<>EPA
United States
Environmental Protection
Agenc:
R.S. Kerr Environmental
Research Laboratory
Ada, Oklahoma 74820
August 1986
BASIC DESIGN
RATIONALE
FOR ARTIFICIAL
WETLANDS
-------
BASIC DESIGN RATIONALE
FOR
ARTIFICIAL WETLANDS
by
John Zirschky and D. Donald Deemer
ERM-Southeast, Inc.
Marietta, Georgia 30066
Contract No. 68-01-7108
Work Assignment Manager
Lowell Leach
R. S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
U. S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 74820
R. E. THOMAS, PROJECT OFFICER
OFFICE OF MUNICIPAL POLLUTION CONTROL
U. S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D. C. 20460
-------
TABLE OF CONTENTS
Page
LIST OF FIGURES iii
LIST OF TABLES iii
Section
1 INTRODUCTION 1
1.1 Background 1
1.2 Objectives 2
1.3 Scope of Work 2
1.4 Definition of Artificial Wetlands 3
1.5 Limitations of the Technology 4
1.6 Report Format 5
2 LITERATURE REVIEW 7
2.1 Sources of Information 7
2.2 Wetlands Hydrology 8
2.2.1 Hydrologic Budget 8
2.2.2 Detention Time 9
2.3 Fate of Wastewater Constituents 11
2.3.1 BOD Removal 11
2.3.2 Suspended Solids 17
2.3.3 Nitrogen 18
2.3.4 Phosphorus 20
2.4 Case Studies 21
2.4.1 General 21
2.4.2 Listowel, Ontario 22
2.4.3 Santee, CA 24
2.4.4 Arcata, CA 25
2.5 Summary 27
3 DESIGN PROCEDURES 29
3.1 Alternatives Available 29
3.2 Options for Preapplication Treatment 31
3.3 Practical Design Guidelines 36
3.4 Limiting Constituent Method 40
3.5 Detention Time 44
3.6 Plug-Flow First-Order Design Procedures 47
3.7 Non-Discharge Design 51
3.8 Summary of Design Methods 53
-------
Section
Table of Contents (cont'd)
RELATED PROCESSES 54
4.1 Introduction 54
4.2 The Root Zone Method (RZM) 54
4.2.1 Process 54
4.2.2 Design Procedures 55
4.2.3 Operating Guidance 58
4.3 Rock/Reed Filter (RRF) 59
4.3.1 Process Description 59
4.3.2 Design Information 60
RESEARCH CONSIDERATIONS 61
5.1 General 61
5.2 Economic Evaluation 61
5.3 Field Research Program 62
BIBLIOGRAPHY 63
11
-------
LIST OF FIGURES
Figure
2-1
3-1
3-2
Effect of BOD Loading on BOD Removal
Effluent BOD Concentration vs. Influent
BOD Loading Rate
Detention Time Required to Achieve Secondary
Quality Effluent in an Artificial Wetland
Page
16
43
45
Table
2-1
3-1
3-2
3-3
3-4
LIST OF TABLES
Average Influent and Effluent Values for
Arcata, California
Characteristics of a Food Processing Wastewater
used for Design Examples
Design Guide and Checklist for Aquatic Systems
Pretreatment Design Parameters for Use with
Wetland Treatment Systems
Loading Rates for Wetland Systems
Page
26
30
32
37
41
111
-------
SECTION 1
INTRODUCTION
1.1 Background
The U. S. EPA Robert S. Kerr Environmental Research
Laboratory (RSKERL) and the Beijing (Peoples Republic of
China) Municipal Research Institute of Environmental
Protection (BMRIEP), as part of a cooperative wastewater land
treatment program, are currently evaluating the feasibility
of using artificial wetlands for the treatment of high
strength wastewaters. Of particular interest is the use of
wetlands for treating brewery and food processing
wastewaters. The effluent from the wetlands could then be
used for crop irrigation. Re-use of the effluent as
manufacturing process water or as a drinking water source are
also being considered.
Although some research on the efficiency of artificial
wetlands treatment has been conducted in the United States,
specific design guidance has not been published.
Furthermore, operational data from artificial wetland systems
are sparse. By attempting to compile a basic guideline for
the design of artificial wetlands for the treatment of high
strength organic wastewaters, areas in need of further
investigative study can be identified. The design rationale
developed can be used to determine if further co-operative
research of this treatment mode is warranted, based upon any
deficiencies identified in the design procedures.
-------
1.2 Objectives
The objectives of this work assignment were first to review
the available literature on the design of artificial
wetlands, and second, to prepare a document outlining basic
design rationale for utilizing artificial wetlands to treat
high strength wastewaters from brewery or food processing
plants.. The criteria developed would serve two purposes.
First, they will assist in determining if further cooperative
research on wetlands systems should be conducted. Second,
these criteria will also assist EPA's Office of Water Program
Operations in implementing innovative and alternative aquatic
treatment systems through the Construction Grants program.
1.3 Scope of Work
In order to develop the design rationale, a computer abstract
search was conducted to identify literature related to the
design and operation of artificial wetlands treatment
systems. In addition, contact was made with several
individuals known to have conducted research related to
artificial wetlands. Design alternatives were then generated
based upon the information collected and a draft report
summarizing these alternatives was submitted to EPA. This
report incorporates the comments received on the draft
report. To facilitate further research in this area, a
bibliography of wetlands references was also prepared and is
included in this report.
-------
1.4 Definition of Artificial Wetlands
The computer abstract search was designed to include
information on wastewater treatment in natural wetlands as
well as in artificial wetlands. The purpose of including
natural wetlands in the literature review was to ensure that
as much data as possible relevant to the design of artificial
wetlands was obtained. Upon reviewing the literature, it
became apparent that a number of different types of systems,
such as marshes, swamps, peat bogs, and water hyacinth
systems were referred to as wetland treatment systems. The
size, shape, and water depth of these systems are highly
variable.
In this report, the term artificial wetlands is limited to
systems with the following characteristics:
1) Areas in which rainfall and the influent wastewater
are the only water sources. Water losses occur
only by treated wastewater outflow and
evapotranspiration.
2) Systems which are based upon plug flow hydraulics
with a single outflow. Systems which discharge
primarily to ground water, such as cypress domes,
are not included.
3) Systems where the water surface is at or above the
ground surface long enough each year to maintain
saturated soil conditions and the related
vegetation (Reed, pers. comm.).
-------
1.5 Limitations of the Technology
Although limited data are available on the use of wetlands
for treating primary effluent, artificial wetlands have been
used in a number of locations for polishing secondary
effluent. Applied BOD concentrations are, thus, typically on
the order of 30 mg/L or less. Food processing wastewaters,
on the other hand, may contain BOD concentrations ranging
from 300 to greater than 30,000 mg/L. In addition, some food
processing wastewaters may contain significantly higher
concentrations of solids, nitrogen, and phosphorus than
typical domestic sewage while others are substantially lower
in nutrients than domestic sewage. Design procedures for
utilizing wetlands, normally a polishing process, as the
primary treatment process for high strength wastewater were
not found in the literature.
Individuals active in the research of artificial wetlands as
well as the environmental representatives of several food
processing industries and trade associations were also
contacted. None of these individuals were optimistic about
the use of artificial wetlands alone for treating high
strength wastes. The concensus of opinion was that
pretreatment would be required in order to use artificial
wetlands.
It was the original intent of this report to exclude systems
in which only subsurface flow occurs (e.g. rock/reed filters,
the Root Zone method) . In a strict sense, such systems are
not artificial wetlands because the water surface is never at
or above the ground surface and the removal rates in these
systems are substantially different from those found in
-------
artificial wetlands. The objective of this report is to
present potential design procedures for artificial wetlands.
As requested by EPA, however, design information on the root
zone method and the rock/reed filter are also presented
herein (Section 4).
It was also initially intended to present as many design
procedures as possible to assist the RSKERL and the BMRIEP in
developing a joint research program. After discussions with
EPA, however, several of these methods have been eliminated.
The amount of information or assumptions required to use such
methods would seldom, if ever, be available to a design
engineer. It is important to note that some of the design
procedures contained herein may also be inappropriate, since
many of these procedures have never been verified. One goal
of any subsequent research on artificial wetlands should be
to attempt to validate these methods and determine which
methods yield accurate results.
1.6 Report Format
This investigation has concluded that artificial wetlands
alone will not provide adequate treatment of high strength
wastewaters. Thus, in order to be used, artificial wetlands
must be included as one step in an overall treatment scheme
which would include pretreatment and perhaps post-treatment.
In Section 2, a brief review of the literature related to
artificial wetlands is presented. Some information on
natural wetlands is also included in this section since the
treatment processes in artificial and natural wetlands are
similar; although, the rate at which these processes occur
-------
may differ significantly between artificial and natural
wetlands. In Section 3, potential design methods are
presented along with a brief discussion of pretreatment
options. Section 4 discusses other treatment processes which
are related to wetlands and which could be used for treatment
of food processing wastes (after pretreatment). Research
recommendations are presented in Section 5. Finally, a
bibliography of wetlands and related information is presented
in Section 6.
-------
SECTION 2
LITERATURE REVIEW
2.1 Sources of Information
A computer abstract search was conducted to identify
references related to the design and performance of natural
and artificial wetlands treatment systems. Abstract sources
searched were NTIS, Pollution Abstracts, and Agricola. In
addition, research facilities and individuals known to have
sponsored or conducted research on artificial wetlands were
also contacted. Facilities contacted included RSKERL, the U.
S. Army Cold Regions Research and Engineering Laboratory
(CRREL), and several universities. The results of this
literature review are summarized briefly in this section.
Literature on the design of artificial wetlands for treating
high-strength wastewaters was not found. Research on
artificial wetlands has focused primarily on the treatment of
secondary municipal effluents. Very limited data are
available on the treatment of raw and primary wastewaters
using artificial wetlands. The systems treating raw
wastewater are not conventional artificial wetlands. For
example, in the Root Zone method (Lawson, 1985), wastewater
flows beneath the soil surface through the root zone of the
vegetation. Surface flow does not occur, and vegetation
plays a minimal role in purification of the wastewater.
Lawson (1985) himself states that the Root Zone method is
distinctly different from artificial wetlands. Discussion of
the Root Zone method is thus presented separately from
conventional artificial wetlands. The following discussion
is based upon key literature pertaining to the treatment of
wastewater in artificial wetlands, as well as in natural
-------
wetlands. A general discussion of the fate of contaminants
is presented first. Brief case studies of several key
research investigations are then discussed.
2.2 Wetlands Hydrology
2.2.1 Hydrologic Budget
The performance of any artificial wetlands system is
dependent upon the system hydrology as well as other factors.
Precipitation, infiltration, evapotranspiration, hydraulic
loading rate, and water depth can all affect the removal of
organics, nutrients, and trace elements not only by altering
the detention time, but also by either concentrating or
diluting the wastewater. A hydrologic budget should be
prepared to design an artificial wetlands treatment system
properly. Changes in the detention time or water volume can
significantly affect the treatment performance.
For an artificial wetlands as defined in Section 1.4, the
water balance could be expressed as follows:
Qi - QQ + P - ET = dV/dt [Eq 2-1]
where:
Q. = influent wastewater flow, vol/time,
Q = effluent wastewater flow, vol/time,
P = precipitation, vol/time,
ET = evapotranspiration, vol/time,
V = volume of water, and
t = time.
-------
Artificial wetlands designed for wastewater treatment are
generally constructed above an impermeable surface (e.g., a
liner or natural soil barrier), and under such conditions,
ground water inflow and infiltration would not occur. Thus,
these components of flow are not included in Equation 2-1.
