5106.31 5108.33 5108.03 5107.10 5108.20
5104.69 5106.44 5105.24 5103.37 5103.94 5103.44 5103.97 5103.44 5103.52 5104.56 5107.26 5104.14 5105.48 5107.88 5108.03 5107.10 5108.20
5104.52 5106.44 5105.24 5105.96 5104.35 5104.35 5104.72 5103.44 5103.86 5104.39 5107.43 5104.06 5105.89 5108.38 5108.03 5107.02 5108.37
5104.77 5106.44 5105.24 5104.46 5104.27 5103.77 5104.22 5103.44 5103.97 5104.56 5107.51 5104.23 5105.73 5108.21 5108.86 5107.10 510S.O-!
5104.77 5106.44 5105.33 5104.29 5104.02 5103.52 5104.22 5103.61 5104.19 5105.31 5107.43 5104.39 5106.39 5108.30 5107.70 5105.94 5108.37
5104.81 5106.44 5105.33 5104.04 5104.35 5103.69 5194.31 5103.86 5103.94 5105.39 5107.76 5104.48 5106.31 5108.33 5108.03 5107.10 5108.20
5104.77 5106.44 5105.24 5104.46 5104.27 5103.77 5104.22 5103.44 5103.97 5104.56 5107.51 5104.23 5105.73 5108.21 5107.86 5107.10 5108.04
5104.77 5106.44 5105.33 5104.29 5104.02 5103.52 5104.22 5103.61 5104.19 5105.31 5107.43 5104.39 5106.39 5108.30 5107.70 5106.94 5108.37
5104.81 5106.44 5105.33 5104.04 5104.35 5103.69 5104.31 5103.86 5103.94 5105.39 5107.76 5104.48 5106.31 6108.39 5108.03 5107.10 5108.20 |
5104.60 5105.61 5105.16 5105.04 5104.19 5103.52 5104.31 5103.87 5103.77 5104.73 5107.68 5104.27 5105.73 5108.21 5107.78 5107.02 5107.87 |
ro
ft x 0.305 > m
-------
TABLE 4. BASE FLOWS
Base Flow
Basin (cfs) (m3/sec)
1 0.013 3.68 x 10-1*
200
3 0.025 7.08 x lO'4*
through the soil, was collected by underdrains, and was pumped to the surface
and discharged into Boulder Creek. With each wastewater application, the
discharge flow of Basin 1 exhibited a rapid increase to a peak within 12
hours after loading. This was followed by a gradual decline in flow as is
shown by a representative discharge hydrograph in Figure 11. Prior to basin
modification, a representative hydrograph for Basin 2 exhibited a rise and
fall similar to that shown in Figure 12. While the peak occurred at about
the same time after loading, the curve was much broader, indicating a lower
percolation rate, or slower water mass flow through the soil. Similarly,
Basin 3 had comparable hydraulic characteristics, but with a dampened peak
occurring at about 15 hours as shown in Figure 13. To facilitate comparison
of the profiles for each bed, the flows from Basins 2 and 3 were normalized to
that of Basin 1 by multiplying their flows by the fraction of loading time
for Basin 1 divided by the loading time for each of the other basins. This
was done to correct the hydrographs for the different amounts of water
applied to each basin. With this normalization, the hydraulic dampening of
Basins 2 and 3 was more apparent, as can be seen in Figure 14. The lower
infiltration rates which are suggested by these curves for Basins 2 and 3
were also indicated by ponding of the water in these basins for several days
after loading. It can be speculated that this extended ponding time further
lowered the infiltration rates because of increased algae growth, and the
accompanying fouling of the surface with suspended solids.
In addition to the differences in infiltration rates which have been
noted for the three basins, the infiltration rates of all the basins gradu-
ally declined during the first few months of operation. This trend is
indicated in Figure 15, which is a plot of the peak flow of the underdrain
discharge during each of the first several loading cycles. The reduced peak
flow rates were characteristic of similar reductions which occurred in the
acceptable application rates during the first months of operation. The ini-
tial loading rates in May, 1976 were equivalent to 48.5 m/yr (159 ft/yr) on
Bed 1, 36.0 m/yr (118 ft/yr) on Bed 2, and 53.6 m/yr (176 ft/yr) on Bed 3.
After two weeks of loading at these rates, ponding conditions developed on
all three beds. As a result, all of the beds were allowed to dry for one
week and then scarified. Subsequently, the loading rates were reduced to
27.4 m/yr (90 ft/yr), 12.2 m/yr (40 ft/yr) and 15.2 m/yr (50 ft/yr) for
beds 1, 2, and 3, respectively. Bed 1 functioned acceptably at this load-
ing rate, and the rate was increased to 30.5 m/yr (100 ft/yr) in October,
26
-------
ro
- 25
- 20 _
ro
O
X
o
15 21
U (/)
CO
E
- 10 ^
- 5
36 48 60
Time after Loading (hours)
Figure 11. Typical discharge hydrograph for Basin 1.
-------
0.4
.3
q-
£> .2
o
o
12
24 36 48
Time after Loading (hours)
60
72
10.0
7.5
5.0
2.5
m
O
X
-------
ro
IQ
0.4
.3
«*-
o
10.0
7.5
5.0
O
0)
2.5
I
I
12
24
60
72
36 48
Time after Loading (hours)
Figure 13. Typical discharge hydrograph for Basin 3 prior to modification,
84
-------
in
to
(13
CO
-a
N
03
O
0 Basin 1
Basin 2
0 Basin 3
12
24 36 48
Time after Loading (hours)
60
72
- 20
- 15
- 10
- 5
84
CO
O
O
d>
c
•r-
VI
CO
•a
a)
N
o
Figure 14. Normalized discharge hydrographs prior to modification.
-------
to
M-
O
1.0
.9
.8
.7
.6
U- .5
(O
CD
D_
.4
.3
.2
.1
0
I I I
o Basin 1
a Basin 2
a Basin 3
=— Drying Period
I
J_
I
1 8 15 23 29
I August
Figure 15. Peak discharge flows prior to basin modification.
18 25
July
5 12 19
September
25
20
ro
O
15
tfl
CO
10 o
fO
cu
D_
26
-------
1976 without causing any hydraulic problems. However, Beds 2 and 3 continued
to demonstrate hydraulic difficulties, and by December, 1976 their loading
rates had been effectively reduced to 3.96 m/yr (13 ft/yr) and 5.79 m/yr
(19 ft/yr), respectively.
INFILTROMETER STUDIES
It was theorized that the surface layer of soil was restricting the
wastewater infiltration and, thus, was responsible for the poor hydraulic
performance of Basins 2 and 3. The Niwot Series soil of the area has been
characterized by the Soil Conservation Service as consisting of 0-0.30 meter
(0-12 inch) depth loams to clay loams, and 0.30-1.52 meter (12-60 inch)
depth coarse sand (1). If the surface soils of Beds 2 and 3 were indeed
clay loams, it was suspected that the theory would be correct. As a result
infiltrometer tests were performed on all three beds to determine- (1) the*
existing infiltration rates on the surface, (2) the type and depth of the
overlying soil layer, and (3) the infiltration rate of the underlying soil
and its soil classification. A double ring infiltrometer was used for per-
formance of these tests. The inner ring was 0.20 meters (8 inches) in
diameter and the outer ring was 0.38 meters (15 inches) in diameter, with
the length of both at 0.36 meters (14 inches). The rings were concentric,
and were connected by metal plates welded between the two rings.
Test sites were located within each of the basins which had no
significant surface disturbance, and which had soil textures representative
of the area. At these sites, the infiltrometer was pressed into the soil to
a depth of approximately 0.15 meters (6 inches). Installation was performed
with care to minimize the soil disturbance around the cylinder. The area
between the inner and outer ring provided a buffer pond which served to mini-
mize the radial flow of water away from the inner ring. This area was filled
first with water and kept at a constant level throughout the testing. The
water used in the testing was unchlorinated secondary effluent from the 75th
Street Trickling Filter Plant; the same wastewater which was applied to the
basins. After the outer ring was filled, the inner ring was filled to a
0.15 meter (6 inch) depth with care taken to minimize the disturbance of the
soil surface. The water level in the inner ring was measured at the start of
testing and at time intervals ranging from 1 minute to 30 minutes, depending
upon the infiltration rate. Sites with high infiltration rates were tested
several times, with the last test used to determine the average infiltration
rate. The average infiltration rate was determined by dividing the total
drop in the water level by the corresponding elapsed time. These infiltra-
tion rates were reported in centimeters per hour. The above procedure was
patterned after similar procedures described by Haise and Johnson (6,7).
Infiltration tests were performed at each of the locations indicated by
a dot in Figure 16. The lettered dots represent sites tested prior to surface
modification of the beds, and the numbered dots represent sites tested after
some soil was removed from the surface of Beds 2 and 3. Surface infiltra-
tion tests were performed in Bed 1 at the locations identified by points A,
D, and F in Figure 16. These tests yielded infiltration rates ranging from
0.58-7.13 cm/hr (0.23-2.81 in/hr), with an average of 4.42 cm/hr (1.74 in/hr).
32
-------
GO
No Scale
- I
Pump House
LEGEND
Infiltration Test
Location
Soil Log Location
Figure 16. Infiltrometer test locations.
-------
Details of the rates at each site have been included in Table 5. The soil
within this basin was a silty loam which extended from the surface to
0.38-0.66 meters (15-26 inches) from the surface. The underlying soil was
well rounded gravel mixed with sand and silt-sand loam. Infiltration tests
at a 0.38 meter (15 inch) depth at location C and a 0.66 meter (26 inch)
depth at location E yielded rates of 110 cm/hr (43.5 in/hr) and 88 cm/hr
(34.5 in/hr), respectively. However, the presence of gravel at these depths
prevented the desired 0.15 meter (6 inch) penetration of the double ring
infiltrometer, which may have contributed to the very high observed rates.
