Constructed Wetlands for
Animal Waste Treatment
..-•-* jp*«v
A Manual on Pe|fo:piiance* Design, and Operation
".-X With Case Histories
Prepared far th^
V ; '••'-.:-- v-"..V. v •-•'' •
Chdtyf Mexico
Conervation
' •
Industry
{ASWCC)
. /
- :" and the
Nation^itnci;the jJff and Paper
r A if in^Btream ilrfipro&ment, (NCASI)
'"•' - "--'-- ---; '-"•:-.;"-' •"-:,'-.:'-. '.'"-''•• -*^ /".' •" -. ' ' • V ' '
Hill
-------
Constructed Wetlands for
Animal Waste Treatment
A Manual on Performance, Design, and Operation
With Case Histories
Prepared for the
Gulf of Mexico Program
Nutrient Enrichment Committee
Under a contract to
Alabama Soil and Water
Conservation Committee (ASWCC)
and
National Council of the Pulp and Paper
Industry for Air and Stream Improvement (NCASI)
By
Payne Engineering
and
CH2MHUI
June 1997
-------
Acknowledgments
Funding for this report was provided by the U. S. Environmental Protection Agency's Gulf of
Mexico Program (GMP) and the GMP's Nutrient Enrichment Committee. Director of the Gulf of
Mexico Program is Dr. James Giattina. Co-chairs of the Nutrient Enrichment Committee are Lon
Strong with the Natural Resources Conservation Service (NRCS) and Dugin Sabins of the
Louisiana Department of Environmental Quality. Project Officer for this project is Dr. Fred
Kopfler.
A portion of the project was funded through a grant to the Alabama Soil and Water Conservation
Committee (ASWCC), Stephen M. Cauthen, Executive Director. Victor W. E. Payne, Jr., PE, of
Payne Engineering was project leader and co-editor for this publication on behalf of the ASWCC
and was co-author of Section I.
The National Council of the Paper Industry for Air and Stream Improvement (NCASI) also
provided support for the project. Dr. Robert L. Knight, Senior Wetlands Ecologist with CH2M
Hill in Gainesville, Florida, coordinated work on behalf of NCASI and was a co-editor of the
publication and co-author of Section I.
A number of researchers from around the country and Canada provided case histories for their
projects and completed the standard forms provided to them. The names of the various
contributors have been noted in the Case Histories Section of the report.
Mr. Don Stettler, Environmental Engineer, National Water and Climate Center, NRCS, Portland,
OR, provided technical review and support for the project on behalf of NRCS.
Appreciation is extended to the Non-Point Source Section of the Alabama Department of
Environmental Management for their encouragement in pursuit of this project.
-------
-------
CONTENTS
Constructed Wetlands for Treating Animal Wastes
Section I: Performance, Design and Operation
Introduction 1
Management of Liquid Wastes . 1
The Constructed Wetland as a Treatment Component 2
The Value of Constructed Wetlands 2
Nutrient matching 2
Pollutant reductions 4
Odor control 4
Economics 4
Reduced labor 5
Aesthetics 5
Potential problems 5
Disease transmission 5
Damage by animals 6
Types of constructed wetlands 6
Surface Flow (SF) Constructed Wetlands 6
Subsurface Flow (SSF) Wetlands 8
Floating Aquatic Plant (FAP) Systems 8
Treatment Mechanisms within Wetlands .9
Biochemical conversions 9
Settling/filtration 10
Accretion 10
Volatilization 10
Adsorption 13
Evapotranspiration . 14
Nutrient uptake 14
Vegetation 14
Types of wetland plants 14
Rooted plants for surface flow constructed wetlands 15
Submerged aquatic plants 15
Floating, rooted aquatic plants 15
-------
Emergent herbaceous plants 15
Emergent woody plants 16
The role of emergent herbaceous wetland plants in the treatment process ... 16
As a source of microbial substrate 17
As a facilitator of nitrification / denitrification 17
As facilitators of soil adsorption 18
As a user of nutrients 18
As a filter 18
As a source of shade 18
As a source of new soils and sediment 18
Planning for a SF Constructed Wetland 19
Pretreatment 19
Wastewater characterization 19
Site evaluation 20
Soils 20
Wastewater storage 21
Topography 22
Land area 22
Surface water 22
Groundwater 22
Floodplains 23
Fencing 23
Jurisdictional wetlands 23
Sociological factors 23
Hydrologic and elimatologic data 23
Regulatory requirements 25
Design of SF Wetlands for Livestock Wastes 25