Historical climatic records can be used to estimate
precipitation and evapotranspiration. Empirical methods such
as the Thornthwaite equation can be used to estimate
evapotranspiration (Thornthwaite and Mather, 1957). Pan
evaporation measurements may be useful if the wetlands will
contain a significant percentage of open water areas. If
required, estimates of water losses due to infiltration can
be obtained by conducting infiltration tests such as outlined
in the EPA Process Design Manual for Land Treatment of
Municipal Wastewater (EPA, 1981). Then, if the system
operates at a relatively constant water depth (dV/dt=0), the
effluent flow rate can be estimated using Equation 2-1.
2.2.2 Detention Time
Treatment performance in artificial wetlands is a function of
the detention time, among other factors. Ground slope, water
depth, vegetation, areal extent, and geometric shape control
the flow velocity and, thus, the detention time through a
wetlands treatment system.
In both natural and artificial wetland systems, estimating
the detention time can be difficult for several reasons.
First, large dead spaces may exist in the wetlands due to
differences in topography, plant growth, solids
sedimentation, and the degree of flow channelization (i.e.
short-circuiting). Only a fraction of the surface area, in
both artificial and natural wetlands, may be available for
-------
wastewater flow. Wile et al. (1985) reported that in spite
of accurate leveling of an artificial wetlands,
short-circuiting still occurred.
The ratio of the surface area available for flow to the total
surface area is termed the void fraction. Using the void
fraction, residence time for wetlands can be defined as
follows (Hammer and Kadlec, 1983):
[Eg. 2-2]
where :
0 = residence time,
V = volume of water in the wetland,
Q = volumetric flow rate,
A = surface area of the wetland,
d = average water depth, and
0 = void fraction of surface area available for flow.
The void fraction, 0, is a difficult parameter to estimate.
Hammer and Kadlec (1983) cite values for the void fraction of
0.1 to 0.3 for shallow water sedge meadows to approximately
1.0 for deep, open water wetlands. Black et al. (1981)
reported that dense cattail growths could occupy between
one-third to one-half of the wetland volume. Dead spaces
within the cattails can also occur.
Since the void fraction can be difficult to estimate,
especially when data on the topography and water depth are
10
-------
lacking, a superficial residence time, 6s, is sometimes
used, which is defined by Equation 2-3 (Hammer and Kadlec,
1983) .
9s = [Eq. 2-3]
In estimating the actual or superficial residence time, the
flow rate, q, used should be an average of the influent and
effluent flow rates since water losses (or gains) can be
significant.
Treatment performance is a function of detention time in the
wetland. Wile et al. (1985) reported seven days to be
optimal for the treatment of advanced primary and secondary
wastewater. Shorter detention times do not provide adequate
time for pollutant degradation to occur; whereas, longer
detention times can lead to stagnant, anaerobic conditions.
Two climatic factors can significantly affect the detention
time at a constant hydraulic loading rate. In warm weather,
evapotranspiration can significantly increase the detention
time. On the other hand, ice formation in cold weather can
significantly decrease the detention time. Maintaining
liquid depths of less than 10 cm in warm weather and greater
than 30 cm in cold weather was recommended to minimize the
effect of climate on the detention time.
2.3 Fate of Wastewater Constituents
2.3.1 BOD Removal
The treatment processes which occur in an artificial wetlands
are similar to those which occur in other forms of land
11
-------
treatment. Removal of settleable organics occurs primarily
as a result of sedimentation and aerobic/anaerobic
decomposition in the sediments. Removal of colloidal and
soluble organics occurs primarily by aerobic microbial
oxidation. Few data are available for the removal of BOD and
COD in wetlands, and there are some limitations in the data
that are available. The primary limitation is that much of
the available data are from natural or artificial wetland
systems used to polish municipal advanced secondary quality
effluent, primarily for nutrient removal. BOD loadings are
thus considerably less than would be experienced in a wetland
treating food processing wastes. Some of these systems
report BOD removals of over 80 percent (Gersberg, et al.,
1984a; Fetter, et al., 1978), while other systems report
essentially no BOD removal (Stowell, et al., 1980).
The low influent BOD concentrations applied to many wetlands
systems discussed in the literature may be very close to
background levels; thus, little or no reduction in BOD
concentration would be expected. When significant water
losses due to evapotranspiration occur, however, the mass
removal of BOD may be significant.
Data on the removal of BOD from primary municipal wastewater
is available primarily from two systems: Santee, California,
and Listowel, Ontario. No data were found on the treatment
of raw municipal wastewater (note: methods such as the Root
Zone treatment scheme used in Europe are discussed
separately). Black et al. (1981) reported 87 percent BOD
removal for an influent BOD quality of 65.5 mg/L (effluent -
9.0 mg/L). The estimated average detention time in the
wetland system was 7.8 days. Additional and more detailed
information on the performance of this system will soon be
published (Herskowitz, pers. comm.).
12
-------
At Santee, California, a mixture of primary and secondary
municipal effluents was treated using an artificial wetlands.
The average influent BOD concentration was 49 mg/L while the
effluent BOD concentration was approximately 3.2 mg/L.
Wastewater was applied at the rate of 18 cm/day. When
primary wastewater alone was treated (influent BOD = 148
mg/L), the effluent BOD averaged approximately 33 mg/L.
Primary wastewater was applied at a hydraulic loading rate of
6 to 8.3 cm/day. Better year-round performance occurred at a
hydraulic loading rate of 6 cm/day, which corresponded to a
detention time of 1 day (Gersberg et al.f 1984a). Additional
data from the Santee and Listowel research projects are
discussed in Section 2.4.
BOD removal in wetlands has been described by a first-order
model as follows (Lakshman, 1980) :
ln(C/Co) = -kt [Eq. 2-4]
where
C = BOD concentration at time t,
C = initial BOD concentration,
o
k = decay constant, and
t = time.
The first-order kinetics of BOD removal in wetlands systems
are similar to those found with both overland flow and lagoon
systems, both of which are somewhat analogous to artificial
wetlands. The data from Listowel and Santee presented above
can be used to estimate the decay constant for Equation 2-4.
At Listowel, the decay constant would be equal to 0.25/day
13
-------
(year-round average) . At Santee the average decay rate was
1.5/day for primary wastewater. Lakshman (1980) reported
decay constants equal to 0.25/day (September) and 0.14/day
(October) for initial BOD concentrations ranging from 30 to
100 mg/L.
Sufficient information, such as temperature data, are not
currently available to permit an in-depth comparison of these
decay rates. One would expect a lower year-round decay rate
for Ontario (Listowel) than for California (Santee). An
investigation of the effects of the wetlands and climatic
characteristics on the decay rate may be a prerequisite to
developing rational design procedures. The forthcoming
report on the Listowel system may contain such information.
Gearheart, et al. (1983), conducted a pilot-scale test of
artificial wetlands for polishing secondary effluent and also
found BOD removal to be described by a first-order kinetic
model:
-In ^ = (fhko) Q
S = effluent concentraton, mg/L,
S. = influent concentration, mg/L,
h = thickness of slime, m,
W = width of section (width of cell) , m,
Q = volumetric flow rate, m /day,
Z = filter depth (length of cell), m,
k = maximum reaction rate, /day,
o
f = proportionality factor
14
-------
The term (fhk ) was found to be approximately equal to 4.95
m/d when oxygen was not limiting. Under oxygen limiting
conditions, BOD removal occurred at a significantly slower
rate. The decay rate was not reported, but it appeared from
the data presented to be approximately one-fourth of the
non-oxygen limited decay rate.
Equations 2-4 and 2-5 are very similar. If the slime
thickness in Equation 2-5 was known, one could rearrange
Equation 2-5 into the form of Equation 2-4 and then compare
the rate constants in each equation. The slime thickness,
however, is not defined by Gearheart et al. (1983); thus,
such a comparison is not possible.
The decay rates used in either Equation 2-4 or 2-5 may not be
indicative of the performance which might be expected for a
wetland system treating higher than advanced primary strength
wastewaters. Higher BOD influent concentrations could create
extensive anaerobic conditions, and BOD oxidation would be
expected to be less under anaerobic conditions than under
aerobic conditions. If anaerobic conditions do occur, odors
could be a significant problem. To reduce the potential for
odors, Stowell, et al, (1980) recommend BOD loading rates of
60 to 70 kg/ha/d or less; although, one example of a wetlands
system treating 100 kg/ha/day without odors was cited.
The relationship between BOD loading and removal rates has
been evaluated by Stowell et al. (1980) and is presented in
Figure 2-1. It is important to recognize that only five
systems were evaluated in compiling Figure 2-1 and that these
systems treated secondary municipal or better quality
15
-------
40
30
Q
O
CD
h- 20
UJ
D
_l
U_
u_
UJ
10
DATA FROM 17 DIFFERENT STUDIES
o
o o
o
o
o
50 100 150
INFLUENT BOD (mg/L)
200
Figure 2-1.
Effect of BOD Loading on BOD Removal
(Marsh and Peatland Systems; Stowell,
et al., 1980).
16
-------
effluent. The data suggest that the degree of removal is a
fairly linear function of the loading applied. Data beyond
an applied loading rate of 100 kg/ha/day are not presented in
Figure 2-1. The implications of this data base are that if a
desired effluent load is selected, the influent loading rate
could be determined. The relationship may continue to be
linear well beyond the range of the data, and when higher
effluent values can be tolerated, the influent loadings
could, perhaps, extend well above 100 kg/ha/day. For
example, Gersberg et al. (1984b) used BOD loading rates of
116 kg/ha/day without any reported problems. The effluent
flow rate at this loading rate is not presented, however, to
permit calculation of a mass removal rate.
2.3.2 Suspended Solids
The suspended solids present in a wastewater will be removed
by primary sedimentation in a wetland. If the influent
solids loading is high, significant accumulation of solids
could occur near the influent discharge points. An ugly gray
slime accumulation was reported to have occurred near the
influent to a wetlands system in Ontario, Canada, receiving a
high solids wastewater (Huggins, personal communication).
Degradation of the solids can also create a significant
oxygen demand. As a result, anoxic conditions could develop,
thereby generating odors and killing vegetation. Black et
al. (1981) noted such conditions occurring as a result of
solids accumulation at the Listowel system.
If the water in the wetlands is not shielded from sunlight by
the vegetation, algae could also be a problem in warm months.
Algal solids will not settle readily and could lead to high
suspended solids concentrations in the effluent. If the
effluent is to be used for crop irrigation, however, algae
may not be of concern. Furthermore, during daylight hours,
17
-------
the algae can generate significant quantities of oxygen. At
night, however, respiration can consume the oxygen, leading
to anaerobic conditions. Gearheart, et al. (1983) reported
that 20 to 40 percent open water was optimal for wetlands as
the resulting phytoplankton can help control pathogens
without creating significant algae problems.
2.3.3 Nitrogen
Microbial action (nitrification-denitrification) and plant
uptake are the primary nitrogen removal mechanisms in
wetlands. Depending upon the detention time and pH, ammonia
volatilization may also be significant. Stowell et al.
(1980) reported nitrification rates ranging from 0 to 45
kg/ha/day and denitrification rates ranging from 0 to 40
kg/ha/day. Vegetation, however, was reported to remove only
approximately 6 kg N/ha/day. Wile et al. (1985) reported
that vegetation removed only 8 percent of the applied
nitrogen at the Listowel system.
Nitrification and denitrification occur primarily at the
interface between the soil particles or vegetation and the
water. Thus, the depth of water, the area inundated, and the
presence of vegetation can affect the rates of nitrification
and denitrification. Weber and Tchobanoglous (1982) reported
nitrification to be a function of the application rate in
water hyacinth systems. As the application rate increases,
the transport of ammonia to the nitrifying organisms which
grow on the soil and plant surfaces also increases. A
similar phenomenon may occur in artificial wetlands. For
example, shallow water depths and even flow distribution in
wetlands encourage contact between the wastewater and the
microorganisms which grow on the soil particles and plant
surfaces (Kadlec and Tilton, 1979) . Shallow water depths
18
-------
also promote the transport of oxygen from the water surface
to the nitrifying organisms.