A soil log 2.44 meters (8 feet) in depth revealed the same gravel-sand soil
mixture from a 0.51 meter (20 inch) depth to the invert of the underdrain
system at approximately 2.44 meters (8 feet). It can be seen in Figure 17
that a significant increase in the measured infiltration rate was noted once
the surface soil layer of silty loam was penetrated. Because of the rela-
tively homogeneous nature of the soil below this point, it was expected that
the infiltration rate would become relatively constant at depths of greater
than 0.61 meters (2 feet). Based on the favorable infiltrometer results,
and the successful operation at a loading rate of 30.5 m/yr (100 ft/yr), it
was decided to continue loading Bed 1 at 30.5 m/yr (100 ft/yr) with no
physical alteration of the bed surface.
Infiltration tests on the surface of Bed 2 at points A, D, and E
yielded infiltration rates ranging from 0.10-1.41 cm/hr (0.04-0.16 in/hr),
with an average rate of only 0.23 cm/hr (0.09 in/hr). The soil on the sur-
face of this bed appeared to be a silty loam with some heavy clay. At
depths of 0.46 meters (18 inches) and 0.76 meters (30 inches), the measured
infiltration rates increased to 0.89 cm/hr (0.35 in/hr) and 0.99 cm/hr
(0.39 in/hr), respectively. These measurements were made at points C and B.
The soil texture at each location was similar to the silty-clay loam of the
surface soil, but also contained some sand. At point H and a depth of 1.07
meters (42 inches), the soil was a sandy loam, and an infiltration rate of
22.6 cm/hr (8.89 in/hr) was observed. A soil log on Bed 2 showed that
gravel and a sandy loam existed from approximately 0.91 meters (3 feet) to
the invert of the underdrain system at about 2.44 meters (8 feet). A plot of
the infiltration rate as a function of depth is shown in Figure 18. From
this figure it was determined that the top 0.91-1.07 meters (3-3.5 feet) of
overburden were unsuitable for an infiltration-percolation system and should
be removed.
As has been indicated, the method which was ultimately utilized to
physically alter Bed 2 consisted of stripping the top 0.61 meters (2 feet)
of soil and then constructing a ridge and furrow system, as was outlined in
Section 4. Following bed modification, six infiltration tests were per-
formed on the bottom of the 0.46 meter (18 inch) deep furrows, at the
locations numbered 1 through 6 in Figure 16. The results of these tests
yielded infiltration rates which varied from 3.2 cm/hr (1.25 1n/hr) to
27.9 cm/hr (11.0 in/hr), with an average of 12.1 cm/hr (4.76 in/hr). The
predominant soil type found in the bottom of the furrows was a sandy loam.
Loading on Bed 2 following the surface modification was limited to the
amount of wastewater which filled the furrows without overflowing onto the
ridges. These loadings were applied on a bi-weekly basis. This practice
34
-------
TABLE 5. SUMMARY OF INFILTROMETER TEST RESULTS
Test
Bed No. Location
1
Soil Log 1:
0-0.51 m - silty-loam
0.51-2.42 m - sand and
gravel
2
Soil Log 2:
0-0.91 m - silty-loam
w/ heavy clay
0.91-2.43 m - gravel
and sandy loam
3
Soil Log 3:
0-0.46 m - silty-sandy
loam w/clay
0.46-2.44 m - sandy loam
A
D
F
B
C
E
A
D
E
F
G
C
B
I
H
1
2
3
4
5
6
A
F
c
D
G
E
2
3
4
5
6
1
B
Depth from
Original Surface
(meters)
Surface
Surface
Surface
0.051
0.381
0.660
Surface
Surface
Surface
0.051
0.051
0.457
0.762
0.864
1.067
1.067
1.067
1.067
1.067
1.067
1.067
Surface
Surface
0.254
0.254
0.457
0.330
0.457
0.457
0.457
0.457
0.457
0.559
0.610
Infiltration
Rate
(cm/hr)
5.5
0.6
7.1
0
110.5
87.6
0.20
0.10
0.41
0.13
0.30
0.89
0.99
-
22.6
27.9
41.9
9.1
3.5
16.7
3.2
2.0
0.53
0.97
4.2
52.0
18.1
4.7
3.0
1.3
3.5
15.2
11.4
38.1
35
-------
Soil Depth (m)
0.2 0,4 0.8 1.0
O)
O
120
100
80
0)
60 §
40
20
2 3
Soil Depth (ft)
Figure 17. Infiltrometer test results for Basin 1.
36
-------
25
20
£ lOh
0.2
Soil Depth (m)
0.4 0.8
1.0
±
60
50
40
(O
ce.
30 o
+»
£
•p
•r—
q-
20 ~
10
"01234
Soil Depth (ft)
Figure 18. Infiltrometer test data on Basin 2 prior to modification,
37
-------
resulted in an effective loading rate of 41.2 m/yr (135 ft/yr) over the
furrow bottom area, or 12.8 m/yr (42 ft/yr) over the entire area of Bed 2.
This proved to be a successful operating level.
Surface infiltration tests performed on Bed 3 at locations A and F
resulted in infiltration rates of 2.0 cm/hr (0.78 in/hr) and 0.53 cm/hr
(0.21 in/hr), respectively. Location A consisted of silty-sandy loam, while
location F consisted of a clay-silt loam. At a depth of 0.25 meters
(10 inches), an average infiltration rate of 2.59 cm/hr (1.02 in/hr) was
observed, and the soil was predominantly a silty-sandy loam. These tests
were made at points C and D. Infiltration tests performed at sites E and G
were at a depth of 0.46 meters (18 inches), and yielded an average rate of
35.0 cm/hr (13.8 in/hr). At the 0.61 meter (24 inch) depth, an infiltration
rate of 38.1 cm/hr (15.0 in/hr) was measured at point B, with the soil con-
sisting primarily of a sandy loam containing some clay. Figure 19 summarizes
these results graphically as a plot of the variation of infiltration rate
with soil depth.
In order to improve the hydraulic performance of Basin 3 it was decided
that the top 0.46 meters (1.5 feet) of tight soil should be removed The
stripping operation was performed without developing a ridge and furrow
system as was constructed in Bed 2. Following the stripping, surface infil-
tration tests were performed at each of five representative locations,
points 2-6, within Bed 3. The infiltration rates measured at these sites
yielded an average infiltration rate of 5.6 cm/hr (2.19 in/hr). The soil at
this depth was predominantly a sandy loam with varying amounts of clay also
present. The loading rate of 48.8 m/yr (160 ft/yr) which was applied fol-
lowing bed modification was determined by an empirical method used for esti-
mating loading rates for infiltration-percolation systems (Personal communi-
cation, R.E. Thomas). The loading rate was established at ten percent of
the theoretical loading for one year at the average infiltration rate of
5.6 cm/hr (2.19 in/hr) as shown below:
Loading rate = 0-1(2.19 in/hr) (8760 hr/.vr).
Loading rate = 160 ft/yr =48.8 m/yr
When loading was resumed, Bed 3 was erroneously loaded at a rate of 57.3 m/yr
(188 ft/hr) for 3 weeks because of a failure to account for the reduced sur-
face area developed as a result of incorporating a side slope into the
excavation. At this loading rate, the bed failed to drain between succes-
sive loading cycles. When the loading was reduced to 48.8 m/yr (160 ft/yr)
on the actual area, the basin operated In a successful manner.
Following the physical alteration of Beds 2 and 3, infiltration rates
of the system were recorded during the operation of all three beds. This
was done by measuring the change in water level with time after inundation
of each of the beds. The average infiltration rates were determined by
dividing the decline in water level by the elapsed time. A summary of these
infiltration rates is shown in Table 6 for the three beds. The average
38
-------
25
20
- 15
o
•^
•P
S 10
Soil Depth (m)
0.2 0.4 0.8
1.0
60
50
40
30
20
10
§
O£
4J
01234
Soil Depth (ft)
Figure 19. Infiltrometer test data on Basin 3 prior to modification.
39
-------
TABLE 6. SUMMARY OF INFILTRATION RATES
Loading
Date
M3
Loaded
Infiltration
Rate (cm/hr)
Bed 1
2-8-77
2-15-77
2-22-77
3-1-77
3-8-77
3-15-77
4-12-77
4-19-77
5-10-77
5-17-77
5-31-77
6-7-77
6-14-77
6-21-77
6-28-77
7-5-77
Average
1024
1256
1158
1166
1193
1240
817
1146
1195
957
986
1079
1212
1257
1239
1114
1128
0.99
0.84
0.84
0.64
0.48
0.48
2.5
0.38
2.9
0.71
2.8
2.0
1.7
1.6
1.4
1.3
1.3
Loading
Date
M3
Loaded
Infiltration
Rate (cm/hr)
Bed 2
4-13-77
4-20-77
4-27-77
5-4-77
5-18-77
6-1-77
6-8-77
6-15-77
6-22-77
6-29-77
7-6-77
Average
255
273
-
263
261
238
320
248
250
525
253
289
9.5
2.2
2.4
2.5
4.8
5.1
4.8
5.4
5.4
3.3
5.1
4.6
Loading
Date
M3
Loaded
Infiltration
Rate (cm/hr)
Bed 3
2-10-77
2-17-77
2-24-77
4-14-77
4-21-77
5-12-77
5-19-77
6-2-77
6-9-77
6-23-77
6-30-77
7-7-77
Average
1262
997
1330
1220
984
1191
1080
1186
1160
946
1169
1147
1139
3.2
0.58
0.58
0.61
0.56
2.3
1.1
2.3
1.9
1.6
1.4
1.2
1.4
-------
infiltration rates are also shown for Beds 1, 2, and 3 at 1.4 cm/hr
(0.53 in/hr), 4.6 cm/hr (1.81 in/hr), and 1.45 cm/hr (0.57 in/hr),
respectively.
Comparison of the recorded infiltration rates with the average rates
obtained in the infiltrometer tests shows a consistent relationship between
the two. The average infiltration rate for the surface of Bed 1 was found
to be 4.4 cm/hr (1.74 in/hr) in the infiltrometer tests. The corresponding
recorded infiltration rate for Bed 1 was 1.4 cm/hr (0.53 in/hr), or 30% of
the infiltrometer test value. Similarly, the recorded infiltration rate for
Bed 3 was 26% of the infiltrometer test value, and the recorded infiltration
rate for Bed 2 was 38% of the infiltrometer test value.