Determining wetted surface area 25
Methods available for sizing 25
Presumptive Method 26
Modified Presumptive (MP) Method 29
Field Test Method #1 31
Field Test Method #2 32
Comparing methods 33
Selecting a method (recommendations) 35
Bottom slope/maximum length 36
Hydraulic retention time 37
Hydraulic loading rate 37
Layout of the wetland 39
-------
Embankments 39
Liners 41
Met / Outlet Structures 41
Water budget .44
Operation and Maintenance 44
Final Comments 46
References 47
Section II. Case Histories
Swine;
Duplin County, NC
F. Humenik, M. Rice, M. Cook, S. Broome, P. Hunt, A. Szogi,
G. Stem, M. Sugg, and G. Scalf 1
Sand Mountain, AL
T. A. McCaskey and T. C, Hannah 5
Dairy:
Tom Brother's Dairy, IN
R. P. Reaves and P. J. DuBowy 9
Alan Scott Dairy, MS
C. M. Cooper and S. Testa III 14
Maiden Valley Dairy, Ontario, CN
P. Hermans and J. Pries 25
Oregon State University Dairy, OR
/. A. Moore and S. F. Niswander 30
Poultry:
Auburn University Poultry Research Unit, AL
D. T. Hill and J. W. Rogers 34
-------
Appendices
Questionnaires for Case Histories A
Typical aquatic plants used in animal waste constructed wetlands B
As-excreted waste values for livestock and poultry C
Conversion tables D
-------
Section I
Performance, Design
and Operation
" turn
-------
-------
Constructed Wetlands for Treating Animal Wastes
Section I: Performance, Design, and Operation
Victor W. E. Payne, Jr., PE, and Robert L. Knight, PhD.*
Introduction
Waste management can be a major problem
for Ivestock producers who grow animals in
buildings, pens, or other confined areas. The
waste produced in these types of facilities
can be in liquid, slurry, or solid form. Solid
wastes, such as that produced at broiler
facilities, are fairly easy to manage. These
wastes, which might include manure and
bedding, are simply scraped and hauled to
land application sites at relatively infrequent
intervals.
Slurries or semi-solid wastes are thick fluids
usually collected in fabricated pits or tanks.
These wastes are often scraped or pumped
from a pit and hauled directly to the land
when conditions permit. Slurry wastes are
apt to have more odors when land applied
than either dry or liquid wastes.
Liquid systems can be more complicated to
manage and can create a much higher risk of
polluting surface and ground waters. These
systems are usually associated with swine,
layer hens, the confinement portion of
dairies, certain beef feedlot operations, and
other less common facilities.
The constructed wetland, as will be shown in
this publication, can be an important tool in
the management of liquid animal wastes.
The recently published Constructed
Wetlands for Livestock Wastewater
Management; Literature Review, Database,
and Research Synthesis (CH2M Hill and
Payne Engineering, 1997), referred to
hereafter as the CWLW report, demonstrates
that constructed wetlands technology has
been used for many years in municipal waste
systems but is relatively new with regard to
animal waste systems; the vast majority of
constructed wetlands for animal waste
treatment have-been installed since 1989.
The CWLW report also illustrates that high
levels of treatment can be achieved with this
technology.
This document is intended to be a state-of
the-art users manual on constructed wetlands
for those engineers, planners, technicians,
and Ivestock producers who have more than
just a casual interest in the subject of con-
structed wetlands for treating livestock
wastes. It win condense some of the
information contained in the CWLW report,
a more data-intensive companion publication
to this document. It will also provide other
practical information intended to help the
user understand the value of constructed
wetlands and how they might be included as
part of an animal waste management system.
The user should refer to the CWLW report if
detailed background tables and other
supporting material are needed.
Management of Liquid Wastes
Liquid wastes from confined animal facilities
include manure, contaminated water, and
other liquids and solids that enter the waste
*Enviroiimental Engineering Consultant, Auburn, AL, and Senior Environmental Scientist, CH2M
Hill, Gainesville, FL, respectively.