Zoltek, et al. (1979) conducted a laboratory study of the
effects of water depth, pH control, and vegetation on
nitrification rates in natural marsh soils. At ammonia
nitrogen concentrations of 40 to 50 mg/L, ammonia removal
rates of 4.9 (15 cm water depth), 5.7 (30 cm), and 7.7 (0 to
30 cm variable depth) kg N/ha/day were found with plants.
Lower rates were generally reported in the systems without
plants. Controlling the pH to the range of 7 to 8.5
increased ammonia removal to 14.3 kg N/ha/day in the systems
with plants. As the strength of the influent wastewater
increases, the nitrification rates might be expected to
decrease since oxygen may become too limiting for
nitrification to occur. More complex organics may be broken
down into ammonia and, without significant nitrification,
ammonia may accumulate.
Some ammonia could be removed via volatilization depending
upon the detention time and the pH. Ammonia volatilization
increases with increasing pH and detention time. The pH of
one artificial wetland was found to be neutral to slightly
acidic (Gearheart et al., 1983), and since artificial
wetlands typically have detention times less than 10 days,
ammonia volatilization would not be expected to be a
significant removal mechanism. Procedures for estimating the
amount of ammonia volatilized are presented by Reed (1984).
The presence of high ammonia concentrations is of concern
because ammonia is toxic to aquatic life. Fish are often
used for mosquito control in wetlands, and high ammonia
levels could cause fish kills in the areas where there is
sufficient dissolved oxygen to support aquatic life.
19
-------
Gersberg et al. (1984a,b) found that by increasing the
influent BOD concentration (by selective blending of primary
and secondary wastewater), nitrogen removal in wetlands could
also be increased. The addition of plant biomass also
stimulated nitrogen removal. Total nitrogen removal
efficiencies of 90% were obtained in wetlands where both
primary wastewater and plant biomass were used as a carbon
source. Although higher strength wastewaters should provide
sufficient carbon for denitrification (and thus nitrogen
removal) to occur, the resulting anaerobic conditions could
limit nitrification. In fact, limiting the extent of
anaerobic conditions over much of the wetlands may be
difficult in a wetlands treating higher than primary strength
wastewater. Gersberg et al. (1984a) found that nitrogen
removal decreased to only 25 percent when primary strength
wastewater was treated.
Only the Listowel data are sufficient for estimating the rate
of nitrogen removal. At a detention time of approximately
7.8 days, the total nitrogen concentration was reduced from
21.1 to 8.45 mg/L. If first-order kinetics are applicable,
then the first-order decay rate for nitrogen removal would be
approximately 0.12/day (year-round average).
2.3.4 Phosphorus
Removal of phosphorus in wetlands occurs primarily as a
result of soil adsorption and to a lesser extent by
precipitation and plant uptake. Phosphorus removal is
enhanced by: 1) organic soils in poor nutrient regimes, 2)
vegetation which is either limited in phosphorus supply or
capable of luxury consumption, and 3) the presence of iron
and aluminum (Heliotis, 1982). As the water depth in the
wetland decreases, the potential for contact with the soil
20
-------
surface and, thus, adsorption increases. With time, however,
the available adsorption sites will become saturated and
phosphorus removal will decrease. Hammer and Kadlec (1983)
presented a procedure for estimating the rate of phosphorus
saturation in natural wetlands. This procedure may also be
applicable to artificial wetlands since the chemical/physical
processes should be similar in both natural and artificial
wetlands.
Phosphorus removal is affected by temperature with most
phosphorus removal occurring in summer, and some phosphorus
release occurring in winter (Fetter et al., 1978), due
perhaps to release from decaying vegetation. Black et al.
(1981) also reported that phosphorus removal decreased during
colder weather, as did Gearheart et al. (1983). Wile et al.
(1985) reported that vegetation removed only 10 percent of
the applied phosphorus. Reed (pers. comm.) recommends that
to be conservative, one should assume that no phosphorus
removal will occur.
2.4 Case Studies
2.4.1 General
Several artificial wetland systems are in operation in the
United States. Most of these systems, however, are designed
either to treat secondary or higher quality wastewater or to
be non-discharging systems. Examples of such systems are:
Incline Village, Nevada (Williams et al., 1985); Mountain
View, California (Demgen, 1979; Hyde, et al, 1984);
Vermontville, Michigan (Williams and Sutherland, 1979; Bevis,
1981; and Hyde et al., 1984); Neshaminy Falls, Pennsylvania
(Hyde et al., 1984); Las Galinas Valley Sanitary District,
California (Demgen, 1984); Cannon Beach, Oregon (Demgen,
21
-------
1984); and Arcata, California (Gearheart et al., 1983).
Only two systems treating primary strength municipal
wastewater were found in the literature: Listowel, Ontario
and Santee, California. Both of these systems are briefly
discussed herein. The Root Zone treatment method used in
Europe is not discussed in this section because such systems
are not artificial wetlands (Lawson, 1985) . The Root Zone
method is discussed separately in Section 4.
2.4.2 Listowel, Ontario (Black et al., 1981;
Wile et al., 1985)
In the late 1970's, the town of Listowel's wastewater
treatment system was becoming hydraulically overloaded. The
treatment system consisted of an aerated lagoon with a
detention time of 3.5 days followed by two facultative
lagoons in series, each with a detention time of 35 days.
Upgrading the system to treat the increasing load and meeting
water quality standards by conventional means appeared to be
uneconomical for this small community. A pilot study of
using artificial wetlands for polishing the lagoon effluent
was thus undertaken in hopes of finding a more economical
means of upgrading the treatment system.
The pilot system was constructed to treat the effluent from
both the aerobic and facultative lagoons. The performance
of the systems treating the aerated lagoon effluent are
applicable to this report. Two basic wetland configurations
were utilized, one a series of five long narrow basins with a
length to width ratio of 75:1 and the other, one basin with a
length to width ratio of 4.5:1. The series of long narrow
basins consistently performed better than the one long basin,
and the researchers recommended that future systems utilize a
22
-------
similar geometry. Thus, this basin configuration is
discussed herein.
In the literature related to this system, the authors
describe the effluent from the aerobic lagoon as "raw,
aerated sewage" (Wile et al. 1985), thus perhaps implying
that the wastewater only undergoes minimal treatment prior to
the wetlands. Data from the first year of operation
indicates that the effluent BOD and TSS from the aerated cell
are approximately 65.5 and 119 mg/L, respectively (Black et
al., 1981). The wastewater thus undergoes significant
treatment prior to the wetlands.
Wile et al. (1985) present performance data for this system
for approximately the first 1-1/2 years of operation (August
1980 to April 1982). During this time period the average
effluent (from the wetland) BOD concentration was
approximately 10.7 mg/L for an average removal efficiency of
approximately 86 percent. The effluent TSS concentrations
averaged 8.2 mg/L for a 94 percent removal efficiency. Total
nitrogen removal averaged 66 percent for an effluent
concentration of 8.3 mg/L. The influent concentrations for
these parameters were not reported. From the removal
efficiencies cited, however, the approximate influent BOD,
TSS, and total nitrogen concentrations were approximately 76,
130, and 25 mg/L, respectively. The average hydraulic
loading rate was 2.5 cm/day yielding organic and nitrogen
loading rates of 18.4 kg BOD/ha/day and 6,04 kg N/ha/day.
Total phosphorus removal was cited as 98 percent. Alum,
however, was used to precipitate the phosphorus. A report
summarizing the four years of experimental data at this
system should be available in September, 1986.
23
-------
2.4.3 Santee, California (Gersberg et al.,
1983,1984a,b)
Several studies on the performance of artificial wetlands
have been conducted at the Santee, California Water
Reclamation Facility. Most of the investigations conducted
at this facility focused on nitrogen removal in artificial
wetlands. Secondary effluent was primarily treated during
these investigations; although, some research was conducted
using primary and a blend of primary and secondary effluent.
Two beds, each 71 m long by 11.6 m wide (L:W=6.1:1) were
used. Wastewater was applied at hydraulic loading rates of 6
to 20 cm/day.
The objective of mixing primary and secondary effluents was
to provide a source of carbon to enhance nitrogen removal in
wetlands; thus, the majority of the investigations focused on
nitrogen removal. Gersberg et al. (1984b) reported that if
the goal of the wetlands treatment is to produce an effluent
quality to less than 10 mg/L of total nitrogen, then blend
ratios of 1:2 (pr imary: secondary) can be used at hydraulic
loading rates of 18 to 19 cm/day. At this blend, the
influent BOD quality tested was approximately 63.5 mg/L,
which corresponds to a mass loading rate of approximately 116
kg BOD/ha/day. At this loading rate, approximately 89
percent of the influent BOD was removed yielding an effluent
concentration of approximately 7 mg/L. Total nitrogen was
reduced from 24 to 5 mg/L at a loading rate of 44.4 kg
N/ha/day (79% removal).
An artificial wetlands treating a 1:2 blend with an influent
BOD concentration of 49 mg/L at a loading rate of 18 cm/day
was reported by Gersberg et al. (1984a) to produce an
effluent quality of 3.2 mg/L (93% removal). The organic
24
-------
loading rate was approximately 88.2 kg BOD/ha/day. At the
1:3.8 primary to secondary blend at a loading rate of 18-19
cm/day, influent BOD concentrations were reduced from
approximately 40.9 mg/L to 4.1 mg/L (90% removal). The BOD
loading rate was approximately 77 kg BOD/ha/day. Total
nitrogen reduction data were not reported for the 1:3.8
blend. Straight primary wastewater with a mean BOD of 148
mg/L was treated at a hydraulic loading rate of 6 to 8.3
cm/day (106 kg/ha/day). The mean effluent concentration was
33 mg/L corresponding to 78 percent BOD removal. Treatment
performance decreased, however, when the hydraulic loading
rate was increased from 6 to 8.3 cm/day. Total nitrogen
removal decreased to only 25 percent. Neither detention time
data nor hydrologic data were presented to permit calculation
of the decay rates or mass removals.
2.4.4 Arcata, CA (Gearheart, et al., 1983)
The city of Arcata, California, conducted a three-year pilot
project of wetlands treatment of secondary effluent. The
major goal was to prove that the effluent from a marsh
receiving secondary effluent could enhance the water quality
in Humboldt Bay. A discussion of this system is included
primarily to provide supporting information for Equation 2-5.
The marsh project was constructed adjacent to the city
oxidation ponds. The marsh was divided into twelve cells,
each 6.1 m (20 ft.) wide by 61 m (200 ft.) long by 1.2 m (4
ft.) deep. Non-chlorinated oxidation pond effluent was
pumped to a stilling tower which delivered the flow to three
division boxes feeding four cells each. Two operating
depths, 30.5 cm (12 in.) and 61 cm (24 in.), and four
theoretical hydraulic detention times ranging from 28 to 224
hours were used in the project experiments. Weirs controlled
25
-------
the flows and the depths in the cells. The cells were
planted in alkali bulrush (Sc i rpus robustus) or hardstem
bulrush (Scirpus acutus) in the spring of 1981. Numerous
wetland vegetation appeared voluntarily, primarily cattails.
Table 2-1 summarizes the influent and effluent quality data
for the Arcata system as a function of hydraulic loading
rate. The data presented in Table 2-1 represent the average
performance of the test systems over a two year period.
Organic loadings of 166 kg/ha/day (148 Ib/ac/day) or less
proved to be the most effective within the range of loadings
utilized. BOD removal rates of 50% or greater were achieved
as long as dissolved oxygen levels remained above 1.0 mg/L.
The best removal rates occurred during the fall and winter.