The higher infiltration rates obtained from the infiltrometer tests may
have been associated with any of several factors. However, the major factor
which contributed to the high rates was believed to be lateral flow from the
infiltrometer rings. The combination of lateral and vertical flow through
the soil medium served to increase the rate at which the soil would accept
the applied water. The U.S. Salinity Laboratory Staff has reported that the
impact of lateral flow increases as the infiltration area decreases (7). The
infiltration area used in the Boulder infiltration-percolation project was
very small compared to the total area of the basins. Although lateral flow
was minimized by the outer buffer pond in the double ring infiltrometer, it
was still felt to be primarily responsible for the higher rates observed in
the infiltrometer studies.
FACTORS AFFECTING INFILTRATION RATES
The infiltration rates in an infiltration-percolation system are
apparently affected by operation and management practices, and wastewater
characteristics, 1n addition to site conditions at a specific location, as
reflected by the data collected during this study.
Operation and management practices of an infiltration-percolation
system consist mainly of the loading schedule, the basin surface management,
and the method of wastewater application. Varying the loading schedule,
i.e., the inundation and drying period of an infiltration-percolation system,
has been shown to affect the infiltration rates of that system. The extent
of this impact is difficult to determine, but it has been stated in the
literature that maximum infiltration rates are achieved by shorter inunda-
tions and longer drying times (8,9,10,11).
The longer drying times allow for aeration of the soil, and enhance the
decomposition and dessication of organic material deposited during the inun-
dation period. The microbial population will also decrease as substrate is
utilized, thereby increasing the pore space available for infiltration (11).
In the Boulder study, the loading schedule was not intentionally varied, but
observations made during bed operation showed the approximate drying times
for Beds 2, 1, and 3 to be 3 days, 2 to 2.5 days, and 1 to 2 days, respec-
tively. A plot of the recorded infiltration rate as a function of time is
41
-------
shown for Basins 1, 2, and 3 in Figures 20, 21, and 22, respectively. From
these figures it can be seen that the infiltration rate following drying and
scarification was the highest. In Basins 1 and 3, the high initial infiltra-
tion rate was followed by a significant rate decrease as the loading cycle
progressed. The high initial infiltration rate was attributed to the soil
aeration and scarification procedure which served to remove any organic
matter that may have formed during the preceding loading cycle. The highest
infiltration rate for Basins 2 and 3 followed a long drying period which
included the period of bed modification. These high rates, observed on
February 10, 1977 for Basin 3, and April 13, 1977 for Basin 2, resulted primarily
from the bed modifications which created a new soil-water interface for
infiltration. Since the literature suggests that soil clogging is a surface
phenomena, it follows that stripping of the beds removed the clogged surface
and created a "fresh" surface (12, 13).
The effect of basin surface management on the infiltration rate was
investigated to only a limited degree at the Boulder site. Basin surface
management practices may include the growth of vegetation on the surface,
placement of a layer of sand or gravel, or operation with a bare soil sur-
face. Surface management may also encompass the type and frequency of scari-
fication during the drying periods. All beds in the Boulder system had bare
surfaces, so comparison with vegetated or gravel surfaces was not possible.
The scarification practiced in this study undoubtedly contributed to the
restoration of high infiltration rates observed at the beginning of each
loading cycle. However, since the same type and frequency of scarification
was used throughout the operation of the basins, comparison with other methods
was not possible.
A gradual decline in infiltration rates was exhibited as the loading
cycle progressed on Beds 1 and 3. Many factors likely contributed to this
decline in infiltration rates. Other investigators have suggested that the
major factors contributing to such declines are the accumulation of suspended
solids, and the microbial activity which is effective 1n degrading the
applied organics. However, Bed 2 did not exhibit these declining trends,
even though it was subjected to application with the same strength waste-
water. The differences in operation between basins which may have caused the
infiltration rate in Basin 2 to remain constant were: (1) the water was
applied to Beds 1 and 3 by total surface flooding, and (2) Bed 2 was loaded
at a lower loading rate than Beds 1 and 3. Although Bed 2 was not scarified
after the drying period, while Beds 1 and 3 were, this fact would only
affect the initial infiltration rate and would not be responsible for main-
taining infiltration rates during the loading cycle. The fact that Bed 2
was a ridge and furrow system may explain why infiltration rates were main-
tained during the loading cycle. It has been suggested by McGaughey that a
ridge and furrow system would not be affected by suspended solids to the
same degree as a flat surface (8). Since the solids that settle will only
alter the surface at the bottom of the furrow, the sides of the furrow
remain relatively free of clogging solids and, therefore, allow the Initial
Infiltration rate to be maintained. Figures 23, 24, and 25 show the Infil-
tration rate plotted as a function of accumulated suspended sol Ids for each
of the three basins. From these figures it can be seen that the accumulated
42
-------
2.0
co
(U
>o
CtL
O
1.5
1.0
0.5
0
i.
.c
E
(U
C
O
4J
-------
Drying and
Bed Modification
u
to
cc.
c
o
(O
S-
M-
c
2.0
July
12 19 26 2 9 16 23 30 6 13 20 27 6 13 29 27 10 17 24 1 8 15 22 29 5 12 19 26 3 10 17
December January February March April May June
1976 1977 1977
Time (weeks)
Figure 21. Pattern of infiltration rate decline in Basin 2.
-------
cn
o
0)
4->
(0
CC.
O
£
0
12 19 26 2 9 16 23 30 6 13 20 27
December January February
1976 1977
6 13 20 27 3 10 17 24 1 8 15 22 29 5 12 19 26 3 10
March April May June July
1977
Figure 22. Pattern of infiltration rate decline in Basin 3.
-------
2.0
1.5
Q)
rtJ
o:
to
t-
1.0
0.5
0
0
100
_L
1000
Accumulated Suspended Solids (mg/m2)
200 300 400 500
1
T
T
LOADING CYCLE OF 5-31-77 to 7-5-77
600
_L
JL
700
5.0
01
4->
-------
2.0
1.5
O
1000
2000 3000 4000 5000 6000 7000 8000
Accumulated Suspended Solids (mg/ft2)
Figure 24. Infiltration rate as a function of suspended solids loading in Basin 2.
-------
00
100
2.0
IT
JC
3 1,5
o>
+J
-------
suspended solids did not appear to have an effect on the infiltration rates
in Basin 2, but appeared to affect the Infiltration rates of Basins 1 and 3.
Based on these data, it was concluded that the existence of a ridge and fur-
row system in Basin 2 was partially responsible for maintaining the relatively
high initial infiltration rates.
The lower loading rate applied to Bed 2 was undoubtedly responsible, in
some part, for the relatively constant infiltration rates. This lower load-
ing rate permitted longer drying times between the loading times on Bed 2,
when compared to Beds 1 and 3. As stated previously, Bed 2 was dry for
approximately 3 days while Beds 1 and 3 were dry from 2 to 2.5 days, and 1 to
2 days, respectively. These longer drying times allowed for a more complete
aeration of the soil in Basin 2 and enhanced the decomposition and dessication
of the organic material deposited during the flooding. Thus, more pore vol-
ume became available for the transport of water, resulting in the mainten-
ance of initial infiltration rates throughout the cycle.
The influent temperature also appeared to have some effect on the
infiltration rate. Figure 26 indicates the average infiltration rate over a
complete loading cycle as a function of the average temperature of the
influent over the same period. The high Infiltration rates observed in Basins
2 and 3 immediately following their modification were not included in these
averages because these high rates resulted mainly from the bed modification,
as was previously discussed. The trends illustrated in Figure 26 were attri-
buted to the lower viscosity of the water at higher temperatures, and to the
fact that the quality of the wastewater influent to the basins was generally
better during the warmer weather.
While temperature did seem to affect the infiltrative capacity of the
soil, the cold temperatures encountered did not necessitate discontinuing the
operation of the Infiltration-percolation system. Even though there was some
freezing of the beds during February and March, the ice did not appear to
Interfere with bed performance. The ice layer apparently served to insulate
the underlying water and collapsed as the water level declined. Subsequent
loadings melted the broken ice, after which a new ice layer formed and the
cycle was repeated. Prolonged periods of sub-freezing temperatures could
freeze the beds solid, which would cause severe operational problems. How-
ever, this did not occur in the Boulder situation.
49
-------
JL.U
s_
^.
c
&a
(0 Q£ U • 0
OJ C
> o
1 o
1— 1
1 1 1 1
Basin 1
_
^^°
r^
- ^^"^
^^^^ -
^^^
O
1 1 1 1
0 5 10 15 20
"t
2. of
u
o>
(O
i.og
•r-
+J
OS
t-
•p
c
C
l— r
Ave. Temperature - C°
2.0
*£
c
7 1-5
0) •»->
CO «
o
5 i.o
c
c 1
1— 1
o
i 1 I 1
Basin 2 y°
/
/
/
J
f
J
J
J
f
O _
t
, +
S_
JC
"e
4.0^
•p
IO
a:
c
o
•I—
3.0 «
£
* c
1* ""
i i i i T
0 5 10 15 20
Ave. Temperature - C°
~ 1.0
.c
c
a» as
as ce.
OJ c 0. b
3^ O
2.0^
S.
c
o
1.05
to
fv
-p
c
10 15 20
Ave. Temperature - C°
Figure 26. Infiltration rate as a function of wastewater temperature.
5Q
-------
SECTION 7
TREATMENT PERFORMANCE OF
THE INFILTRATION-PERCOLATION SYSTEM
GENERAL CONSIDERATIONS
In assessing the treatment performance of the infiltration-percolation
system, the primary focus was directed toward monitoring the fate of the fol-
lowing major wastewater constituents: phosphorus, organics, nitrogen, hard-
ness, alkalinity, and coliform organisms. In addition, the behavior of
selected heavy metals was assessed in a specific short term study. The aim
of the study was to evaluate the treatment efficiency and its variation with
seasonal and climatic changes, as well as long term equilibration effects.
Table 7 summarizes the average seasonal constituent concentrations in
the secondary effluent applied to the beds. Seasonal values were derived
by averaging the constituent concentrations of all the secondary effluent
samples taken during a particular season. For example, the average fall COD
concentration was the average of the COD concentrations in all of the secon-
dary effluent sampled between September 22 and December 22. Some of the
seasonal trends observed in the renovated water quality, e.g., COD, were a
result of the seasonal variation of the pollutant concentrations in the
applied secondary effluent. Other seasonal trends observed, such as the
phosphate level in the renovated water, seemed to be less affected by the
average seasonal pollutant concentration in the applied secondary effluent.