-------
stream, such as spilled milk and feed,
bedding material, animal hair, feathers, and
broken eggs. In many cases, the volume of
contaminated water in liquid systems is much
greater than the volume of manure.
Contaminated water includes flush water
used to remove wastes and clean houses and
milking facilities, spilled drinking water,
runoff from open lots and buildings, and
direct precipitation on lagoons and other
open waste storage facilities.
There are instances in 'which
pumping from a lagoon and
directly applying wastewater to the
land may not be the best option.
Over the years, livestock producers have
used a wide variety of techniques and
methods to manage and control liquid
wastes. Many approaches to waste
management have been ill conceived and
have resulted in pollution of streams and
takes, as weE as odor problems with
neighbors.
Four factors will determine the success or
failure of a liquid waste management system:
proper planning, design, installation, and
management. Failure in any one of these
areas will result in failure of the system.
The development of a manageable method of
handling liquid wastes should be addressed
during initial planning using a systems
approach. This approach recognizes that
there is not a single, all-purpose method that
applies to all livestock facilities and that the
system developed must be site specific.
Furthermore, it views all structural,
vegetative and management elements
associated with waste management as part of
a well organized and interrelated system.
Development of the system requires careful
planning and takes into account all factors
associated with collecting, treating, storing,
and land applying the waste. Such factors as
regulatory restrictions, economics,
manpower requirements, operation and
maintenance, and environmental consid-
erations are also part of a well planned
system.
Table 1 illustrates the major elements and
some of the components commonly used in
animal waste management systems. The
constructed wetland is shown as a treatment
component because its purpose is to reduce
the pollution potential of wastewater. If a
constructed wetland is added to an existing
system, which has occurred in most
situations around the country to date, this
addition will affect nutrient management,
water management, and O&M.
The Constructed Wetland as a
Treatment Component
D The Value of Constructed Wetlands
It would appear that the most straight-
forward and useful way to dispose of liquid
wastes is through direct pumpout or hauling
from the lagoon or other storage structure to
a land application site. In most cases this
may be the best approach. However, there
are instances in which this approach cannot
be used or, possibly, should not be used. The
following situations, summarized from Miller
et al., (1996), Hughes et aL (1996), and
Payne et al. (1996), illustrate the usefulness
of a constructed wetland in a waste
management system.
1-2
-------
Table 1. Agricultural Waste Management System Functions and Components
System
Function
Production
Collection
Waste
Transfer to
Storage or
Treatment
Storage
Treatment
Wastewater
recycling or
transport to
fields
Utilization or
disposal*
Component
Category
Structural
Vegetative
Management
Structural
Vegetative
Management
Structural
Vegetative
Management
Structural
Vegetative
Management
Structural
Vegetative
Management
Structural
Vegetative
Management
Structural
Vegetative
Management
Typical Components
Roof gutters, downspouts, diversions
Grassing of diversions and around production facilities
Maintenance of leaky waters, recycling flush water, measures to
reduce feed spillage, maintenance of structural components
Alleys & gutters; slatted floors; scrapers and flush systems; curbs;
pumps & pipelines; fences; solids storage pads; diversions.
Grassing of diversions
Maintenance of all structural and vegetative components.
Sumps; pumps; gravity and pressure pipes; flumes; valves; weirs.
Vegetation around sump boxes and along flumes
Maintenance of structural and vegetative components
Waste storage ponds, waste storage tanks, waste stacking facilities;
fences; loading ramps; pumps; pipes and pipelines.
Grassing around facilities for erosion control
Maintenance of structural components and vegetation; managing
water levels.
Waste treatment lagoons, composters, solid/liquid separators, settling
basins, constructed wetlands, overland flow treatment
Grassing of lagoon and wetland embankments; plants in the
constructed wetlands.
Maintenance of structural components and vegetation; maintenance
of water levels in lagoons and constructed wetlands
Pumps; pumphouses; pipelines; valves; hauling equipment;
Grassing around pumphouses, over pipelines.
Maintenance of structural components; preventing plugging of pipes
and pumps with debris and struvite.
Irrigation and hauling equipment; biogas generators; refeeding;
bedding; monitoring stations for permitted discharge.
Vegetation at spreading site; diversions; filter strips; riparian zones.
Soil testing; cutting and managing vegetation; maintenance of
structural components; collecting water samples.
*Disposal refers to a permitted discharge. Although possibly allowed, it is not usually recommended.