TABLE 2-1
AVERAGE INFLUENT AND EFFLUENT VALUES FOR
ARCATA, CALIFORNIA
Effluent Concentration
Hydraulic Loading (MGD/acre)
Concentration
BOD, mg/L 26
NH -N, mg/L. 12.8
pH, su 7.4
P04-P, mg/L . 6.28
0.06 0.12 0.24
10.7 13.3 12.8
9.6 11.6 9.8
6.6 6.6 6.5
6.05 6.17 7.0
Based upon data from September 1980 to September 1982
26
-------
Ammonia-nitrogen levels were effectively reduced at loadings
of 0.1 to 2.5 kg/ha/day (0.09 to 2.24 Ibs/acre/day). Ammonia
nitrogen made up approximately 80 percent of the total
Kjeldahl nitrogen in the marsh effluent. Over 70 percent of
the organic nitrogen and 30 percent of the total Kjeldahl
nitrogen was removed. The higher hydraulically loaded cells
produced lower concentrations of nitrates. During periods of
high dissolved oxygen, nitrate levels increased from the
normal 1.0 mg/L to 3-4 mg/L. Some denitrification did occur,
but was not a major factor.
Phosphorus was only minimally removed, with the average
removal being five percent. The major portion of phosphorus
removal occurred during the three to four months growing
season. Phosphorus levels did not noticeably increase during
non-growing seasons.
Non-filterable residues (NFR) were reliably removed over the
entire project period, with an overall average of 85 percent
removal. The hydraulic loading did not effect the removal
efficiency. Removal efficiency did increase as the amount of
vegetation increased from the beginning of the project.
Based upon extrapolation of pilot project data, organic
loadings of 200 to 250 kg/ha/day (178 to 223 Ib/acre/day) are
the maximum loadings at which a marsh treatment system can
produce an effluent low in NFR.
2.5 Summary
The literature indicates that advanced secondary to tertiary
quality effluent can be produced for advanced primary quality
municipal wastewater (BOD of 40 to 80 mg/L). Secondary
27
-------
effluent quality can be produced from primary wastewater (BOD
of approximately 150 mg/L). Total nitrogen removal decreases
as the influent BOD concentration increases, probably as a
result of decreased oxygen levels limiting nitrification.
Artificial wetlands are not effective for phosphorus removal.
Alum can be used, however, to increase phosphorus removal.
Virtually all of the literature sources discuss treatment of
municipal wastewaters. Even where primary effluents were
treated, the strength of those effluents was significantly
less than the strength of most food processing and brewery
wastewaters. The evidence strongly suggests that a
substantial level of preapplication treatment may be required
for those higher strength wastes.
28
-------
SECTION 3
DESIGN PROCEDURES
3.1 Alternatives Available
The objective of this section is to present alternative
procedures for designing an artificial wetlands treatment
system for high strength wastewaters. The data presented in
Section 2 indicate that preappl ication treatment: would be
required in order to use artificial wetlands because
artificial wetlands alone are not suitable for high strength
wastewaters. Very limited information on the design of
wetlands treatment systems is available, and most of the
available information relates to the use of wetlands for
treating municipal wastewaters. The design procedures
presented herein should, therefore, be considered as
preliminary in nature. All of the design procedures
presented herein can be found in the literature. Even though
these methods have not been validated, these methods are
presented herein to suggest options for future research work.
One of the goals of any co-operative investigation between
the U.S. and China may be to eliminate or validate some of
these design methods.
Five general design procedures are presented. The first
procedure is based upon the practical experience gained at
the systems described in Section 2. The second approach
utilizes constituent loading rates, primarily BOD, for
design. The third approach is based upon the detention time
required for treatment. A first-order plug-flow kinetic
model forms the basis of the fourth approach. Finally, the
fifth method involves performing a water balance to determine
the surface area required to create a non-discharge system.
29
-------
For the design procedures developed herein, an attempt was
made to conduct a limited validation of these methods by
comparing the results obtained with practical guidelines
developed from experience. The pretreated effluent quality
presented in Table 3-1 was selected so as to be within the
range of operating data from the Listowel and Santee systems.
Thus, one would expect a valid general design procedure to
yield results similar to those found at Listowel and Santee.
Detention time and organic loading rate are used to compare
the design procedures with the existing data.
The advantages and potential limitations associated with each
system are presented with the discussion of each method. For
each design example, the wastewater characteristics presented
in Table 3-1 have been used as a typical food processing
wastewater.
TABLE 3-1
CHARACTERISTICS OF A FOOD PROCESSING
WASTEWATER USED FOR DESIGN EXAMPLES
Parameter
(mg/L)
BOD
TSS
Total N
Influent Flow
a
Range
(mg/L)
<50 to >30,000
300-500
10-100
. b
Design
Value
100
50
25
1 MGD
Effluent Flow
(3785 M /day)
0.85 MGD
(3200 M3/day)
a) of raw wastewater; b) after pretreatment
30
-------
Table 3-2 presents a guide and checklist for the design of
aquatic treatment systems, including artificial wetlands.
Although not all of the information required in this
guideline is currently available, the design process outlined
herein summarizes the factors which should be considered in
designing artificial wetlands.
Other approaches could be used for the design of an
artificial wetland including an oxygen balance, the
first-order model presented by Gearheart et al. (1983,
Equation 2-5) , or adapting procedures from other forms of
aquatic or land treatment systems. The data are not
available to support these methods, or the methods require
parameters which would be difficult to measure or estimate.
For example, Hammer and Kadlec (1983) developed a mass
transport approach for use in the design of wetlands, and the
reference should be consulted for a complete development of
this approach. In order to use this approach, however, five
constants for the wetlands must be known including mass
transfer coefficients. These constants could vary
significantly between wetlands, and it does not appear that
there is any means to estimate these constants at this time.
Thus, a further discussion of this method is not presented.
3.2 Options for Preapplication Treatment
The data presented in Section 2 and conversations with
several other wetlands experts all indicate that artificial
wetlands are not suitable for treatment of high strength
wastewaters. Data presented by Gersberg et al. (1984a)
indicates that even primary municipal wastewater (BOD-148
mg/L) can significantly stress a wetland. If wetlands are to
be included in a process scheme for high strength
31
-------
TABLE 3-2
DESIGN GUIDE AND CHECKLIST FOR AQUATIC SYSTEMS
(Stowell, et al., 1980)
"Preliminary Investigation
1. Survey sites suitable for an aquatic system.
2. Review and summarize climatic factors: temperature,
precipitation, and wind.
3. Survey local flora: growth season, vegetative structure
through winter, and potential to affect aquatic
environments.
4. Review discharge requirements: are seasonal
requirements possible, if justified?
5. Review literature with emphasis on type of project being
envisioned, contaminant removal mechanisms, and
treatment performance and reliability.
6. Formulate preliminary aquatic system design, including
possible use of conventional processes.
7. Analyze cost-effectiveness of alternative wastewater
treatment processes: go or no-go on aquatic system
design.
32
-------
TABLE 3-2 (cont'd)
Aquatic System Design
1. Itemize expected flows, influent concentrations, and
discharge requirements.
2. Identify contaminant removal mechanism(s) and
environment(s). (Complete this analysis for each
contaminant to be removed.)
a. Removal mechanism(s).
1) Identify removal mechanism(s) operative on
contaminant of concern.
2) Summarize kinetics of removal mechanism(s) and
factors affecting the kinetics: contaminant
concentration, temperature, pH, etc.
b. Aquatic-environment(s) required for removal
mechanism(s) to be operative.
1) List requirements for aquatic environment(s).
2) List factors affecting these requirements:
climatic factors, organic loading rate, plant
species, chemical and equipment input, etc.
3. Conceptualize aquatic process units (APU's) necessary.
(Complete this analysis for each contaminant to be removed.)
the aquatic
a. Plant selection.
1) What components necessary to
environment will it provide?
2) Seasonality in the local environment?
3) Local diseases and predators: seasonal or
year round?
33
-------
TABLE 3-2 (cont'd)
4) Nuisance potential?
b. Chemical and equipment input.
1) What components necessary to the aquatic
environment will it provide?
2) Continuous, seasonal, or emergency use.
c. Management to maintain aquatic environment (e.g.,
harvesting).
d. Rate of contaminant removal and APU surface area
requirements on a seasonal basis.
4. Design aquatic system.
a. Integrate conceptual APU's into designed APU's.
1) Maximize concomitant treatment, process
performance and system reliabilty.
2) Minimize redundant aquatic environments and
costs.
b. Complete quantities mass balance (in terms of
removal mechanism kinetics) for each contaminant
for each APU for the seasons of the year.
c. Are the APU's mutually compatible?
d. Are the discharge requirements met?
e. What measures can be taken to increase the
reliability of the system?
34
-------
TABLE 3-2 (cont'd)
f. Management.
1) To maintain the system.
2) Courses of action in case of system upset.
5. Make final check on cost-effectiveness of aquatic
system.
6. Design and operate pilot-scale facility . (This step is
necessary until contaminant removal mechanism rates
and factors affecting these rates are quantified.
If the pilot does not function as anticipated,
aquatic system design steps 2 and 6 will have to be
modified as necessary. Failure of the pilot
facility could result in the aquatic system
alternative being dropped from consideration for
the particular project.)
7. Design and construct full-scale aquatic system."
35
-------
wastewaters, then some form of preapplication treatment will
be required.
Stowell et al. (1985) provide guidance and design parameters
for pretreatment options which can be considered for use in
conjuction with wetland treatment systems. These include
Imhoff tanks, aerated lagoons, primary sedimentation, and
primary filtration. These options, however, are based
primarily for the treatment of domestic sewage, and the
applicability of these methods to food processing wastewaters
will depend upon the characteristics of the specific
wastewater. The design parameters for these pretreatment
methods are presented in Table 3-3. Wile et al. (1985)
recommend using a complete mix, aerated lagoon with an 8 to
10 day detention time for pretreatment of municipal
wastewater. For a high strength wastewater, additional
detention time may be required. Other pretreatment options
include slow rate (with underdrains) and overland flow land
treatment.
3.3 Practical Design Guidelines
The early design procedures for other forms of land treatment
systems were primarily based upon practical experience gained
at existing systems. For municipal wastewater, hydraulic
loading rates were often used for design. The hydraulic
loading rate used successfully at one system could easily be
adapted for use at other systems due to the relatively low
range of variability in the concentrations of raw municipal
wastewater. The range of wastewater characteristics within
the food processing industry, however, is much wider. The
variability of wastewater characteristics, however, will
largely be eliminated by the requirement for pretreatment
prior to the wetlands. If primary to advanced primary
36
-------
TABLE 3-3
PRETREATMENT DESIGN PARAMETERS FOR USE WITH
WETLAND TREATMENT SYSTEMS (Stowell et al., 1985)
Pretreatment
Design Parameters
Units
Value
Imhoff tank
Overflow rate
Detention time
Sludge storage capacity
Lagoon
Aerated
Detention time
Depth
Facultative
Detention time
Depth
BOD loading
Anaerobic
Detention time
Depth
BOD loading
Primary sedimendation
Overflow rate
Detention time
Depth
Primary filtration
Surface loading rate
gal/ftVd
h
mo
d
ft
d
ft
Ib/acre/d
d
ft
Ib/acre/d
gal/ftVd
h
ft
2
gal/ft /min
600
3
6
5
12
10
5-6
50
30
14-16
300
1000
2
12
37
-------
quality wastewaters are discharged to the wetlands, the
following guidelines should be applicable for designing a
system to produce secondary quality efflue;nt. These
guidelines are based upon the research conducted by Black et
al. (1981), Wile et al. (1985), Gersberg et al. (1983,
1984a,b), Stowell et al. (1980), Gearheart et al. (1983), and
Reed (pers. comm.).
Detention time - Seven to ten days.
Hydraulic loading rate - 2 to 20 cm/day, use the low end
of the range for primary
wastewater.
Water depth - <10 cm in summer, >30 cm in winter.
Physical configuration - a length:width ratio >15:1,
channel width of approximately
3 m for ease of harvesting.
Vegetation - Cattails (Typha spp.)
Organic load ing rate - 15 to 120 kg BOD/ha/day, use
lower values of this range for
primary wastewater, and values in
the middle of this range for
nitrogen removal from advanced
primary wastewater.