Figure 27 summarizes the annual precipitation pattern as well as the
minimum weekly temperature in the renovated water from Basin 1 for the first
year of study. These data have been presented for reference, since the mini-
mum renovated water temperature reflected an environmental condition within
the soil system that may have affected the removal efficiencies for certain
constituents. The weekly precipitation represents a moisture effect that
may have been effective in causing some of the observed seasonal variations.
PHOSPHORUS BEHAVIOR
Data compiled during the course of this study showed the phosphorus
concentrations in the basin effluents to vary over quite a broad range.
This variation is demonstrated in the total phosphorus concentration profiles
of Figure 28.
In all three of the basins, there appeared to be a significant seasonal
51
-------
TABLE 7. SEASONAL AVERAGES FOR WASTEWATER
CONSTITUENTS IN BOULDER SECONDARY EFFLUENT
Constituent
Total Nitrogen1
Ammonia
Organic Nitrogen
Nitrate
Total Phosphate2
COD
Alkalinity3
Hardness3
Cal ci urn
Magnesium
Chlorides
Suspended Solids
Total Solids
Spring
13.8
7.3
4.7
1.8
4.8
58.0
128
132
34.6
11.0
27.4
14.6
311
Summer
13.2
6.0
4.7
2.5
5.1
58.1
118
154
35.6
14.8
23.2
11.7
333
Fall
14.6
5.3
6.6
2.7
7.2
72.1
125
119
-
-
32.1
21.6
286
Winter
24.5
11.7
12.2
0.6
7.9
118.3
164
112
32.2
7.4
27.3
25.4
302
Annual
16.5
7.6
7.0
1.9
6.2
76.6
134
129
34.1
11.1
27.5
18.3
308
nitrogen concentrations in mg/£ as N
2mg/£ as P
3mg/«, as CaCO
52
-------
CO
C 2.0-
_o
'£
o
0)
2
2.
July August Sept. Oct. Nov. Dec.
1976
a.
•r-
O
Jan. Feb. March April May June July
1977
Figure 27. Weekly precipitation and minimum temperature of renovated water.
-------
in
I I I I I I I I 1 I I I ! I I I I
Basin
Modification
J A S 0 N D
1976
I I I TTTTTT'l
JF N A M J JA S ON D
I I I I 1
J F M A M J
1977
1978
Figure 28. Effluent phosphorus as a function of time.
-------
pattern to the leakage of phosphorus through the basins. The concentration of
total phosphorus appearing 1n the renovated water increased markedly during
the winter loading cycles when the wastewater temperatures were the lowest,
and the influent phosphorus concentration was the highest because of reduced
infiltration in the municipal collection system and lower phosphorus uptake
in the trickling filter system.
It has been demonstrated that phosphates introduced to the soil system
are initially adsorbed in the mineral or organic fraction of the soil
(14, 15). The adsorbed phosphate is subsequently precipitated or mineral-
ized as inorganic compounds into the soil matrix by a slower reaction. This
precipitation reaction apparently serves to release the adsorption sites
within the soil for additional phosphorus removal. The adsorption process
will limit the rate at which phosphorus can be removed by the soil until the
soil adsorption sites become saturated. Further phosphorus removal will be
controlled by the slower rate of precipitation of relatively insoluble com-
pounds (16). The increase in the phosphorus levels applied to the beds
during the fall and winter apparently saturated the phosphorus adsorption
capacity within the soil system. At this point, the slower precipitation
reaction effectively controlled the rate and limited the effectiveness of
the phosphorus removal process. This effect was shown most dramatically in
the winter of 1977-78 after Basins 2 and 3 had been modified. During modi-
fication, Basin 3 had most of the fine grained soil removed from the upper
portion of the soil profile. Following modification, this basin was loaded
at the highest equivalent hydraulic loading rate of 48.9 m/yr (160 ft/yr).
With this combination of basin modification and loading characteristics, the
phosphorus removal efficiency was reduced in Basin 3 to about 40% during
January of 1978. Conversely, Basin 2, which was loaded at the lowest
rate, demonstrated a much higher and more consistent phosphorus removal
capability. These data suggest that by increasing the detention time in the
soil system, the phosphorus removal efficiency might be controlled at a very
high level.
Analysis of the phosphorus profile for Basin 1 indicates that the phos-
phorus leakage peaked during both winters of operation. However, the
effluent concentration during the second year of operation was about twice
that observed during the first winter. These data suggest that the system
had not fully equilibrated in the first year of operation. This finding is
consistent with the findings reported from the Lake George, N.Y. system by
Aulenbach (17, 18). Apparently, the regeneration of adsorption sites by
mineralization is not complete, and a continual degradation of system per-
formance can be expected in phosphorus removal.
In addition to the seasonal trends which have been discussed, signifi-
cant variation was also noted in the phosphorus leakage during different
loadings within a cycle. In most of the loading cycles, the concentration
of total phosphate in the renovated water increased from the first loading
of the cycle to the last. Selected examples showing this trend have been
presented in Figure 29. In keeping'with this trend, the total phosphate
concentration in the renovated water from the first loading of a loading
cycle was usually less than the phosphate concentration in the renovated
water from the last loading of the previous loading cycle. These
55
-------
en
2.0
s-
o
CL
I/)
1.0
I I
ol
8/30 9/6 9/13 9/20 10/11 10/18 10/25 11/1 11/8 11/29 12/6 12/13 12/20 12/27 1/3
1976 1976 1976 1977
(A) Basin 1
10/14 10/19 10/26 11/2 8/25 9/1 9/8 9/15 6/2 6/9 6/16 6/23 6/30 7/7 8/4 8/11 8/18 8/25
1976 1976 1977 1977
(B) Basin 2 (C) Basin 3
Figure 29. Effluent phosphorus variation within loading cycles.
-------
observations can be explained in terms of the slow precipitation of phosphate
in the soil matrix. During the resting period between loading cycles,
rejuvenation of the soil phosphate adsorption capacity apparently occurred.
The soil was able to remove the initial application of phosphate most
efficiently during the first loading of the new cycle. The subsequent appli-
cations of effluent during the cycle showed that the capacity of the soil to
remove phosphate was reduced as the cycle continued. By the last loading of
the loading cycle, the soil adsorption capacity for phosphates was in its
most exhausted state, and the greatest leaching of phosphates through the
soil system occurred at that time.
REFRACTORY ORGANICS
Refractory organics were monitored during the study period by sampling
the secondary effluent during application to each bed, and the renovated
water from the basin 24 hours after application commenced. Figure 30 pre-
sents a summary of the average COD level for each loading cycle in the basin
effluents. As can be seen from this figure, the infiltration-percolation
system was generally effective in reducing the product COD to a residual
level of 10-20 mg/x,. These levels represented a reduction of the influent
COD by 70-80%. The only significant deviation from this treatment pattern
occurred in the December-February period of 1976-77. During that time the
influent had unusually high levels of COD which were attributed to highly
concentrated industrial waste discharges into the Boulder Treatment Plant.
The successful operation of the basins through the winter demonstrated
the fact that cold conditions did not reduce the capability of the soil for
removing organics from the applied wastewater effluents. This performance
supports the findings of other authors regarding the temperature insensi-
tivity for organic removal in high rate systems. In one study of this
phenomenon, Thomas, et al. loaded six silica sand lysimeters with secondary
and septic tank effluent to evaluate the effect of temperature on COD
removals (19). Two lysimeters were incubated at 28°C, two at 18°C to 30°C,
and two at 18°C to 35°C. All six lysimeters produced comparable COD removal
efficiencies. Schwartz, et al. also reported that cold climate did not
restrict the COD removal capability of a high-rate land treatment system.(20).
The relative positions of the three curves in Figure 30 suggests that
the organic content of the product water may bear some relationship to the
hydraulic loading Of the basin. Following bed modification, Basin 3 was
loaded at the highest rate, and the product water from this basin was
typically the highest in COD. Consistent with this observation, Basin 1, the
next most heavily loaded, had a lower effluent COD, and Basin 2 had both the
lowest hydraulic loading and the lowest product COD level.
NITROGEN
Nitrogen was the most difficult constituent to evaluate with respect to
the treatment performance of the infiltration-percolation basins. The reason
57
-------
en
00
o
8
1 I—I—I
o Basin 1
Basin
Modification
JASONDJFMAMJJASOND
1976
1977
J F M A M J
1978
Figure 30. Effluent COD as a function of time.
-------
for this is that the nitrogen cycle contains several pathways for transfor-
mation of the organic and ammonium nitrogen applied to a soil system in
secondary effluent. The three major sequences involve the following
reactions: (1) The ammonium ions may be held in the soil by adsorption or
fixation processes; (2) The ammonium ions retained within the soil by adsorp-
tion or fixation may be nitrified to nitrate and leached through the soil
system with the renovated water; and (3) The nitrate, once formed, can be
denitrified in the presence of a carbon source and anoxic conditions. The
The adsorption and nitrification processes were clearly the most important
during the operation of the Infiltration-percolation basins in this study.
However, the relative Importance of the fixation process, which allows
Indefinite nitrogen storage, and the den1tr1fication process, was difficult
to evaluate.
Nitrate and ammonium ions were the principal nitrogen forms discharged
in the basin effluents, with the nitrate ion accounting for about 98% of the
total nitrogen discharged from Basin 1. Figures 31 and 32 show the average
ammonium ion and nitrate concentrations, respectively, for each of the load-
ing cycles during this study. The average nitrate and ammonium ion levels
for a given weekly loading period were calculated by a flow weighting pro-
cess. This involved development of a hydraulic discharge curve by plotting
the hydraulic discharge from the basin as a function of sampling time. The
total basin discharge was determined from this figure by integrating under
the hydraulic discharge curve. The total pounds of nitrogen discharged by a
basin during the week were determined by a similar method. A nitrogen mass
flow curve was generated by plotting the product of the ammonium or nitrate
nitrogen concentration and the flow rate as a function of sample time. Fol-
lowing normalization of units, the area under this curve yielded the total
pounds of the ammonium or nitrate nitrogen discharged by the basin during
the loading period. The average ammonium or nitrate concentration was found
by dividing the pounds of ammonium or nitrate nitrogen discharged by the
total hydraulic discharge.