1-3
-------
1, of a
to the
at of the and at
If 30 is
to a
but 20 is the
has a of on
to this (1) to a
can
(2) the
or (3) add to
the
to or
not be i if a
be if
a a in
the or
or if the and
are
the Is out, the
is the
he a as to the
of foi
aid
the are io a
of this
2. of
is not a
do not
in any
A of 13 in
May et ai» that
a
in a hut the
a
and
for TSS,
and the use of a
to
in
tint or
of
the If a the
to be to
the and
si ihe
restrictions.
'
3, die
or be-
to and in
the
is
ihe he
in a and
to the site
by
4. A will
the of
a and,
the of at the
in cm the of
tor or It can
1-4
-------
also allow the use of smaller and more cost-
effective spreading equipment. The loss in
nutrient value and the loss associated with
land possibly taken out of production to
install the wetland might be balanced against
the lower capital costs of equipment and
reduced labor requirements (Hughes et at,
1996). From an economic standpoint, each
system must be evaluated on its own merits
to determine if the installation of a
constructed wetland will provide an
economic advantage.
5. Reduced labor: Installation of a con-
structed wetland could reduce land
requirements at the application site to the
extent that the producer could install a
simple solid set irrigation system as opposed
to a more labor intensive traveling gun or
center pivot irrigation system. Even if the
economics do not favor the wetland/solid set
irrigation system, many producers would be
willing forgo a small measure of economic
benefit to reduce the amount of time spent in
handling wastes.
6. Aesthetics: The constructed wetland can
be a nice addition to the farm enterprise.
Some very attractive, flowering aquatic
plants have been used successfully to treat
wastewaters; however, not all are suitable
for the high strength wastes often associated
with livestock wastewaters. The planner will
need to determine the strength of the waste
and the suitability of decorative or exotic
plants to survive in such an environment.
Even if the more colorful plants cannot be
used, traditional plants such as cattail (Typha
spp.) and bulrush (Scirpus spp.) are
attractive and suitable for treating most
wastewaters.
D Potential problems
Disease transmission: Occasionally,
questions are raised about the chance of
diseases being transmitted by wildlife which
enter the constructed wetland and then move
to another location. There is no doubt that
certain birds, land mammals, reptiles and
insects are attracted to treatment wetlands.
In the case of municipal wetland treatment
systems, which have been used worldwide
for decades, no disease transmission
problems caused by migrating animals have
ever been reported, even though many of
these systems have been used as nesting sites
for waterfowl.
If the wetlands do not include open water
areas, bird populations will be limited to
nonaquatic species, such as redwing
blackbirds and yellow headed blackbirds. In
other words, these birds will not normally
have contact with the polluted water. Such is
usually the case with wetlands for treating
animal wastes.
In non-wetland situations, it should be noted
that kiUdeer often walk on the flotsam of
animal waste lagoons and egrets dine on the
droppings of cattle without concern for
disease transmission. It is also noted that a
number of scientists and researchers have
described the positive benefits to wildlife
associated with municipal waste treatment
wetlands (Guntenspergen et al., 1993;
Lofgren, 1993; Kadlec and Knight, 1996).
Thus, based on extensive experience with
municipal wetlands, the limited access of
waterfowl to animal waste wetlands, the
already extensive access of birds to other
sources of animal wastes, and the apparent
lack of evidence linking treatment wetlands
1-5
-------
to health hazards, it appears that potential
problems regarding disease transmission by
birds associated with animal waste wetlands
are minimal.
There appears to be no evidence that other
mammals, reptiles, or insects have contracted
any diseases from constructed wetlands or
that they will migrate to other locations and,
thereby, transmit diseases.
Damage by animals: Both nutria and
muskrats have been a problem in animal
waste constructed wetlands. These animals,
if not controlled, may burrow between
wetland cells and may also destroy vege-
tation. Vigilance is required to assure that
these types of invaders are controlled. Where
these problems are known to have occurred,
the animals have been removed and measures
were taken to prevent farther access.
Cattle have also entered treatment wetlands
and damaged vegetation. If grazing animals
could be a problem, fencing may be required
to protect vegetation, embankments, and
water control structures.