Mosquito control - required at all but remote locations.
Influent structures - multiple flow diffusers
recommended.
Effluent structures - collection manifold recommended.
38
-------
Design Procedures;
1. Calculate the wetland volume required to achieve
the desired detention time;
Seven days was selected as the design detention
time.
V = (3785 m3/day)(7 days) = 26495 m3
2. Select an operating depth and calculate the
required area.
A depth of 20 cm as a year-round average was
selected.
a = 26495/0.2 = 132000m2 = 0.029 m2 = 13.2 ha
3. Calculate the hydraulic loading rate.
Loading rate = 3785m3/day/132000m2 = O.,029m/day =
2.9 cm/day
2.9 cm/day is within the acceptable range.
4. Calculate the organic loading rate.
mass of BOD applied = 378.5 kg/day
378.5/13.2 = 28.7 kg/ha/day, which is within the
acceptable range.
39
-------
3.4 Limiting Constituent Method
Other forms of land treatment systems (e.g., slow rate, rapid
infiltration) are often designed by determining the
acceptable mass loading of organics, nutrients, and trace
elements. Then, by knowing the concentration of each of the
parameters of concern in the wastewater, the acceptable
hydraulic loading rate associated with each parameter can be
determined. The hydraulic loading rate is selected either
based upon the hydraulic loading rate of the limiting
constituent or the infiltration capacity of the soils,
whichever is less. The required land area is then calculated
from the selected hydraulic loading rate.
A similar approach can be used for the design of wetlands.
Table 3-4 sumarizes the reported removal efficiencies and
organic loading rates reported in the literature. These data
were developed from systems where the wastewater ranged from
primary to secondary (40 mg/L) quality. Gersberg et al.
(1984a) found that nitrogen removal decreased as the BOD
concentration increased even if the BOD loading rate was
roughly the same. Thus, judgement is required when using the
data presented in Table 3-4. Another potential problem with
the data presented in Table 3-4 is that the effect of
operating parameters such as water depth and residence time
is unknown.
40
-------
TABLE 3-4
LOADING RATES FOR WETLAND SYSTEMS
Parameter
BOD
Nitrogen
Loading Rates
(kg/ha/day)
18.4
77
88.2
106
116
6.04
44.4
2
Removal Efficiency
89%
90
93
78
86
86
79
(63.5 to 7 mg/L)
(40.9 to 41)
(49 to 3.2)
(148 to 33)
(76 to 10.7)
(25 to 8.3)
(24 to 5)
Developed from data presented in Section 2.
2
Numbers in parentheses represent reductions from influent to
effluent.
The following design example is presented using the
wastewater characteristics presented in Table 3-1.
Design Procedures:
Wastewater Characteristics (from Table 3-1).
Q = 106 gallons/day = 3.785 x 106 L
BOD = 100 mg/L
Total N = 25 mg/L
1. Select design wastewater loading rates
The following values were selected based upon judgement
and the data presented in Table 3-4.
BOD: 75 kg/ha/day
N: 6 kg/ha/day
41
-------
Low BOD and nitrogen loading rates were selected because it
was felt that the high BOD concentration could create
anaerobic conditions which would inhibit nitrification.
2. Calculate mass loadings rates
BOD: (3.785 x 106L/d)(100 mg/L)(10~6 kg/mg) =
378.5 kg/day
Total N: (3.785 x 106L/d) (25 mg/L) (10~6 kg/mg) =
94.6 kg/day
3. Calculate Required Surface Area
BOD: (378.5 kg/day)/(75 kg/ha/day) = 5 ha
Total N: ( 94.6 kg/day)/(6 kg/ha/day) = 15.7 ha
Thus, approximately 16 ha of wetlands would be required.
4 . Estimate treatment performance
The effluent quality can then be estimated using Figure 2-1,
or Figure 3-1. It is important to note, however, that
neither of these figures were developed for primary strength
wastewaters. The effluent BOD concentration can be estimated
directly from Figure 3-1, which yields an effluent
concentration of approximately 30 mg/L. This concentration
roughly agrees with the reduction of an influent BOD of 148
to 33 mg/L at a loading rate of 106 kg/ha/day reported by
Gersberg et al. (1984a) . Figure 2-1 indicates that 95% BOD
removal will be obtained at a loading rate which corresponds
to an effluent concentration of approximately 5 mg/L. The
latter procedure (Figure 2-1) probably yields an unrealistic
estimate of the effluent quality.
42
-------
45 •
40--
35
CD 30
£25
Q
O
m
+. 20
c
-------
5. Select a water depth and calculate detention time
Using a water depth of 20 cm yields a water volume in the
wetlands of 31,400 m . The theoretical detention time would
thus be approximately 8.3 days which is within the optimum
range of 7 to 10 days reported in Section 3.3. Since the
same data base was used to develop this method and the method
described in Section 3.3, however, the results should roughly
agree.
3.5 Detention Time
Stowell, et al. (1985) presented design curves for sizing
wetland systems to produce secondary quality wastewater by
treating raw, primary, and advanced municipal wastewaters,
but not high strength wastewater (Figure 3-2). This curve is
also limited for use at a loading rate of 100 Ib BOD/ha/day
and a water depth of one foot (30 cm). No data are presented
to support this design curve. This method is presented for
information purposes only and is not recommended for design
use until supportive data are available.
Design Procedures;
1. Estimate the critical wetlands temperature
For the detention times generally encountered in artificial
wetlands, the water temperature should be roughly equal to
the design air temperature, to a minimum of 32 F. The
average air temperature during the critical time of year
44
-------
0
10
o
0
J_
13
•+-•
03
0)
Q.
E
15
20 -
25 —
30
BOD Loading Rate = 100 Ib/acre.d
Depth = 1.0 ft
»= 1.10
Decreasing Temperature
Untreated Wastewater
BODs=220 mg/L
•Primary Effluent
BODs= 145 mg/L
-Advanced Primary
BODs = 80 mg/L
'Increasing Pretreatment
i I i i i i I i i i i
i i
0
10
15
20
25
30
Figure 3-2. Detention Time Required to Achieve Secondary
Quality Effluent (Stowell, et al., 1985).
45
-------
should be used. A second possibility would be to use the
temperature of nearby streams or wetlands. A temperature of
10 C was selected for this example.
2. Determine the detention time
Figure 3-2 presents a relationship between the wastewater
temperature and the required detention time for three levels
of pretreatment (none, primary, and advanced primary).
By interpolating between the curves, the detention times
required for treatment of other influent BOD concentrations
can be found.
Using the temperature determined in step 1, use Figure 3-2 to
determine the required detention time. At 10°C, the required
detention time to produce secondary quality effluent is
approximately 8 days. This detention time agrees with the
recommendations presented in Section 3.3.
3. Determine the land area required
A = Qt/d
where,
2
A = surface area, m
Q = influent flow rate, m /d
t = detention time, d
d = water depth, m
A = (3.785 x 106 L/d)(8 d)(10~3m3/L)/O.3m =
1.01 x 105m2
A = 10 ha.
46
-------
This surface area yields a BOD load rate of 38 kg/ha/day,
which is in the acceptable range.
Several problems exist with this method. First, one cannot
specify a desired effluent quality and design the system
accordingly. Second, one is limited to a loading rate of 112
kg/ha/day (100 Ibs/acre/day) , as stated on the design
curves. Finally, too many parameters may be specified on the
design curves. For example, the design curve specifies a BOD
loading rate of 100 Ib/acre/day (11 kg/ha/day). The result
of this design example indicates a loading rate of 34
Ib/acre/day (38 kg/ha/day). This design procedure thus does
not appear to be consistent. The primary advantage of this
method is that temperature can be included as a design
parameter, and temperature fluctuations may have a
significant effect on performance.
3.6 Plug-Flow First-Order Design Procedure
To some extent, especially for shallow water depths, an
artificial wetland system is analogous to an overland flow
land treatment system. First-order plug-flow models have
been successfully used to describe pollutant removal in both
overland flow systems (Abernathy, et al., 1985; Smith, et
al., 1983; Martel, et al., 1982) and artificial wetlands
(Lakshman, 1980; Gearheart, et al., 1983). In overland flow
systems, both slope and detention time have been used as the
independent variables. Slope length has yielded better
results to date due perhaps to the inability to predict
detention time on an overland flow slope accurately
(Abernathy et al. 1985). Sufficient data are not available
to determine the kinetic removal rates as a function of
distance in a wetland. Furthermore, detention time on an
overland flow slope is controlled primarily by application
47
-------
rate and slope. In a wetlands, however, detention time is
significantly affected by water depth. Detention time may
thus be a more suitable basis for using a kinetic model for
wetlands design than the length of the wetland.
The design model to use would be of the following form:
ln(C/Co) = -kt [Eq. 2-4]
where C, Co, k and t are defined previously.
The potential advantage of using this approach is that the
detention time required to achieve a specified effluent
concentration can be calculated if C, Co, and k are known or
specified.
t = -(l/k)ln(C/Co) [Eq. 3-2]
Values for the decay constant, k, for wetlands are scarce to
non-existent. Lakshman (1980) reported BOD decay rates of
0.14 to 0.25 per day for wetlands treating influent BOD
concentrations of 30 to 100 mg/L. The data from Listowel and
Santee indicates a range of BOD decay rates of from 0.25
(cold climate) to 1.51 per day (warm climate). A nitrogen
decay rate of 0.12/day was found to describe the performance
of the Listowel system. Lakshman (1980) reported that total
nitrogen decay rates were approximately one-third of the BOD
decay rates for an AWT wetlands.
BOD removal will occur both by solids settling and microbial
oxidation. If a significant portion of the influent BOD is
associated with solids, then two decay rates (one each for
solid and soluble BOD) may be needed. Such data, however,
are not currently available.
48
-------
Design Example
Suppose that the influent wastewater has the characteristics
presented in Table 3-1 and that an effluent BOD and nitrogen
quality of 30 mg/L and 10 mg/L, respectively, is desired.
The following procedure could be used to size the wetland
system.
1. Select a BOD and nitrogen decay rate
It has been assumed that the BOD decay rate will be 0.25/day
and that the nitrogen decay rate will be 0.12/day, based upon
data presented in Section 2. The lowest values found were
used to be conservative. If the wetlands water temperature
will be significantly different than 20°C, the BOD decay rate
can be adjusted using the following relationship:
T-20
KT = (K2Qo) 9 [Eq> 3_2]
where
K = reaction rate constant at wetlands temperature ( C)
in days
K,, o = reaction rate constant @ 20 C.
9 = temperature coefficient, 1.085
o
T = water temperature in C.
Stowell et al. (1985) used a temperature coefficient of 1.1,
which is close to the value of 1.085 used above.
49
-------
2. Estimate the required detention time
Using Equation 3-2,
for BOD removal:
t = (1/k)(lnC/Co)=-(l/0.25)(In 30/100) = 4.8 days
for N removal:
t = (1/0.13)ln(10/25) = 8 days
Thus, nitrogen removal will control the land area
requirements.
3. Calculate the volume of water required to achieve
the desired detention time
From Table 3-1, the influent flow rate is 3,785 m /day.
The required water volume is thus:
V = (8 days) (3785) = 30,280 m3
4. Select a design water depth
A water depth of 20 cm was selected to promote contact
between the soil and wastewater.
5. Determine the required surface area
A = V/d = 30,280 m3/0.2m = 15.1 ha
50
-------
6. Compute organic loading rate
The influent mass of BOD to the system, from design example
1, was 378.5 kg/day. The organic loading rate is thus:
378.5 kg BOD/day/15.lha = 25 kg/ha/day
The results of this method also appear reasonable when
compared to the practical guidelines.
3.7 Non-Discharge Design
All of the previous design approaches assume treatment is the
desired objective of the wetlands system. However,
wastewater disposal or volume reduction may be the desired
goal. If so, a hydrologic balance can be prepared to
calculate the required surface area.