Through an analysis of the input and output nitrogen quantities, a
nitrogen balance was made for Basin 1 which covered the first full year
of operation. Figure 33 shows the cumulative nitrogen balance through
Basin 1 during this period. The upper line is the cumulative total nitrogen
applied, and the lower line is the cumulative nitrogen discharged from the
basin. The difference between the lines at any given date shows the total
nitrogen storage or loss in the basin from July 12, 1976 through the loading
cycle beginning on that date.
From these curves it is apparent that the basin tended to store nitro-
gen during the cold winter months of the first year and release it as the
water temperature wanned in the spring. This figure suggests that the
nitrification process was inhibited during the coldest period and ammonium
nitrogen was stored by fixation or exchange 1n the basin. With the onset of
the warming trend, nitrification of the ammonium stored 1n the basin seemed to
be more complete, and nitrogen was released from the basin as the mobile
nitrate Ion. This observation Is reinforced by the increased NH* leakage
59
-------
0?
U>
tTt
O
5 -
4 -
J •*••*
2 -
0
I
Basin Modification
J A S 0 N D
1976
JFMAMJJASONDJFMAMJ
1977
1978
Figure 31. Effluent ammonium nitrogen as a function of time.
-------
CT>
I co
O
01
I—I—I—I—I—I
Basin Modification
"- 10 —
i • i AW
JASONDJFMAMJJASOND
J F M A M J
1976 1977 1978
Figure 32. Effluent nitrate nitrogen as a function of time.
-------
2000
ro
1000 -
- 800
o Nitrogen Applied
• Nitrogen Discharged
Figure 33. Cumulative nitrogen applied and discharged from Basin 1.
-------
from Basins 1 and 2 during the colder winter months. A similar observation
and explanation has been offered by Bouwer, et al. (21).
"This seasonal trend is probably caused by greater drying of
the soil between flooding periods in the summer than in the winter,
which allowed more oxygen to diffuse to greater depths in the soil
in the warmer months. Thus, more adsorbed NHj could be nitrified
in the summer, causing more NH$ to be adsorbed during flooding and,
consequently, a decrease in the NHiJ level in the renovated water.
This continued until the fall when poorer drying limited the amount
and depth of penetration of oxygen. The seasonal trend could also
be the result of the higher amount of SS in the effluent in fall and
winter. These solids began to accumulate on the soil during winter
and early spring. The resulting sludge layer could have acted as
an oxygen sink during drying, thus reducing the amount of oxygen
entering the soil."
This comment was made in relation to the Flushing Meadows system, but
also appears applicable to the infiltration-percolation system operated by
the City of Boulder. A plot of the suspended solids in the secondary efflu-
ent during the study period, Figure 34, did show that an increase in
suspended solids occurred in the winter, as was the case at Flushing Meadows.
Seasonal nitrogen data from the Lake George system demonstrated the same
general variations in bed performance which have been noted in this study.
Ammonium concentrations taken from two sample wells were highest in the win-
ter and lowest in the summer. Conversely, the nitrate concentrations were
generally higher in the summer and fall. A similar ammonium leakage trend
occurred during the operation of the Brookings infiltration system (22). The
ammonium ion concentrations in the product water were lowest in the summer
and highest in the winter. This trend likely reflects both the variation of
the ammonium concentration in the applied wastewater, as well as the seasonal
variation in the ability of the basins to convert ammonium to nitrates.
The nitrogen behavior in the basins during the second year of operation
showed a somewhat less predictable pattern than that of the first year. The
level of nitrification seemed to be higher and somewhat more consistent
throughout the year, although there was a slight tendency toward peaking of
effluent nitrates in the spring, and increased leakage of ammonium ions dur-
ing the coldest winter months. However, in Basin 3, which was the most
heavily loaded during this period, there seemed to be some very
significant departures from the expected pattern. The ammonium level in the
effluent was significantly higher than in the other two basins, and increased
rather dramatically during the last 7-8 months of operation. This was
accompanied by a nitrate level that was substantially below that of the dis-
charge from Basins 1 and 2. These data Indicate that the basin was unable to
effect nitrification of an increasing amount of the applied nitrogen. This
pattern seems to suggest that the basin was being stressed to the point that
the hydraulic, ammonium, and organic loads exceeded the capacity of the sys-
tem for transferring adequate oxygen to permit complete oxidation of the
applied wastewater ammonium. Clearly, this loading would have to be reduced
to provide acceptable ammonium removals on a long term basis.
63
-------
30
T—i—I—i—r
T—i—i—i—i—r
i—i—i—r
20
0)
4J
«/»
-------
In addition to the observed seasonal trends, nitrate and ammonium
concentrations in the renovated water followed a general pattern within the
individual loading cycles. The average nitrate concentrations tended to
decline with successive loadings within a cycle. Figure 35 shows data selec-
ted from loading cycles that clearly illustrate this trend. The maximum
nitrification occurred when oxygen penetrated deepest into the basins during
maximum drying. For this reason, the maximum nitrate concentrations in the
renovated water occurred during the first loading following the extended dry-
ing periods. Subsequent loadings during the cycle tended to limit the basin
aeration and reduced the degree of nitrification that could occur. As a
result, the basins produced lower nitrate levels as the loading cycle
continued.
A reverse trend was observed with respect to the ammonium ion concentra-
tion, particularly during the winter loading cycles. This is shown for two
cycles in Figure 36. As shown in this figure, the ammonium ion level in the
product water increased with each successive inundation within a loading
cycle. This trend was apparently caused by a reduction in the available
ammonium adsorption capacity within the soil. Rejuvenation of the soil ammo-
nium adsorption capacity generally occurred during the drying period, when
the ammonium ions held on adsorption sites were nitrified. This nitrifica-
tion freed the adsorption sites in the soil for readsorption of ammonium ions
during subsequent loadings of the basin. When ample drying of the basin did
not occur between inundations, rejuvenation of the soil ammonium adsorption
capacity did not occur. In these cases, greater amounts of ammonium leached
through the soil profile with each succeeding loading period, resulting in
an increase in ammonium concentration in the renovated water with successive
loadings of the loading cycle. The most obvious examples of this trend
occurred during the winter when an abnormally high ammonium concentration
was present in the applied effluent. During this time, the ammonium adsorp-
tion capacity was the most heavily taxed because of high ammonium loading and
low rates of nitrification. Apparently, the ammonium adsorption capacity of
the soil was not fully taxed during the other seasons, and the indicated
trend did not occur except in the winter.
DISSOLVED SALT SPECIES
Water percolating through the infiltration-percolation soil system
carried with it soluble salts. The dissolved inorganic constituents moni-
tored in this study were hardness, calcium, magnesium, alkalinity, and
chloride. An effort was made to correlate the concentrations of the various
salt constituents leached from the soil with operating conditions and the
seasonal variations of environmental conditions.
The average hardness concentrations of the renovated water, and the
changes observed in hardness as the water percolated through the basins are
shown in Figures 37 and 38, respectively. The most pronounced characteris-
tic of both these curves was the peaking pattern observed during the period
February-May, 1977. This peaking was apparently related to the addition of
alum in the secondary treatment system for suspended solids control during
65
-------
10
i
15
10
I I I I
9/ 9
13 /20 15*
i i i
1976
1977
10
(0
.oi
£
i i i r
I i i
i i^»^
i i i i
1976
1977
Basin 1
Basin 2
Basin 3
1976 1977
Figure 35. Nitrate variation in effluent within loading cycles.
66
-------
2.0
I r
E
1.5
1.0
0.5
Basin 1
I I I I I
2/8 2/15 2/22 3/1 3/8 3/15
1977
2.0
1.5
re
o?
1.0
0.5
T
2/10
2/17
1977
2/24
Figure 36'. Effluent ammonium variations with loading cycles.
67
-------
350
00
vt
"O
- Basin 2
- Basin 3
0—av-fr-..
150 -
100
JASON DJFMAMJJASOND
1976
1977
J F M A M J
1978
Figure 37. Effluent hardness as a function of time.
-------
250
VO
en
to
ca
•o
I I I I I I I I 1 I I I I I I I | I I I I I
Basin 1
---- Basin 2
O—— Basin 3
J F M A M
°T 150 -
en
1
100 -
50 _
J A
ONDJFMAMJ
1978
Figure 38. Change in hardness concentration through the basins.
-------
this period. With the reduction in pH which accompanies alum addition, it
was not surprising to observe increased leaching of hardness ions. The
increased uptake of calcium and magnesium from the soil profile can also be
seen by analyzing the changes in concentration of these two constituents from
the influent to the effluent, as is shown in Figures 39 and 40.
Accompanying the release of the cations causing hardness, the alkalinity
of the water collected in the underdrain system also increased markedly dur-
ing the period February-May, 1977. This is shown very clearly in the curves
of alkalinity change presented in Figure 41. The behavior of both the hard-
ness and alkalinity constituents during the period of chemical addition
apparently showed the impact of solution pH variations on the inorganic qual-
ity of the percolate water.
Apart from the unusual data developed during the spring of 1977, there
seemed to be a general decline in the level of leaching of hardness and alka-
linity from the soil profile. In fact, during the last two loading cycles of
Basins 2 and 3, there was little change in the concentration of hardness and
alkalinity from the influent to the effluent. This would seem to indicate
that an equilibrium condition was being approached between the applied waste-
water and the soil profile. The reasons for the deviation from this pattern
in Basin 1 were not readily apparent.
The chloride concentration of the renovated water was monitored
to assess the level of interaction between the ground water and the
infiltration-percolation system. As can be seen in Figure 42, the chloride
concentration in the water percolating through the basin seemed to show little
change over the period of this study. The only significant exception to this
was during the first six months of operation of Basin 3. During this
period before the bed modification, the infiltration rate was very low, and
the rate of application was continually reduced to the point that the
equivalent loading in December, 1976 was only about 4.6 m/yr (15 ft/yr).
With this low wastewater loading, dilution of the wastewater chloride con-
centration by the limited flow of ground water coming through the clay dike
apparently made a detectable impact on the underdrain chloride concentration.