D Types of constructed wetlands
Three types of constructed wetlands could
be used for treating animal wastes: surface
flow (SF), subsurface flow (SSF), and
floating aquatic plant (FAP) systems (Figure
1). While natural wetland systems are in
some cases used for municipal treatment,
they are not considered to be "constructed"
wetlands, and they are not likely candidates
for the treatment of animal wastes. Therefore
the following discussion describes only the
principal types of constructed wetlands.
1. Surface Flow (SF) Constructed Wetlands:
The SF wetland is the most commonly used
wetland for treating animal wastes and is the
only one currently recommended by the
USDA Natural Resources Conservation
Service (USDA NRCS, 1991).
SF constructed wetlands are shallow
impoundments planted with rooted,
emergent vegetation. (Emergent means the
plant structure extends above the water
surface.) Wastewater is treated as it passes
over the bottom of the wetland and through
the plants and bottom litter.
Surface flow constructed
wetlands are shallow
impoundments planted with
rooted, emergent vegetation.
Plant uptake of nutrients by the aquatic
vegetation is very small in relation to the
total nutrient load in the water column;
therefore, removal of nutrients from the
wetland by plant harvesting is considered
unnecessary. Instead, nutrients and
biodegradable organics in the wastewater are
efficiently converted and removed in the SF
wetland primarily through the natural
assimilative capacity of the mierobial flora
(principally bacteria and fungi). The various
mechanisms involved in treatment are
discussed in detail under Treatment
Mechanisms.
The advantages of the SF wetland include
(1) their ability to efficiently treat high
strength wastes associated with discharges
from animal waste lagoons and other
1-6
-------
Distribution Pipe
Outlet Weir
Surface Flow Constructed Wetland
Distribution Pipe
gravel substrate
Subsurface Flow Constructed Wetland
lined basin
Floating Aquatic Plant (FAP) System
Figure 1: Types of Constructed Wetlands for Waste Treatment
1-7
-------
pretreatment facilities; (2) their relatively low
cost compared with subsurface systems; (3)
their relative ease of management; and (4)
their ease of repair and maintenance should
problems occur. For these reasons and the
fact that USD A currently recommends only
SF systems for animal waste treatment, the
surface flow constructed wetland will be the
focus of this report.
An on-site constructed wetland successfully
treats domestic wastewater
2. Subsurface Flow (SSF) Wetlands:,
The SSF wetland contains gravel, rock or
soil media, placed below ground level,
through which the wastewater passes in a
horizontal direction. The water level remains
just below the surface elevation of the
porous bed. Emergent hydrophytic plants are
grown on the surface with the roots pene-
trating the saturated, porous medium. The
bed and the penetrating roots contain a large
surface area on which bacteria grow; thus,
the system functions somewhat like a rock
trickling filter at a municipal wastewater
treatment plant. But unlike the trickling
filter, the roots appear to provide micro-
scopic zones of aeration which aid the
treatment process. (See Vegetation.)
SSF wetlands have an advantage in cold
climates because treatment occurs below the
ground surface, and bacterial communities
are thereby insulated somewhat from the
frigid air. In addition, SSF systems have
virtually no odors, and mosquitoes are not a
problem. When properly designed, gravel
based wetlands are highly efficient at
removing biodegradable organic matter and
nitrates from wastewater.
When used to treat dilute wastewaters, SSF
wetlands can be planted with various types
of attractive plants such as canna lilies,
elephant ear, and spiderwort (Canna
flaccida, Colocasia esculenta, and Trades-
cantia spp., respectively). Such systems are
being used successfully in rural areas to treat
domestic wastewater, especially for single
family residences.
A major disadvantage of the SSF wetland is
the potential for plugging, causing water to
pool on the surface. The potential for
plugging would be much higher for livestock
systems, which usually contain very high
solids concentrations. In addition, the
installation cost is typically at least five times
more than for the same area for SF systems
(Kadlec and Knight, 1996). Thus, because of
the potential for plugging and the high costs
of installation, SSF systems are not being
seriously considered for the treatment of
livestock wastes. If these systems were to be
used at all, it would likely be in colder
climates and only for certain components of
small scale livestock facilities having
wastewaters with a low solids content and
low water volume.