It has been assumed that water losses due to infiltration
will be insignificant in accordance with the definition of a
wetland (see Section 1.4). Without infiltration, the water
balance becomes simpler:
change in water volume = Wastewater inflow +
precipitation - evaporation
Obviously, for the system to be non-discharging without
infiltration occurring, evaporation must exceed the sum of
precipitaton and wastewater inflow.
51
-------
1. Perform a water balance
Using either evapotranspiration and precipitation data or
estimating evaporation from empirical formulas, determine the
amount of water lost per day per unit area. EPA (1981)
presents methods for calculating a water balance. For this
example, assume that 25 cm (10 inches) has been selected as
the design evapotranspiration rate.
2. Calculate the land area required
The volume of wastewater generated each year is:
V = 3.785 x 102m3/d(365 days) = 1.38 x 105m3/year
The amount of water evaporated per square meter is, from
3 2
above, 0.25 m /m /year.
The land area required is thus:
A = (1.38 x 106m3/yr)/(0.25 m3/m2/yr) = 55.3 ha
3. Calculate the organic loading rate
From Section 3.3, the BOD mass loading rate is:
378.5/55.3 = 6.8 kg BOD/ha/day
This value is well below the loading rates presented in Table
3-2. Thus, it may be more economical to add additional
wetland capacity for polishing (if a discharge of higher
quality effluent is feasible) or to use another form of land
treatment.
52
-------
3.8 Summary of Design Methods
All of the design methods yielded reasonable results. The
practical design guidelines (Section 3.3) is the recommended
method until the other methods can be validated. These
practical guidelines were developed from the operating
experience at several systems which have performed well.
The other more rational design procedures need to be
subjected to a more rigorous validation procedure. One goal
of a future research project may be to conduct such a
validation. Additional research considerations are outlined
in Section 5.
53
-------
SECTION 4
RELATED PROCESSES
4.1 Introduction
Two processes somewhat similar to artificial wetlands are
briefly discussed in this section. These processes are the
Root Zone Method (RZM) and the Rock/Reed Filter. In the
purest sense, the RZM is not an artificial wetlands since the
wastewater is maintained below the ground surface. The data
from operational systems in Europe indicate that this method
could be effective for the treatment of food processing
wastewaters. A discussion of this method is presented
herein. For more detailed design information, the reader
should refer to Boon (1985) and Lawson (1985) .
A process similar to the RZM is the Rock/Reed Filter (RRF;
Wolverton, 1981, 1986). Wolverton describes the RRF as an
anaerobic filter and not as a wetland. This process,
however, may also have some applicability to a high strength
food processing wastewater treatment scheme. A discussion of
the RRF is also presented herein.
4.2 The Root Zone Method (RZM)
4.2.1 Process Description
The RZM is a subsurface treatment system where wastewater
flows through the root zone of a reed (Phragmites) bed. The
reed roots serve a number of functions. First, the roots
increase the permeability of the soil. Second, the reeds
tranport oxygen from the atmosphere to the roots and into the
54
-------
root zone. The RZM is thus an aerobic and anaerobic
treatment process. Nitrification occurs in the aerobic zone
around the root, while denitrification occurs in the
remaining anaerobic areas. Phosphorus removal occurs by
soil adsorption and chemical precipitation; although, as with
wetlands, the phosphorus removal cabability may eventually
become exhausted. Higher influent BOD concentrations (on
the order of 600 mg/L) can be treated by this method due to
the more intimate contact between the wastewater and the
microorganisms. Three years are required, however, for the
roots to create sufficient soil permeability for treatment to
occur. Design considerations for this method are presented
below.
4.2.2 Design Procedures
The following design steps for the RZM are presented by Boon
(1985) :
"1. Selection of a Site
The site should be an open ground not shadowed by
trees or shrubs. The slope of the ground should
not be excessive; between 2 and 6 percent would be
acceptable. If possible, the soil covering the
site (up to a depth of at least 1.5 m) should have
a hydraulic conductivity of about 10 to 10 m/s.
A 'light1 clay or a 'heavy1 top soil would be best;
such soil could be brought to the site if
necessary.
55
-------
2. Preparation of a Site
Existing soil has to be removed to a depth of about
1.5 m below the level of incoming sewage so that
sewage will gravitate through the reed bed. A
sealant of clay (with a hydraulic conductivity of
less than 10~° m/s), bentonite, synthetic fabric,
or asphalt is required to retain water in the
reed-bed and prevent contamination of ground water.
After sealing the site, the soil in which the reeds
will be planted has to be replaced. A 'light'
clay, containing some quartz and chalk would be
ideal. This soil should be carefully placed to a
depth of about 0.6 m, avoiding the use of heavy
vehicles which might compress the soil and reduce
its hydraulic conductivity.
At the inlet and outlet ends of each bed, trenches
should be left into which stones, granite, or slag
can be placed. The nominal diameter of the medium
(similar to that used for biological filters) would
be 60 to 100 mm. Each trench should be sited
across the width of a bed and should be about 0.35
m wide and 0.65 m deep. Final effluent would be
collected into the outlet trench and flow out via a
20 cm diameter pipe, the discharge level of which
should be capable of variation. During normal
operation, the level of water would be maintained
at about the top of the bed or just below (1-2 cm).
56
-------
3. Design Details for a RZM Works
The width of a bed would depend upon the slope from
inlet to outlet. The cross-sectional area
can be calculated from the equation.
(A )
Ac - 0
Ac " °-
where
- 4~1]
kf dH/ds
is the average flow-rate of sewage (m /s)
32
is the hydraulic conductivity (m /m s)
of the fully developed rhizosphere (i.e.
k
- 3 3 2
is equal to 10 m /m s) , and
dH/ds is the slope of the bed (m/m).
This equation applies to a RZM bed treating
effluent, raw or settled sewage, or a readily
biodegradable wastewater with a bed depth about 0.6
to 0.65 m. For wastewaters which are difficult to
biodegrade, the depth may have to be reduced to 0.3
m.
The area of a bed (A, ) can be calculated from
n
the equation.
^ = 5.2 Q0 (In.C -In.C )
n dot
(m) [Eq. 4-2]
57
-------
where
Q, is the average flow-rate of sewage {m /d)
C is the average BOD of the wastewater
(rag/L) , and
is the
(mg/L) .
C is the average BOD of the effluent
The value of 5.2 relates to the removal of BOD from
sewage in the bed which is to be 0.6 m deep and operated
at a minimum temperature of 8°C. For more difficult to
biodegrade wastewaters and lower temperatures the value
will be greater, up to 14 or 15."
Design Example
Using the wastewater characteristics presented in Table 3-1/
one can calculate the land area required for an RZM treatment
system using Equation 4-2, as follows:
A, = 5.2 (378.5) (In 100 - In 30) = 2370 m2
n
A, = 0.23 ha.
This method thus obviously requires less land area than an
artificial wetland for the same wastewater.
4.2.3 Operating Guidance
Three years are required for an RZM system to become fully
operational. During the first year, secondary or higher
quality effluent should be applied. During the second year,
raw (screened and degritted) wastewater should gradually be
added with the effluent. At first, a proportion of 30
58
-------
percent raw municipal' wastewater to 70 percent effluent is
recommended. Gradually, the percentage of raw wastewater
should be increased until by the end of the third year, raw
wastewater may be treated.
Harvesting should not be conducted as cutting the reeds will
reduce the transport of oxygen into the root zone.
Additional design, construction, and operating guidance can
be found in Boon (1985) and Lawson (1985) .
4.3 Rock/Reed Filter (RRF)
4.3.1 Process Description
The RRF filter is essentially a horizontal trickling filter
into which reeds, rushes, or similar plants have been
planted. Wastewater enters the filter after pretreatment to
remove solids and decrease the organic loading and flows
through a rock bed. The bed media consists of a one-foot
layer of 4-inch rock overlain by 4 inches of topsoil. The
length to width ratio of the bed is typically 50:1.
A microbial film grows on the rock surfaces similar to a
trickling filter. As the wastewater passes through the bed,
this microbial film purifies the wastewater. Influent BOD
concentrations of approximately 120 mg/L have been treated
with this method. With a 24-hour detention time, an influent
BOD concentration of 116 mg/L was reduced to 12 mg/L (90%
reduction). Wolverton (1982, 1986) should be consulted for
additional information on the performance of this system.
59
-------
4.3.2 Design Information
Very limited design information is available on the RRF.
Important design parameters identified by Wolverton (1986)
are: rock sizes, retention time, and vegetation type. From
the literature, it appears that a four-inch rock size and a
24-hour detention time are optimal. The vegetation type will
depend upon the climate. A hydraulic loading rate of
approximately 9 inches/day (23 cm/day) for primary wastewater
is also suggested. At this loading rate, approximately 2 ha
would be required to treat the wastewater characterized in
Table 3-1.
60
-------
SECTION 5
-RESEARCH CONSIDERATIONS
5.1 General
The results of this investigation indicate that there are
currently no widely accepted rational design procedures for
artificial wetlands. Furthermore, examples of the use of
wetlands for further treatment of high strength wastewaters
were not found in the literature. Therefore, many of the
design procedures presented in Section 3 may not: be valid for
the design of artifical wetlands for treating high strength
wastewaters.
The literature data that was collected appear to indicate
that wetlands treatment technology is not suitable for high
strength wastewaters. Testing of this technology, at least
on a pilot-scale, however, should perhaps be considered to
further evaluate this technology. A possible research
program is outlined below.
5.2 Economic Evaluation
Before investigation in a field research program, an economic
evaluation of using wetlands for treating food processing
wastewaters should be considered. This treatment technology
can perform adequately providing a low BOD mass loading rate
(kg-BOD/ha/year) is used. The lower the loading rate,
however, the greater the land area required, and hence, the
greater the cost. The cost of pretreatment must also be
considered. The purpose of the economic evaluation would be
to determine, based upon currently available information, the
range of loading rates and perhaps climatic conditions under
61
-------
which wetlands treatment might be more cost-effective than
other forms of conventional treatment or land treatment,
including the costs of pretreatment in the analysis. Any
field research program could then focus on these loading
rates and other conditions. If factors other than economics
are important, such as wildlife habitat, then the economic
evaluation may not be the determining criteria.
5.3 Field Research Program
The next phase of a research program would be to conduct
pilot-scale testing of wetlands under the conditions
identified as being potentially cost-effective in the
research program to permit an accurate determination of:
o a hydrologic budget,
o an oxygen balance
o hydraulic detention time, and
o treatment performance
- through the wetlands,
- influent vs. effluent quality.
The pilot system(s) should have the capability to permit:
o variation of the hydraulic loading rate,
o variation of the organic loading rate,
o recycle of the effluent,
o collection of soil water and ground water samples
beneath the wetlands, and
o variations in the influent distribution system.
All of the above information and process capabilities would
be useful in evaluating the design procedures presented here
and, if necessary for development of new procedures.
62
-------
SECTION 6
BIBLIOGRAPHY
Abernathy, A. R., J. Zirschky, and M. B. Borup, 1985.
"Overland Flow Wastewater Treatment at Easley, S.C."
Journal Water Pollution Control Federation.
57(4) :291-299.
Bender, M. E. and D. L. Correll. 1974. "The Use of Wetlands
as Nutrient Removal Systems." Chesapeake Research
Consortium Pub. No. 29. University of Maryland, College
Park, MD.
Bevis, F. B. 1981. Reuse of Municipal Wastewater by
Volunteer Fresh-Water Wetlands (Vermontville, MI) .
PreparedforNationalScienceFoundation,Washington,
D.C., Williams and Works, Grand Rapids, MI.
Black, S. A., I. Wile, and G. Miller. 1981. "Sewage Effluent
Treatment in an Artificial Marshland." 1981 Water
Pollution Control Federation Annual Conference, Detroit,
Michigan.
Boon, A. G. 1985. Report of a Visit by Members and Staff of
WRc to Germany to Investigate the Root Zone Method for
Treatment of Wastewaters, WRc Processes, Herts, United
Kingdom,63p.