When the basin surface was modified and the loading rate increased, this
effect was not apparent.
COLI FORM ORGANISMS
During the early stages of the study, grab samples were collected
periodically to monitor the performance of the infiltration-percolation sys-
tem in removing bacterial organisms. The data which summarize these analy-
ses are presented in Table 8. The influent values in this table represent
averages for each of the three basins, while the effluent levels are specific
for each basin. As indicated In these data, the basins were very effective in
making a substantial reduction in the coliform concentrations. The effici-
encies shown compare well with those which have been reported for similar
high rate systems. Nevertheless, the effluent from each basin contained a
significant concentration of residual fecal organisms. The difference in
the effluent fecal coliform levels between the Basin 1 effluent and those in
70
-------
60
u>
CO
'£ 50
40
e
5 30
10
I I I I I
I i I I i 1
i i
I I I I I
J A S 0 N D
I I I I I I I I I i I
JFMAMJJASOND
J F M A M J
1976 1977 1978
Figure 39. Change in calcium concentration through the basins.
-------
30
ro
-P
en
8
.c
4-9
20
0)
u
c
o
0 10
CT>
0)
c
-------
200
CO
lO
o
o
(O
o
(0
.01
0}
CO
0)
1
c
•r-
-------
+15
+10
+5
T I I I—I—I—I—I—I—I—I—I—I—i—\—i—i—|—i—i—i—r
Basin Modification
0?
OJ
O
«v
Basin 1
— Basin 2
n— .— Basin 3
1 1 1 1 1
J A S 0 N D
0
i I
F M A
1
1
M J
I
J A
I I
1 1 1 1
SON
D
J
I i
1 1
F M A
1 1 '
M J
1976 1977 1978
Figure 42. Change in chloride concentration through the basins.
-------
TABLE 8. COLIFORM REMOVAL DATA
Parameter Total Coliforms Fecal Coliforms
(per 100 ml) (per 100 ml)
Composite Influent Avg. 2.175 x 106 1 x 105
Effluent Avg.
Basin 1 5700 3800
Basin 2 4025 900
Basin 3 4250 1025
% Removal Total Coliforms Fecal Coliforms
Basin 1 99.7 96.2
Basin 2 99.8 99.!
Basin 3 99.8 99.0
75
-------
Basins 2 and 3 was likely related to the greater soil permeability and
higher hydraulic loading rate in this basin during the period of collection
of these data.
HEAVY METALS
In a separate short term study phase, the removal of several important
heavy metal wastewater constituents was assessed. The details of the
research approach for this portion of the investigation have been presented
previously in Section 5.
The results of this heavy metal investigation have been summarized in
Figures 43 through 48. Each figure presents the concentrations of the dis-
solved, suspended, and total metals in the influent to the basins, and the
total metal concentration in the effluent from the basins. The concentrations
are averages of six determinations from each bed over a six week loading
cycle. The data from all three of the basins were plotted on a single figure
for each metal. The figures show the part per million concentrations as bar
graphs with the means, standard deviations, and ranges tabulated below. The
percent removal of each metal during infiltration-percolation was calculated
and presented next to the mean value of the metal concentration in the
effluent. The concentration of the metals are given in parts per billion (p.p.b.).
These data show that all of the metals studied were removed to some
extent from the wastewater during the infiltration-percolation process. The
levels of removal ranged from high efficiencies of greater than 80% for lead
and zinc, to low levels of less than 50% for nickel and chromium.
As seen in the previous figures, the metals which occurred in the Boul-
der wastewater were rather evenly distributed between the dissolved and sus-
pended fractions. In fact, in several cases the measured mean concentrations
of metals in the dissolved fraction exceeded those in the suspended fraction.
This was not expected since it has been reported in the literature that heavy
metals are usually strongly associated, with the suspended fraction. However,
in most cases the results which have been reported were from studies with
either sludges or raw wastewater having much higher suspended solids
concentrations.
The metals associated with the suspended material undoubtedly accounted
for much of the observed removal. Since most of the suspended solids were
removed from the wastewater during infiltration-percolation, it appears that
a large portion of the metal removal observed in this study was a result of
filtration of the solids. The remaining metal removal was likely due to a
combination of exchange and precipitation reactions.
76
-------
20
15
tL 10
o
o
o
0
Diss
Sus
Influent
Tot.
Tot. (A=68%)
Effluent
Mean
1.26
1.21
2.47
0.78
Stnd.Dev.
Range
1.09
4.50
2.91
11.00
2.65
11.00
1.56
6.00
Figure 43. Wastewater cadmium concentrations.
-------
80 r-
oo
Mean
Diss.
19.75
Sus.
Influent
24.27
Tot.
54.02
Tot. (A=65%)
Effluent
18.75
Stnd.Dev,
13.83
9.67
16.05
8.72
Range
50.00
37.00
52.00
26.00
Figure 44. Wastewater copper concentrations.
-------
80
60
Q.
a. 40
§
o
20
Mean
Diss.
16.93
Sus.
Influent
9.00
Tot.
25.93
Tot. (A-47%)
Effluent
14.60
Stnd.Dev.
Range
14.11
49.00
7.38
22.00
14.10
41.00
9.97
38.00
Figure 45. Wastewater chromium concentrations,
-------
.a
Q.
Q.
O
O
O
00
o
Mean
Diss.
8.59
Sus.
Influent
5.27
Tot.
13.86
Tot. (A=36%)
Effluent
8.94
Stnd.Dev.
5.58
4.36
8.31
4.55
Range
17.00 14.00 31.00
Figure 46. Wastewater nickel concentrations.
12.00
-------
oo
Mean
2.
10.15
Tot. (A=83%)
Effluent
1.75
Stnd.Dev.
Range
2.55
7.00
3.41
11.00
3.75
13.00
2.54
9.00
Figure 47. Wastewater lead concentrations.
-------
400
300
i 200
o
o
00
ro
100
Mean
Diss.
49.17
Sus.
Influent
11.11
Tot.
60.28
Tot. (A=87%)
Effluent
7.94
Stnd.Dev.
Range
41.06
150.00
11.67
25.00
35.76
120.00
15.92
40.00
Figure 48. Wastewater zinc concentrations.
-------
SECTION 8
COLUMN STUDIES
COLUMN OPERATION
In order to assess the patterns of constituent removal with soil depth,
a small column was constructed to simulate the soil profile of Basin 1.
The laboratory column consisted of a 3.7 meter (12 foot) section of 0.20
meter (8 inch) PVC pipe. It was capped on one end-, and a 0.15 meter (6 inch)
layer of gravel was placed in the bottom to serve as an underdrain system.
A 2.1 meter (7 foot)layer of sand was placed directly above the gravel, and
a 0.20 meter (8 inch) silt-loam layer was added on top of the sand. Samp-
ling ports were placed at depths of 0.20 meters (8 inches), 0.56 meters
(1 foot 10 inches), 1.2 meters (3 feet 10 inches), 1.8 meters (5 feet
10 inches) from the top, and at the bottom of the column.
The soil used for the laboratory column study was taken from throughout
the depth of Basin 1. The top silt-loam layer was cohesive and facilitated
collection of an undisturbed sample, while the sand was too loose to permit
collection of undisturbed samples with depth. As a result, composite sam-
ples were collected at several basin depths and Introduced into the column 1n
the same order to simulate the gradation of soils found in the field.
The sample collection apparatus at each port consisted of a 65 mm
funnel attached to a section of rubber tubing at the discharge end. To pre-
vent clogging of the funnel, a small piece of fine screen was placed within
the mouth of each funnel.
Since the amount of sample that could be collected at each sampling was
quite small, it was necessary to make a series of column runs to adequately
evaluate the soil column performance. This series consisted of several
column loading cycles for assessing the organic removal behavior of the soil
column. These were followed by two column runs for evaluating the removal
of phosphorus in the soil. The following two loadings provided the data for
evaluating the flow, nitrate, nitrite, and ammonium profiles. The next
loading was tested for calcium, hardness, alkalinity, and chlorides, and a
run for col 1 form and phosphorus analyses completed the series.
In each series of tests, 0.30-0.33 meters (12-13 inches) of secondary
effluent was applied to the top of the column. During the wastewater appli-
cation, a portable gravel layer was set on top of the soil while the waste-
water was applied slowly through small holes in a plastic bag to prevent
disturbance of the top loam layer.
33
-------
COLUMN PERFORMANCE
The removal of COD, phosphorus, nitrogen, dissolved solids, and
coliforms by the infiltration-percolation soil column was evaluated under a
number of different loading cycles. The results of these evaluations have
been summarized graphically in this section.
Chemical Oxygen Demand
The first column run was made to assess the removal of COD by the pilot
soil column. Samples for testing the levels of COD removal were collected
every 6-12 hours from each of the sample ports. A graphical representation
of the results of this run has been presented in Figure 49. This figure
indicates the COD concentration in the percolate water as a function of depth
and time after loading.
In analyzing the data on this figure, it is apparent that one of the
most significant characteristics of the figure was the large increase in the
COD concentration during passage of the water through the top 0.23 meters
(9 inches) of soil. In fact, the COD of water removed from the 0.23 meter
(9 inch) sampling port never dropped below the soluble COD of the applied
wastewater. Since the COD associated with suspended solids is typically
removed in the top few inches of soil by filtering and straining, it was
expected that the COD in the column would decrease very rapidly to a level
below the soluble COD concentration of the influent (23). However, in this
run there was no net removal of COD until the water had percolated through
approximately 0.46 meters (18 inches) of soil. An increase of COD within the
top few centimeters (cm) of soil has not been reported by previous investigators.
However, it has been noted that the suspended solids are rapidly removed in
the top few centimeters of soil by filtering and straining (23). As oxygen
enters the soil during a subsequent drying period, the retained organics are
usually decomposed (24). Since the oxygen diffusion rates are typically
high at shallow depths, the trapped COD would be expected to decompose
quickly (23). However, in this study the moisture content of the top soil
layer was high during the entire loading schedule. This high moisture con-
tent may have slowed the rates of oxygen transport and thus affected the
rates of organic decomposition (23, 25). This factor may have contributed
to incomplete oxidation of the organics, resulting in some solubilization.