3. Floating Aquatic Plant (PAP) Systems:
Several different FAP systems have been
used for wastewater treatment. These
systems commonly use floating aquatic
1-8
-------
species such as duckweed (Lemna spp. or
Spirodela spp.) or water hyacinths
(Eichomnia crassipes). This vegetation
takes up nitrogen, phosphorus, and metals,
which can be physically removed by plant
harvesting. In addition, microbes attached to
plant roots assimilate biochemical
oxygen-demanding substances, nitrify
ammonium (NH4) to nitrate (NO3), and
denitrify NO3 to nitrogen gas. The dense
vegetative mat that forms on the water
surface effectively shades out algal
populations.
Intensively managed FAP systems can meet
low effluent limits for nutrients without using
chemical additions. Since a limited number of
FAP systems are currently operating, very
little information is available on design,
costs, and performance with highly enriched
livestock wastewaters. Thus, it is difficult to
compare FAP systems with other treatment
wetland technologies.
However, based on data for SF and FAP
municipal systems, FAP systems have lower
reaction rates, higher construction and
operating costs, more sensitivity to cold
temperatures, and more susceptibility to
plant pests and pathogens. Polyculture
(multiple species) systems that use a
combination of floating aquatic plant species
offer an alternative with less intensive pest
management requirements. Also, FAP
systems that use greenhouse enclosures in
colder climates can be considered for small
livestock operations with relatively dilute
wastes.
D Treatment Mechanisms within
Wetlands
A number of physical, chemical, and
biological or biochemical treatment mech-
anisms occur within a treatment wetland.
These mechanisms are often interrelated;
some are simple and some complex. In
addition, some of the mechanisms are not
fully understood technically or in terms of
their overall contribution to the treatment
process.
The mechanisms are listed below with a brief
commentary on each. More detail will
emerge in subsequent discussions.
Some of the treatment
mechanisms within wetlands
are not fully understood
technically or in terms of their
overall contribution to the
treatment process.
1. Biochemical conversions: The largest
single factor affecting treatment is the
conversion of various chemical compounds
through the activity of bacteria and fungi.
Organic compounds, represented analytically
through such tests as biochemical oxygen
demand (BOD5), chemical oxygen demand
(COD), and volatile solids (VS), are reduced
under anaerobic conditions to innocuous end
products such as carbon dioxide (CO2) and
methane (CH4). Under aerobic conditions the
end products are CO2 and water.
Nitrogen is converted by microorganisms
from the organic form (Org-N) to ammonia
forms (NH/ + NH3). If aerobic conditions
are present, the ammonia will then be
converted to nitrite (NOj) and nitrate (NO3),
If the nitrate then enters an anaerobic zone, it
can be converted by bacteria to a gaseous
1-9
-------
Table 2. Typical ammonia concentrations in the supernatant and sludge of anaerobic treatment
lagoons for dairy, swine, and poultry.*
Type
Dairy
Swine
Poultry
NH4* + NH3-N concentrations (mg/L)
Supernatant
200
219
549
Sludge
2498
758
918
*Source: Modified from USDA-NRCS, 1992.
form (principally N^ which can be liberated
to the atmosphere. (See nitrogen cycle,
Figure 2.)
Organic phosphorus and other compounds
also undergo biochemical conversions.
Unlike nitrogen, phosphorus and some other
chemical constituents are conservative and
do not have a gaseous state; thus, they will
be "removed" through other mechanisms
noted below or will be discharged.
2. Settling/filtration: These are perhaps the
simplest mechanisms and involve the
deposition of solids on the floor of the
wetland and entrapment by plant stems and
bottom litter. Both floating matter and other
suspended material may be retained through
these mechanisms. The organic fraction of
the solids will be degraded biochemically.
Some of the inert material or slowly
degradabte material may eventually become
part of the peat bed which forms through
accretion.
3. Accretion: This term refers to the physical
buildup of material on the floor of the
wetland as new soil Recent additions of
loose vegetative litter or thatch are not
considered part of the accreted material The
accretion rate will typically be less than one-
half inch (1.2 cm) per year and will consist of
settled wastewater solids, the remnants of
decayed litter, and microbial biomass.
Accretion will be a principal removal process
for phosphorus and certain metals.
4. Volatilization: This refers to the loss of
constituents to the atmosphere in gaseous
form. The process can be biochemical or
strictly chemicaL The conversion of NO3 to
gaseous nitrogen through denitriflcation has
already been noted under Biochemical
conversions above and wiH be discussed
further under Vegetation. The discussion
here will focus primarily on the volatilization
of ammonia because of its importance in
animal waste management systems.