Boyt, F. L., S. Bayley, and J. Zoltek, Jr. 1977. "Removal of
Nutrients from Treated Municipal Wastewater by Wetland
Vegetation." Journal Water Pollution Control
Federation, 49:789-799.
Chan, E., T. A. Bursztynsky, N. Hantzsche, and Y. J. Litwin.
1981. The Use of Wetlands for Water Pollution Control.
EPA-6 00/2-8 2-086, U~. S~. EPA Municipal Engineering
Research Laboratory, Cincinnati, OH.
Conn, W. M. and A. C. Langworthy. 1984. "Practical Operation
of a Small Scale Aquaculture." Proceedings Water Reuse
Symposium III, San Diego, CA., American Water Works
Association, Denver, CO, pp.703-712.
Cornwell, D. A., J. Zoltek, C. D. Patinely, T. D. Furman, and
J. I. Kim. 1977. "Nutrient Removal by Waterhyacinths."
Journal of Water Pollution Control Federation, 49:57-65.
63
-------
DeBusk, T. A. and J. H. Ryther. 1984. "Nutrient Removal from
Domestic Wastewater by Waterhyacinths: The Importance
of Plant . Growth, Detritus Production and
Denitrification." Proceedings Water Reuse Symposium
111, San Diego, CA, American Water Works Association,
Denver, CO, pp.713-722.
Demgen, F. C. 1979. "Wetlands Creation for Habitat and
i a."
Lnar
EPA
en, r. v_. iy/y. wetxanas creation IOL naDitat ana
Treatment at Mt. View Sanitary District, California."
Seminar
Treatment at Mt. View Sanitary District, Califc
Aquaculture Systems for Wastewater Treatment: i
"Proceedings and Engineering Assessment,
430/9-80-006, U.S. EPA, Washington, DC, pp.61-73.
Demgen, F. C. 1984. "An Overview of Four New Wastewater
Wetlands Projects." Proceedings Water Reuse Symposium
III, San Diego, CA, American Water Works Association,
Denver, CO, pp.579-595.
Dempsey, H. H. and J. E. Shelton. 1983. "Advanced Biological
Treatment of Municipal Wastewater Through Aquaculture."
EPA-660/S2-83-007, U. S. EPA, Cincinnati, OH.
Denny. 1972. "Sites of Nutrient Absorption in Aquatic
Macrophytes." Journal of Ecology. 60:819.
Dinges, W. R. 1978. "Upgrading Stabilizaion Pond Effluent by
Water Hyacinth Culture," Journal of Water Pollution
Control Federation, 50:833.
Dinges, W. R. 1979. "Development of Hyacinth Wastewater
Treatment Systems in Texas." Aquaculture Systems for
Wastewater Treatment; Seminar Proceedings and
Engineering Assessment, EPA-430/9-80-006, pp.193-231.
Duffer, W. R. and J E. Moyer. 1978. Municipal Wastewater
Aquaculture. EPA-600/2-78/110, U. S. EPA, Cincinnati,
OH.
Engler, R.M., D.A. Antic, and W.H. Patrick, Jr. 1976.
"Effect of Dissolved Oxygen on Redox Potential and
Nitrate Removal in Flooded Swamp and Marsh Soils."
Journal of Environmental Quality, 5^:230-235.
Ewel, K. C. and H. T. Odum. 1978. "Cypress Swamps for
Nutrient Removal." Advances in Water and Wastewater
Treatment, Biological Nutrient Removal. M D. Wanielista
and W. W. Eckenfelder (eds.), pp.181-198.
Fetter, C. W. , 1978. "Use of a Natural Marsh for Wastewater
Polishing." Journal Water Pollution Control Federation,
50:290-307.
64
-------
Gaigher, I. G., D. F.
"Preliminary Studies
Brewery -Effluent in
Bacterial/Algal/Fish/Macrophyte
Agricultural Wastes, 12:207-224.
Gearheart, R. A., S,
and S. Sunberg
Marsh Pilot Project
and Management
Arcata, CA
Toerien, and J. U. Grobbelaar. 1985.
on the Treatment of Sorghum Beer
in an Integrated
Culture System."
Design
Works,
. Wilbur/ J. Williams, D. Hull, B. Finney,
j. 1983. "Final Report: City of Arcata
Project Effluent Quality Results—System
agement." City of Arcata, Dept. of Public
Gearheart, R. A., B. Finney, S. Wilbur, J. Williams, and D.
Hull. 1984. "The Use of Wetland Treatment Processes in
Water Reuse." Proceedings Water Reuse Symposium III,
San Diego, CA, American Water Works Association, Denver,
CO, pp.617-638.
Gee
and Jenson
Treatment
Engineers. 1980.
Design Manual
Water Hyacinth Wastewater
Laboratories, NSTL
Engineers, West Palm
Gersberg, R. M., B. V.
"Nitrogen Removal
Research, 17:1009.
Station,
Beach, FL.
Prepared for
MS. Gee and
Elkins, and C.
in Artificial
R. Goldman.
Wetlands."
NASA
Jenson
1982.
Water
Gersberg, R. M., B. V. Elkins, and C. R. Goldman. 1984a.
Wastewater Treatment of Artificial Wetlands." Water
Science and Technology, 17:443-450.
Gersberg, R. M. , B. V. Elkins, and C.
"Use of Artificial Wetlands to
" Journal Water
Wastewater.
Federation,
56:152-156.
R. Goldman. 1984b.
Remove Nitrogen from
Pollution Control
Gersberg, R. M., B. V. Elkins, S. R. Lyons, and C. R.
Goldman. 1984c. "The Removal of Heavy Metals by
Artificial Wetlands." Proceedings Water Reuse Symposium
III, San Diego, CA, American Water Works Association,
Denver, CO, pp.639-648.
Gersberg, R. M., B. V. Elkins, S. R. Lynch,
Goldman. 1985. "Role of Higher Aquatic
Wastewater Treatment by Artificial Wetlands."
manuscript.
and C. R.
Plants in
Unpubl.
Godfrey, P. J., E. R. Kaynor, S. Pelczarsk, J. Benforado.
1985. Ecological Considerations in Wetlands Treatment
of Municipal Wastewaters"!Van Nostrand Reinhold, New
York.
65
-------
Hall, D. H. and J. E. Shelton. 1983. Advanced Biological
Treatment of Municipal Wastewater Through Aquaculture.
EPA-600/2-83-007, U. S. EPA, Cincinnati, OH.
Hammer, D. E. and R. H. Kadlec. 1983. Design Principles for
Wetland Treatment Systems. EPA-600/2-83-026, U. S. EPA,
Cincinnati, OH.
Hantzche, N. N. 1985. "Wetland Systems for Wastewater
Treatment: Engineering Applications." Proceedings of
Symposium on Ecological Considerations in Wetlands"
Treatment of Municipal Wastewater. P. J. Godfrey et al.
(eds.) Van Nostrand Reinhold, New York, pp.7-25
Hardisky, M. A., R. M. Smart, and V. Klemas. 1983. "Growth
Response and Spectral Characteristics of a Short
Spartina Alterniflora Salt Marsh Irrigated with
Freshwater and Sewage Effluent." Remote Sensing of
Environment, 13:57.
Hauser, J. R. 1984. "Use of Water Hyacinth Aquatic
Treatment Systems for Ammonia Control and Effluent
Polishing." Journal of the Water Pollution Control
Federation, 5j5(3) :219.
Heliotis, F. D. 1982. Wetland Systems for Wastewater
Treatment; Operating Mechanisms and Implications for
Design. IES Report #117, Institute for Environmental
Studies, University of Wisconsin, Madison, WI.
Herskowitz, J. Water Resources Branch, Ontario Ministry of
the Environment, personal communication.
Huggins, A. Environmental Scientist, Environmental Resources
Managment, Inc., West Chester, PA, personal
communication.
Hyde, H. C. , R. S. Ross, and F. Demgen. 1984. Technology
Assessment of Wetlands for Municipal Wastewater
Treatment. EPA-600/2-84-154, U. S. EPA, Cincinnati,
urn
Jewell, W. J., J. J. Madros, W. W. Clarkson, H. Pompe, and
R. M. Kabrick. 1983. "Wastewater Treatment with Plants
in Nutrient Films." EPA-600/S2-83-067, U. S. EPA.
Cincinnati, OH.
Kaczynski, V. W. 1985. "Considerations for Wetland Treatment
of Spent Geothermal Fluids." Proceedings of Symposium
on Ecological Considerations of Wetlands of Muncipal
Wastewater, P. J. Godfrey, et al. (eds.), Van Nostrand
Reinhold, New York, PP.48-68
66
-------
Kadlec, R. H., et al. 1981. "The Hydrology of Overland Flow
in Wetlands." Chemical Engineering Communications
2:331.
Kadlec, R. H., C. J. Richardson, and J. A. Kadlec. 1975.
"The Effects of Sewage Effluent on Wetland Ecosystems."
Semi-Annual Report No. 4., NTIS PB249192, Washington,
D.C. University of Michigan, Ann Arbor.
Kadlec, R. H., D. L. Tilton, B. R. Schwegler, and Benedict R.
1979. "Three-year Summary of Pilot Scale Operations of
Houghton Lake." Report to the National Science
Foundation, NTIS PB295965, Washington, D.C.
Kadlec, R. A. and D. L Tilton. 1979. "The Use of Freshwater
Wetlands as a Tertiary Wastewater Treatment
Alternative." CRC Critical Reviews in Environmental
Control, 2(2):185.
Kadlec, R. H., 1979. "Wetlands for Tertiary Treatment."
Wetland Functions and Values: The State of Our
Understanding. American Water Resources Association,
Minneapolis, MN, pp.490-504.
Kadlec, R. H., 1985. "Wetlands Utilization for Management of
Community Wastewater." 1984 Operations Summary, Wetland
Ecosystem Research Group International Report,
University of Michigan, Ann Arbor.
Kadlec, R. H. and D. E. Hammer. 1984. "Wastewater Renovation
in Wetlands: Six Years at Houghton Lake." Proceedings
Water Reuse Symposium III, San Diego, CA, American Water
Works Association, Denver, CO, pp.596-616.
Khettry, R. K. 1984. "The Use of Artificial Marshes for
Wastewater Treatment in Ontario, Canada." Ontario
Ministry of Environment, Toronto, Ontario, Canada.
Lakshman, G. 1979. "An Ecosystem approach to the Treatment
of Waste Waters." Journal of Environmental Quality, 8
(3): 353-361.
Lakshman, G. 1980. A Demonstration Project at Humboldt to
Provide Tertiary Treatment to the Municipal Effluent
Using Aquatic Plants. 1979 Progress Report.
Sasketchewan Research Council. Saskatoon, Sasketchewan,
Canada.
Lawson, G. T. 1985. Cultivating Reeds (Phragmites australis)
for Root Zone Treatment of Sewage. Report totheWater
Research Centre of the Institute of Terrestrial Ecology,
Cumbria, U.K.
67
-------
Martel, C. J., T. F. Jenkins, C. J. Diener, and P. L. Butler,
1982. Development of a Rational Design Procedure for
Overland Flow-Systems, CRREL Report 82-2, U. S. Army
CRREL, Hanover, New Hampshire, 29p.
Mudroch, A., and J. A. Capobianco. 1979. "Effects of
Treated Effluent on a Natural Marsh." Journal of the
Water Pollution Control Federation. 51;2243.
Nichols, D. S. 1983. "Capacity of Natural Wetlands to Remove
Nutrients from Wastewater." Journal Water Pollution
Control Federation, 55:495.
Otta, J. W., T. G. Searle, and S. V. Geddes. 1984. "Land
Treatment Enhances Habitat of the Endangered Mississippi
Sandhill Crane." Proceedings of Water Reuse Symposium
III, San Diego, CA~American Water Works Association,
Denver, CO, pp.649-659.