Subsequent loadings of the secondary effluent could have flushed this solu-
ble COD out of the top few centimeters of soil. In time, as the soluble organics
were flushed from the soil, the COD concentration in the collected water
would decrease. This pattern was observed to occur as can be seen by the
curves developed at the later sampling times.
The COD increase in the upper levels of the column had no effect on the
effluent quality. The high initial COD levels in the samples from the top
sampling port were quickly reduced by the underlying sand layer in the
column. The COD concentration decreased with depth to a constant COD con-
centration of approximately 20 mgA at the bottom of the column. The efflu-
ent level of 20 mg/«, of COD represented 50-75% removal of the applied COD.
84
-------
00
2 -
4 -
6 -
8
Soil Influent, 39.9 mg/£
Influent, 72.3 mg/a
Soil Surface
Time of Sampling
60 80
COD (mgA)
Figure 49. COD Concentration variation with column depth and time.
-------
The results are consistent with those of previous studies and the actual field
operations in this study (19, 20, 26).
Phosphorus
The behavior of phosphorus in the soil column was evaluated through
three separate loadings of secondary effluent. Two of the phosphorus runs,
Nos. 6 and 7, were made early in the loading history of the column, and the
third was made after many column loading cycles. This was done to gain some
insight regarding the phosphorus reduction capacity with repeated loading of
a soil matrix. A graphical summarization of the phosphorus depth relation-
ship has been presented in Figure 50.
The three curves plotted in Figure 50 represent the phosphorus concen-
tration of the soil solution as a function of the soil depth for each of
three separate loadings. In all three of these curves the phosphorus con-
centration of the applied wastewater declined sharply in the top silt loam
layer. This highly efficient removal suggests that the silt loam provided
numerous sites for phosphorus adsorption. The rate of phosphorus removal
with increased depth declined significantly when the water reached the sand
layer. This decline in the removal rate could be due to many factors,
although the most probable cause is related to the characteristics of the
sand media. Because of their relatively large particle size, sandy soils
provide relatively few adsorption sites for phosphorus removal (27). As a
result, the effectiveness of sand for phosphorus removal is not as great as
that of finer textured soils.
While the results of the column study seemed reasonable, the effluents
collected in the field operation had phosphorus concentrations which were
significantly higher than those determined in the laboratory column study.
Since the soil used in the column was taken directly from Basin 1, it was
thought that changes in the soil matrix must have occurred. The soil sam-
ples taken for the column study were obtained in the middle of January, 1977,
while the column construction was not completed until the end of June, 1977.
This resting period may have provided time for nearly all of the adsorbed
phosphorus to become mineralized (28). The result of this phosphorus pre-
cipitation was the freeing of adsorption sites for the subsequent fixation
of phosphorus within the column.
Nitrogen
In a separate loading cycle, secondary effluent was applied to the soil
matrix for analysis of the behavior of the nitrogen species. The data
obtained from this run consisted of time profiles at each sampling port of
the ammonium, nitrite, and nitrate concentrations, and the flow quantities.
The ammonium level of the applied wastewater was very low in this phase
of the study, containing only 5.44 mg/a . As can be seen in Table 9, the
ammonium concentration decreased rapidly to less than 1.0 mg/i as the water
moved through the silt loam layer. The concentration decreased further when
the water entered the sand region. Sampling Port No. 3 yielded ammonium
86
-------
Soil Surface
CO
2.0
7.0 8.0
3.0 4.0 5.0 6.0
Total Phosphorus (mg/Jt)
Figure 50. Wastewater phosphorus concentration as a function of soil depth.
-------
TABLE 9. SELECTED NITROGEN REMOVAL DATA (mg/Jl)
Sample Port
Influent
Port 1
Port 3
Port 5
Collection
Time
Hrs. Min.
3
5
8
19
3
5
8
19
3
5
9
19
00
30
55
40
00
30
55
45
00
30
00
50
Nitrite
.48
.46
.70
.14
.02
.02
.01
.01
.02
.05
.01
.01
.01
Nitrate
1.63
16.59
14.00
8.75
3.08
9.95
12.67
13.61
14.48
29.34
34.02
30.42
22.18
Ammon i a
5.44
.92
.77
-
.22
.23
.09
.08
.07
_
.08
.06
.07
-------
levels in the percolate water of less than 0.10 mg/Ji. This indicates that
most of the ammonium ion was adsorbed or fixed in the upper one-half of the
column depth, with little change in the ammonium levels noted below that
point.
As seen in Table 9, the nitrite level of the influent wastewater was
0.48 mgA as N. At the first sampling level this concentration had
increased slightly to levels ranging from 0.32-0.71 mg/st, as N. Below the
first sampling port the nitrite concentration dropped to less than 0.05 mg/x,
and remained at a very low level throughout the rest of the soil matrix.
The increased levels of nitrite at the top sampling port would point to the
initiation of nitrification in the applied wastewater, or incomplete nitri-
fication during the previous drying period (25). As shown by the extremely
low nitrite concentrations at Port No. 2, a majority of the nitrite pro-
duced in the silt loam layer was quickly converted into the nitrate form.
The nitrate form of nitrogen was present at significantly higher
concentrations than the other nitrogen forms throughout the column depth.
The time profiles of the nitrate concentrations in the water from the various
ports are shown in Figure 51. As can be seen from this figure, the concen-
trations of nitrate in the column increased as the water moved through the
soil. In addition, the shape of the nitrogen discharge curves from Port
No. 1 and the effluent, displayed the characteristic first flush nitrate
peaking which has been reported from several other studies of cyclically
loaded soil systems (10, 29).
Flow measurement on the column effluent provided the data for the curve
of discharge as a function of time in Figure 52. These data were used in
conjunction with the effluent nitrogen data to provide the nitrogen mass dis-
charge pattern shown in Figure 53. The pattern shown in these figures com-
pared favorably with that of the actual field experience.
Dissolved Solids
In the portion of this column study dealing with the behavior of dis-
solved solids in soil systems, analyses were made for calcium, magnesium,
total hardness, alkalinity, and chlorides. These determinations were made on
the total water volume collected from each port during one loading of secon-
dary effluent. The data describing the fate of these constituents have been
summarized graphically in Figures 54 through 58. Each figure presents the
variation in concentration of the dissolved material with soil depth.
The curves of alkalinity, calcium, magnesium, and total hardness dis-
played similar variations of concentration with depth. The concentration of
each of these constituents increased substantially in the top portions of
the soil. The rate of concentration increased, then gradually declined with-
in the 0.61-1.2 meter (2-4 foot) level. At depths greater than 1.2 meters
(4 feet) the concentrations of these materials remained essentially constant.
A slight difference in the concentration behavior of these four consti-
tuents was noted in the top silt loam layer. The concentrations of hardness
and calcium increased significantly as the wastewater moved through the fine
89
-------
Nitrate Concentrations
VO
o
30
.«* 20
0)
-IJ
-------
400
300
rO
200
100
0
0 4
Figure 52.
8
I 1 I 1
Flow Rate of Column Effluent
20
28
12 16
Time (hrs)
Column underdrain flow as a function of time after loading.
-------
10
ro
0)
• 11
10
9
8
7
-------
V£>
OJ
Soil Surface
silt-loam
40 50
Ca+2 (mg/*)
Figure 54. Wastewater calcium concentration as a function of column depth.
100
-------
VO
8
0
Soil Surface
T I v
Silt-loam \
* x
Sand
_L
I
I
10
25
I
30
50
100
Q.
150 Q
200
35
15 20
Mg+2 (rng/A)
Figure 55. Wastewater magnesium concentration as a function of column depth.
-------
vo
Ul
8
100
Soil Surface
150 200
Alkalinity (mg/i as CaC03)
250
Figure 56. Wastewater alkalinity as a function of column depth.
0
300
-------
Soil Surface
Silt-loam
0
50
Sand
to
o
CO
8
I
100
150
200
ex
01
100 200 300
Total Hardness (mg/«, as CaCo3)
Figure 57. Wastewater hardness as a function of column depth.
-------
Soil Surface
Silt-Loam
Sand
o
to
8
10
30
15 20 25
Chlorides (mgA)
Figure 58. Wastewater chloride concentration as
a function of column depth.
50
100
150
200
o
00
35
-------
grain soil. However, the alkalinity and magnesium showed little change in
concentration during percolation through the silt loam layer.
Testing for chlorides in the soil column effluents showed variations
similar to those of the other dissolved constituents, although the changes
in concentration were not as large. The chloride concentration increased
rapidly in the top 0.61 meters (2 feet) of soil and remained essentially
constant throughout the remaining column depth.
The observed variations in concentration of the dissolved materials
within the soil matrix can be attributed to a number of reactions (30). The
dissolved material in a soil solution may interact with the soil matrix by
ion exchange, precipitation or dissolution with the solid phase, ingestion
by microorganisms, incorporation into the soil organic matter, and reactions
with the soil air. The large increases observed in the concentration of
alkalinity, calcium, magnesium, and total hardness indicate that a dissolu-
tion process was likely occurring as a result of the contact between the
soil and the applied wastewater. Chemical principles suggest that mineral
dissolution would be expected when a soil solution contains dissolved salts
at a concentration level which is below equilibrium with respect to any
solid phase or mineral present. Thus, the composition of the renovated
water ultimately would be controlled by the solubility of the various miner-
als within the soil matrix. However, for some precipitation and dissolution
reactions, the rates are extremely slow and are controlled by kinetic and
thermodynamic constraints (30). The materials tested in this study have
been classified as being very reactive within the soil matrix (30). This
would account for the very rapid change of concentration within the top
centimeters of soil.
All of the materials tested did display the same characteristics in the
lower regions of the column. In each case, the dissolved solids showed
little change in concentration once the 1.22 meter (4 foot) column depth was
reached. Thus, the reaction rate of the dissolution reaction had slowed and
the soil solution was near equilibrium. This slowing of the reaction rate
would suggest that the soils below the 1.22 meter (4 foot)'level were not
significantly involved in the dissolution reaction.
Coliforms
The concentrations of both fecal and total coliforms were determined
during the study. A summary of these results has been presented in Table 10.