There are only two natural mechanisms by
which nitrogen can be lost to the atmosphere
from a waste treatment system: ammonia
volatilization and denitriflcation (see Fig. 2).
In order for denitrification to occur (an
anaerobic process), nitrogen must first be
converted to nitrate (NO3) from the ammonia
form (an aerobic process).
Livestock pretreatment lagoons are nearly
always anaerobic; thus, nitrogen will be in
the organic and ammonia forms only.
Typically, the supernatant will contain 20 to
1-10
-------
.^
*^s>
Figure 2: The Nitrogen Cycle (Ref. AWMFH)
i-11
-------
40 percent organic nitrogen and 60 to 80
percent ammonia. The concentrations of
ammonia nitrogen in the supernatant of
animal waste lagoons will be much higher
than those of untreated domestic sewage,
which, for a medium strength waste, will
only be about 25 mg/L. (See Table 2.) Since
there is virtually no oxygen present within
thft water column of most anaerobic animal
waste lagoons, the conversion of ammonia to
nitrate cannot occur and, hence, without
nitrate, denitrification is impossible. While
some nitrification may occur at the air/water
interface of an anaerobic livestock lagoon
where oxygen diffusion could occur, it
appears that volatilization rather than nitrifi-
cation/denitrification may be the principal
mechanism for nitrogen loss within lagoons.
It should be noted, however, that only a
small fraction of dissolved gaseous NH3 is
usually present in livestock wastewaters,
with the amount being a function of pH and
temperature. The higher the pH and the
warmer the water, the large the fraction of
NH3 present. Table 3 illustrates the shift in
concentration between NH4* and NH3.
The pH in animal waste lagoons is usually in
the range of 7.0 to 8.0. From late spring
through fall, water temperatures of lagoons
sampled in Alabama were between 25 and 30
degrees C, with pH ranging from 7.2 to 7.5.
While un-ionized ammonia would be only
about three percent of total ammonia in these
lagoons, much more than this amount could
be lost through volatilization. The reason for
this is explained by the following equation:
ET
Table 3. Percent un-ionized ammonia, (Ntta) in aqueous ammonia solutions.*
Temperature
CO
15
20
25
30
pH
6.0
0.027
0.040
0.057
0.080
6.5
0.087
0.13
0.18
0.25
7.0
0.27
0.40
0.57
0.80
7.5
0.86
1.2
1.8
2.5
8.0
2.7
3.8
5,4
7.5
8.5
8.0
11.0
15
20
*Source: modified from EPA, 1974.
The equation indicates that if NH3 in an
aqueous solution is lost as a gas, the
equation shifts to the right to maintain the
equilibrium. Consequently, as NH3 at the
water/air interface of a lagoon is lost to the
atmosphere, more NH4 is converted to NH3
which, in turn, is available for volatilization.
In municipal treatment, "ammonia stripping"
towers are used to drive off un-ionized
ammonia, a process which is facilitated by
fans or blowers and by raising the pH above
10. In lagoons only the movement of wind
across the surface enhances the volatilization
rate. In addition, temperature is an important
1-12
-------
factor in this process. In stripping towers for
municipal wastewaters, 90 to 95 percent of
the ammonia is typically driven off at 68°F
(20°C), but only about 75 percent is lost at
50°F (10°C) (Corbit, 1989).
It appears, therefore, that the lagoon
accomplishes ammonia stripping but at a rate
much slower than municipal systems on a
unit area basis. While the amount lost over
the total surface area of a lagoon is obviously
great, there is some question as to whether
or not ammonia volatilization is the principal
mechanism in nitrogen removal from
livestock wetlands.
Ammonia gasification over wetlands vege-
tated in rice and fertilized with ammonium
fertilizer has been studied, and the loss rates
were found to be comparable to plant uptake
for dense stands of macrophytes (Freney et
al., 1985). Kadlec and Knight (1996)
summarized other studies dealing with
ammonia gasification for municipal systems
and an acid bog (Billmore et. al., 1994, and
Hemond, 1983). They conclude that
"volatilization typically has limited
importance, except in specific cases where
ammonia is present at concentrations greater
than 20 mg/L."