Phillip J. Clark Engineers & Consultants. 1983. Marsh
Pond/Overland Flow Pilot Plant Project Report. Goshen,
NY.
Phillip J. Clark Engineers & Consultants. 1981. Project
Status Report for the Moodna Basin Marsh/Pond/Overland
Flow Pilot Treatment Plant. Goshen, NY.
Reddy, K. R. and D. L. Sutton. 1984. "Waterhyacinths for
Water Quality Improvement and Biomass Production."
Journal of Environmental Quality, 13;1.
Reed, S., Bastian, R., and Jewell, W. 1980. "Engineering
Assessment of Aquaculture Systems for Wastewater
Treatment: An Overview." Aquaculture Systems for
Wastewater Treatment; An 'Engineering Assessment7
EPA-430/9-80-007, 1-12.
Reed, S. Environmental Scientist, U. S. Army CRREL, Hanover,
New Hampshire, personal communication.
Reed, S., C. Bastian, S. Black, and R. Khettry. 1984.
"Wetlands for Wastewater Treatment in Cold Climates."
Proceedings Water Reuse Symposium III, San Diego, CA.,
American Water Works Association, Denver, CO.
Reed, S. C., 1984. Nitrogen Removal in Wastewater Ponds,
CRREL Report 84-13. U. S. Army CRREL, Hanover, N.H.
26p.
68
-------
Richardson, D. L. and G. T. Daigger. 1983. "Aquaculture:
The Hercules Experience." Journal of Environmental
Engineering Division, Proceeding American Society of
Civil Engineers, 110, 940.
Schwegler, B. R. and R. H. Kadlec. 1978. "Wetlands and
Wastewater." Pamphlet, Wetlands Ecosystem Research
Group, University of Michigan, Ann Arbor.
Shaw, B. H. 1984. "Municipal Wastewater Treatment Using an
Ombrotrophic Bog." Unpublished Paper. University of
Wisconsin, Stevens Point, WI.
Sloey, W. E., et al. 1978. "Management of Freshwater
Wetlands for Nutrient Assimilation." Freshwater
Wetlands. Ecological Processes and Management Potential.
IT. E. Good, et al. (eds.) Academic Press, New York, NY.
Small, M. M. 1978. "Wetlands Wastewater Treatment Systems."
State of Knowledge in Land Treatment of Wastewater,
International Symposium, U. S. Army CRREL, Hanover, NH,
pp.141-147.
Small, M. M. 1983. "Marsh-Pond Systems." Proceedings of the
Low-Cost Wastewater Treatment Workshop, Dept. of
AgriculturalEngineering,ClemsonUniversity, Clemson,
S.C.
Small, M. 1976. "Marsh/Pond Sewage Treatment Plants."
Freshwater Wetlands and Sewage Effluent Disposal.
Edited by D.
Richardson.
pp. 197-213.
, L. T
The
ilton, R. H. Kadlec, and C.
University
of
Ann
Arbor ,
. J
MI
Smith, R. G. et al., 1983. "Performance of Overland Flow
Wastewater Treatment Systems - Summary Report." Dept.
of Civil Engineering, University of California, Davis,
CA.
Spangler, F. L. , C. W. Fetter, Jr., and W. E. Sloey. 1977.
"Phosphorus Accumulation--Discharged Cycles in Marshes."
Water Resources Bulletin, 13(6);1191-1201.
Spangler, F. L. , W. E. Sloey, and C. W. Fetter, Jr. 1976.
"Experimental Use of Emergent Vegetation for the
Biological Treatment of Municipal Wastewater in
Wisconsin." Biological Control of Water Pollution, J.
Tourbier and R. W. Pierson, Jr. (eds.) University of
Pennsylvania Press, Philadelphia, PA., pp.161-171.
69
-------
Stewart, E. A., D. L. Haselow, and N. M. Wyse. 1984. "A
Practical Model for Water Hyacinth Based Wastewater
Management—Design and Operation." Proceedings Water
Reuse Symposium III, San Diego, CA., American Water
Works Association, pp.679-702.
Stowell, R. E., R. Ludwig, J. Colt, and G. Tchobanoglous.
1980. Toward the Rational Design of Aquatic Treatment
Systems. Dept. of Civil Engineering, University of
California, Davis, CA, 59p.
Stowell, R. E., et al. 1981. "The Use of Aquatic Systems for
Wastewater Treatment: An Assessment." Publication No.
65, California State Water Resources Control Board
Sacremento, CA.
Stowell, R., R. Ludwig, J. Colt, and G. Tchobanoglous. 1981.
"Concepts in Aquatic Treatment System Design." Journal
of Environmental Engineering Div., Proc. ASCE. 107:919.
Stowell, R. S. Weber, G. Tchobaoglous, B. A. Wilson, and K.
R. Townzen. 1985. "Mosquito Considerations in the
Design of Wetland Systems for the Treatment of
Wastewater." Ecological Considerations in Wetlands
Treatment of Municipal Wastewater, Godfrey, P. J., et
al. , (eds.) , Van Nostrand Reinhold, New York, 1985,
pp.38-47.
Sutherland, J. C. 1979. "The Vermontville, Michigan,
Wastewater-Grown Volunteer Seepage Wetlands: Water
Quality Engineering Implications (abstract)." Presented
at the Conference on Freshwaer Wetlands and Sanitary
Wastewater Disposal, Higgins Lake, MI, July 10-12.
Swett, D. 1979. "A Water Hyacinth Advanced Wastewater
Treatment System." Aquaculture Systems for Wastewater
Treatment; Seminar Proceedings and Engineering
Assessment, EPA-430/9-80-006,QT S~. EPA, Washington,
D.C., pp.233-255.
Tchobanoglous, G. and G. L. Gulp. 1980. "Aquaculture Systems
for Wastewater Treatment: An Engineering Assessment."
EPA-430/9-80-007, U. S. EPA, Washington, D.C.
Tuschall, J. R., P. L. Brezonik, K. C. Ewel. 1981. "Tertiary
Treatment of Wastewater Using Flow-Through Wetland
Systems." Proceedings of the National Conference of
American Society of Civil Engineers, 198TIAmerican
Socity of Civil Engineers, New York.
70
-------
U. S. EPA, Center for Environmental Research Information.
1981. Process Design Manual for Land Treatment of
Municipal Wast-ewater. EPA 625/1-81-013.
U. S. EPA, Region IV. et al., 1983. Environmental Impact
Statement. Phase 1 Report Freshwater Wetlands for
Wastewater Management. EPA-904/9-83-107, [~. S~. EPA,
Atlanta, GA.
U. S. EPA, Region IV. 1985. Freshwater Wetlands for
Wastewater Management Handbook~. EPA-404/9-85-135, U. S.
EPA, Atlanta, GA.
U. S. EPA, Office of Water Program Operations. 1979.
Aquaculture Systems for Wastewater Treatment; Seminar
Proceedings and Engineering Assessment.
EPA-430/9-80-006, U. S. EPA, Washington, D.C.
U. S. EPA, Office of Water Program Operations. 1980.
Aquaculture Systems for Wastewater Treatment; An
Engineering Assessment. EPA-430/9-80-007, U. S. EPA,
Washington, D.C.
U. S. EPA, Criteria and Standards Division. 1982. Benefits
and Implementation Potential of Wastewater Aquaculture.
UI S~. EPA, Washington, D. C.
U. S. EPA, Region V, Eastern Energy and Land Use Team. 1984.
The Ecological Impacts of Wastewater on Wetlands; An
Annotated Bibliography/EPA-9 05/3-8 4-00 2. 0". S.
Environmental Protection Agency, Region V, and U. S Fish
and Wildlife Services, Eastern Energy and Land Use Team,
Chicago, IL.
U. S. EPA, Region V, Wapora Inc. 1983. The Effects of
Wastewater Treatment Facilities on Wetlands in the
Midwest. EPA-905/3-83-002, U. S. EPA Region V, Chicago,
IL.
Valiela, I., S. Vince, and J. M. Teal. 1975. "Assimilation
of Sewage by Wetlands." Estuarine Processes, M. Wiley,
ed., Academic Press, New York, NY. pp.234-253.
Wang, H. K. 1984. "Sewage Irrigation in China."
International Journal for Development Technology, ^:4.
Weber, S. A. and G. Tchobanoglous. 1982. "Factors Affecting
Nitrogen Removal in Water Hyacinth Treatment Systems."
1981 National Environmental Engineering Conference,
American Socity of Civil Engineers, New York.
71
-------
Wile, I., G. Palmateer, and G. Miller. 1981. "Uses of
Artificial Wetlands in Wastewater Treatment."
Proceedings of the Midwest Conference on Wetland Values
and Management. St. Paul, MM. pp. 255-271.
Wile, I., G. Miller, and S. Black, 1985. "Design and Use of
Artificial Wetlands." Ecological Considerations in
Wetlands Treatment of Municipal Wastewaters , P .' J .
Godfrey, et al. (eds.), Van Nostrand Reinhold, New York,
pp. 26-37.
Williams, R. B., D. J. Reardon, J. Schefchik. 1985. "Design
and Startup of 770-Acre Wetlands Project." Annual
California Water Pollution Control Federation Conference
and Short School, Anaheim, CA. , May 1985.
Williams, T. C. and J. C. Sutherland. 1979. "Engineering,
Energy, and Effectiveness Features of Michigan Wetland
Tertiary Wastewater Treatment Systems." Aquaculture
Systems for Wastewater Treatment: Seminar Proceedings'
and Engineering Assessment, EPA 430/9-80-006,
pp. 141-173.
Wolverton, B. C. 1979. "Engineering Design Data for Small
Vascular Aquatic Plant Wastewater Treatment Systems."
Aquaculture Systems for Wastewater Treatment; Seminar
Proceedings _ and _ Engineering _ Assessment, EPA
430/9-80-006, pp. 179-191.
Wolverton, B. C. 1981. "Hybrid Wastewater Treatment System
Using Angerobic Microorganisms and Reed (Phragmites
Communis)." Economic Botany 36:373-380
Wolverton, B. C. 1985. The Role of Plants and Microorganisms
in Assuring a Future Supply of Clean Air and Water, A
Summary of NSTL Research. National Space Technology
Laboratories, NSTL, MS.
Wolverton, B. C. 1986. ' Microbial-Plant Filters (Artificial
Marshes) for Treating Domestic Sewage and Industrial
Wastewater." Presented at Louisiana Department of
Environmental Quality Seminar, Baton Rouge, LA.
Wolverton, B. C. and R. C. McDonald. 1981. "Natural
Processes for Treatment of Organic Chemical Waste." The
Environmental Professional. 3:99
Wolverton, B. C. and R. C. McDonald. 1982. "The Role of
Vascular Aquatic Plants in Wastewater Treatment." The
Herbarist. 48:25-29.
72
-------
Wolverton, B. C. , R. C. McDonald, and W. R. Duffer. 1983.
"Microorganisms and Higher Plants for Wastewater
Treatment." Journal of Environmental Quality. 12;236.
Wolverton, B. C., R. C. McDonald, and L. K. Marble. 1984.
"Removal of Benzene and Its Derivatives from Polluted
Water Using the Reed/Microbial Filter Technique."
Journal of the Mississippi Academy of Sciences.
29;119-127.
Wolverton, B. C. and R. C. McDonald-McCaleb. 1985.
"Biotransformation of Priority Pollutants Using Biofilms
and Vascular Plants." Submitted to Mississippi Academy
of Science Journal.
Yonika, D. A., et al. 1979. "Feasibility Study of Wetlands
Disposal of Wastewater Treatment Plant Effluent."
Research Project 78-04, Massachusetts Water Resource
Commission, Division of Water Pollution Control.
Zoltek, J. and S. E. Bayley. 1978. Removal of Nutrients from
Treated Municipal Wastewater by Freshwater Marshes.
Centerfor Wetlands,Gainesville,FL.~
73
------- |