As can be seen in this table, the column removed 95% of the total and fecal
coliforrris applied within the first 0.23 meters (9 inches) of the silt loam
layer. As the wastewater moved further through the soil matrix, the coli-
form count was reduced somewhat more, but at a slower rate. These results
were predicted since bacterial organisms are removed in soil by straining,
die-off, sedimentation, entrapment, and adsorption (23, 24,31). The finer
grained silt loam layer would be expected to provide more locations for
these processes to occur. The bottom layer was much coarser sand, and pro-
vided lower removals by the indicated processes. The results of this column
study compare favorably with other data on the removal of bacteria in soils
(32, 33).
98
-------
TABLE 10. COLIFORM REMOVAL WITH DEPTH
Port 1
Port 2
Port 3
Port 4
Port 5
% Removal
Total
95
99.6
—
99.75
99. 5
% Removal
Fecal
95
99.8
—
99.98
99.98
Heavy Metals
Figure 59 provides a summary of the behavior of heavy metals 1n the col
umn as a function of column depth. It 1s clear from this figure that most
of the heavy metal removal occurred 1n the top 25 cm (10 in) of the soil
profile. It is not apparent from these data whether the decreased removal
below 25 cm (10 1n) was a mass action effect, which would be observed 1n
a homogenous soil column, or whether the surface removal was simply due to
the presence of the sllty-loam soil 1n that region. However, 1t 1s likely
that both mass action effects and soil type effects combined to yield the
observed results, with soil type providing the predominant influence.
99
-------
0
50
100
150
200
250
20
9 Cadmium
w Copper
O Nickel
o Zinc
a Chrome
• Lead
40 60
Cone, (p.p.b.)
80
Figure 59. Wastewater Keavy metal concentration
variations with depth.
100
-------
REFERENCES
1. U.S. Department of Agriculture, Soil Conservation Service, "Soil Survey
of Boulder County Area, Colorado." 1975. 86 pp.
2. CH2M-Hill, Inc. and City of Boulder, Colorado. "Wastewater Facilities
Plan for Boulder, Colorado." 1975. 256 pp.
3. U.S. Environmental Protection Agency. "Methods for Chemical Analysis of
Water and Wastes." EPA-625-/6-74-003, National Environmental Research
Center, Cincinnati, Ohio, 1974. 298 pp.
4. Standard Methods for the Examination of Water and Wastewater. Four-
teenth Edition, Academic Press* New York. APHA, AWWA, WPCF. 1976.
1193 pp.
5. Harada, H.M., G.H. Reid, E.R. Bennett, and K.D. Linstedt. "A Modified
Filtration Method for the Analysis of Wastewater Suspended Solids,"
Journal of the Water Pollution Control Federation. 45:1853-1858. 1973.
6. Haise, H.R., W.W. Donnan, J.T. Phelan, L.F. Lawhon, and D.G. Shockley.
"The Use of Cylinder Infiltrometers to Determine the Intake Character-
istics of Irrigated Soils." U.S.D.A. Agricultural Research Service
No. 41-7. 1956. 10 pp.
7. Johnson, A.I. "A Field Method for Measurement of Infiltration." Paper
1544-F, U.S. Geological Survey Water Supply, U.S. Government Printing
Office. 1963. 25 pp.
8. McGaughey, P.M. and J.H. Winneberger. "Studies of the Failure of Sep-
tic Tank Percolation Systems." Journal of the Water Pollution Control
Federation, 36:593-606. 1964.
9. Laverty, F.B., R. Stone, and F.A. Myerson. "Reclaiming Hyperion
Effluent." J. Sanitary Engineering Division, American Society of Civil
Engineers, SA6:1-40.19617
10. Bouwer, H., R.C. Rice, and E.D. Escarcega. "Renovating Secondary Sew-
age by Groundwater Recharge with Infiltration Basins." EPA-16060-DRV-
03/72, U.S. Environmental Protection Agency, Washington, D.C. 1972.
102 pp.
11. Bouwer, H. "Land Treatment of Liquid Waste: The Hydrologic System."
IN: Proceedings of the Joint Conference on Recycling Municipal Sludges
and Effluents on Land, Champaign, Illinois.1973. pp. 103-112.
101
-------
REFERENCES (continued)
12. Rice, R.C. "Soil Clogging During Infiltration with Secondary Effluent."
Journal of the Water Pollution Control Federation, 46:708-716. 1974.
13. McGaughey, P.M. and R.B. Krone. "Soil Mantle as a Wastewater Treatment
System." SERL Report No. 67-11, University of California, Berkeley.
1967. 200 pp.
14. Hsu, P.M. "Adsorption of Phosphate by Aluminum and Iron in Soils."
Soil Sci. Soc. Amer. Proc.» 28:474-478. 1964.
15. Harter, R.D. "Phosphorus Adsorption Sites in Soils." Soil Sci. Soc.
Amer. Proc.. 33:630-632. 1969.
16. Barrow, N.J. and T.C. Shaw. "The Slow Reactions Between Soil and
Anions: 2 Effect of Time and Temperature on the Decrease in Phosphate
Concentration in the Soil Solution." Soil Sci.. 119:167-177. 1975.
17. Aulenbach, D.B., N.L. Clesceri, T.J. Tofflemire, S. Beyers, and
L. Hajas. "Water Renovation Using Deep Natural Sand Beds." J. American
Water Works Association. 67:510-515. 1975.
18. Aulenbach, D.B., J.J. Ferris, N.L. Clescert, and T.J. Tofflemire.
"Protracted Recharge of Treated Sewage into Sand Part III: Nutrient
Transport Through the Sand." Ground Water. 12:161-169. 1974.
19. Thomas, R.E. and T.W. Bendixen. "Degradation of Wastewater Organics in
Soil." J. Water Pollution Control Federation, 41:808-813. 1969.
20. Schwartz, W.A. and T.W. Bendixen. "Soil System for Liquid Waste Treat-
ment and Disposal: Environmental Factors." J. Water Pollution Control
Federation. 42:624-630. 1970.
21. Bouwer, H., J.C. Lance, and M.S. Riggs. "High-Rate Land Treatment II:
Water Quality and Economic Aspects of the Flushing Meadows Project."
J. Water Pollution Control Federation, 46:844-859. 1974.
22. Dornbush, J.N. "Infiltration Land Treatment of Stabilization Pond
Effluent Technical Progress Report #1 and Addendum." Unpublished.
1976. 44 pp.
23. Pound, Charles E. and R. Crites. Wastewater Treatment and Reuse by
Land Application—Volume II. EPA-660/2-73-006b, U.S. Environmental
Protection Agency, Ada, Oklahoma, 1973. 249 pp.
24. Powell, G.M. Land Treatment of Municipal Wastewater Effluents Design
Factors—II. U.S. Environmental Protection Agency Technology Transfer,
January 1976. 72 pp.
102
-------
REFERENCES (continued)
25. Lance, J.C., F.D. Whisler, and H. Bouwer. "Oxygen Utilization in Soils
Flooded with Sewage Water." Journal of Environmental Quality,
2:345-350. 1973.
26. Robeck, G.G., T.W. Bendixen, W.A. Schwartz, and R.L. Woodward. "Factors
Influencing the Design and Operation of Soil Systems for Waste Treat-
ment." J. Water Pollution Control Federation, 36:971-983. 1964.
27. Hinesly, T.D. "Land Treatment Process for Wastewater Renovation."
Public Works, 105(2):62-66. 1974.
28. Ellis, B.6. "The Soil as a Chemical Filter." "Conference on Recycling
Treated Municipal Wastewater through Forest and Cropland." EPA-660/2-
74-003, U.S. Environmental Protection Agency, Ada, Oklahoma, March,
1974. pp. 47-72.
29. Lance, J.C. and F.D. Whisler. "Nitrogen Balance in Soil Columns
Intermittently Flooded with Secondary Sewage Effluent." Journal of
Environmental Quality, 1:180-186. 1972.
30. Lindsay, W.L. "Inorganic Reactions of Sewage Wastes with Soils."
IN: Conference on Recycling Municipal Sludges and Effluents on Land,
Champaign, Illinois. 1973. pp. 91-96.
31. Gerba, C.P., C. Wall is, and J.L. Melnick. "Fate of Wastewater Bacteria
and Viruses in Soil." Journal Irr. and Drain. Div. Proc. Amer. Soc.
Civil Eng., 101:157-174. 1975.
32. Aulenbach, D.B., T.P. Glavin, and J.A.R. Rojas. "Protracted Recharge
of Treated Sewage into Sand—Part I Quality Changes in Vertical Trans-
port through Sand." Ground Water, 12:161-169. 1974.
33. Page, H.G. and C.H. Wayman. "Removal of ABS and Other Sewage Compon-
ents by Infiltration through Soils." Ground Water, 4:10-17. 1966.
103
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-m
2.
4. TITLE ANDSUBTITLE
TREATMENT OF SECONDARY EFFLUENT BY INFILTRATION-
PERCOLATION
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
August 1979 Issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D. G. Smith, K. D. Llnstedt, and E. R. Bennett
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
1O. PROGRAM ELEMENT NO.
City of Boulder, Colorado
and
University of Colorado
Boulder, Colorado 80309
1BC822
11. CONTRACT/GRANT NO.
R803931
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Robert S. Kerr Environmental Research Lab - Ada, OK
Office of Research and Development
U. S. Environmental Protection Agency
Ada, Oklahoma 74820
Final - 8/75 - 9/78
14. SPONSORING AGENCY CODE
EPA/600/15
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Treatment performance of three rapid infiltration basins receiving secondary
treated municipal wastewater is evaluated at the City of Boulder, Colorado, Sewage
Treatment Plant. The prime objectives of the project are evaluation of the pre-
treatment provided, the hydraulic load to the soil, duration of wetting and drying
cycles, and climatic influences on system performance. Two of the basins have
unscarified beds while one has a bed surface constructed in a ridge and furrow
arrangement. Year-round operation allowed evaluation of applied effluent and
infiltrated water collected from underdrains during extreme climatic conditions.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Land use
Ground water
Purification
Quality control
Sewage treatment
Nutrient removal
Land Application
High rate infiltration
Tertiary treatment
Ridge and furrow basins
Underdrain system
Sewage effluents
68D
48B, E, G
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
114
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
104
* u.s.GovEmMcmntiinim;OFFICE: 1979 -657-060/5424
------- |