It would appear, therefore, that ammonia
volatilization may be the most significant
mechanism for nitrogen loss within
constructed wetlands which treat animal
wastes. The principal reasons are as follows:
a. Total ammonia concentrations in the
pretreated wastewater entering livestock
constructed wetlands are nearly always
greater than 20 mg/L, the concentration
above which volatilization becomes
important; often it will exceed 100 mg/L.
b. Most pretreated livestock wastewaters
discharged to treatment wetlands are
typically anaerobic, which means that
nitrification in the surface flow will be limited
and, consequently, any subsequent denitri-
fication will also be limited.
c. Limited research on rice fields treated with
inorganic ammonia fertilizer indicates that
ammonia volatilization is occurring, and it is
expected that the conversion rates would be
even higher for organic wastes with high
ammonia concentrations.
Further research related to ammonia
volatilization from animal waste constructed
wetlands is needed. (See additional
discussion on ammonia losses under
Adsorption, Evapotranspiratipn, and
Vegetation.) ' . ;"
It would appear that ammonia
volatilization may be the most
significant mechanism for
nitrogen loss within wetlands
used to treat animal wastes
5. Adsorption: Adsorption refers to the
binding of one constituent to another
through the chemical attraction of oppositely
charged particles. Positively charged
compounds such as the ammonium ion
(NH4+) are attracted to negatively charged
clay particles. Various forms of phosphorus
can be bound with calcium, aluminum and
iron, al common constituents of soils. It is
likely that initially high rates of phosphorus
and NH4+ removal in the early years of
wetland life may be attributed to adsorption
within the soil matrix. As the wetland
matures and more of the adsorption sites
1-13
-------
become occupied, the apparent net treatment,
efficiency for these chemicals drops.
6. Evapotranspiration: As wastewater flows
over the surface of the soil in the wetland, an
amount of water approximately equal to that
required by the plants for evapotranspiration
(ET) is drawn into the root zone. (This
assumes that the wetland was properly sited
and that no lateral subsurface water enters
the root zone.) The average ET rate for
Lemna is commonly found in animal waste
constructed wetlands.
vegetated wetlands is approximately equal to
lake evaporation rates (70 to 80 percent of
pan evaporation). Thus, a geographic area
with 36 inches (92 cm) of annual lake
evaporation wil extract from the overlying
water about 3 ft3 (2.83 x 10'2 m3) of water
per year per square foot (0.009 m2) of
wetland surface. A livestock wetland with
wet surface dimensions of 100 x 400 ft (30.5
x 122 m) would withdraw about 120,000 ft3
(3,396 m3) or 897,000 gallons of water per
year. Evapotranspiration becomes especially
important in areas with low rainfall and high
evaporation rates.
Evapotranspiration rates will affect hydraulic
retention time (HRT), which is an important
factor in design. In addition, since ET relates
directly to the amount of wastewater drawn
into the soil profile, it is an important factor
in pollutant removal associated with adsorp-
tion and, possibly, nitrification/ denitrifl-
cation that may occur within the root zone
(see Vegetation).
1. Nutrient uptake: Wetland plants extract
nitrogen, phosphorus, potassium and various
minor nutrients from wastewater. These
nutrient removal rates may be significant
during initial development of wetland plant
biomass. However, the majority of these
nutrients will be recycled back to the water
on an annual basis, resulting in a relatively
small net removal rate by the plants
compared with the total amount fed to the
system. For this reason, harvesting is not a
viable alternative for improving nutrient
removal where most non-commercial
wetland species are concerned.
D Vegetation
A wide variety of wetland plant species have
been used in or have invaded and adapted to
the wastewater environment of constructed
wetlands, especially in municipal systems.
Guntenspergen et al. (1989) listed 17
emergent species, 4 submerged species, and
11 floating species that have been used in
constructed wetlands for wastewater
treatment. The pollutant concentrations in
livestock wastewaters are typically much
higher than in municipal systems; thus, some
species which have adapted to municipal
systems will not survive in livestock
wastewaters.
Types of wetland plants: Wetland plants
may be broadly classified as floating or
rooted. The floating, unattached vegetation
includes such common plants as duckweed
(Lemna spp.), water hyacinths (Eichhornia
crassipes), and algae in a wide variety of
1-14
-------
species. Algae and duckweed will often
colonize in the open spaces of livestock
wetlands. The extent to which these species
are successful is limited by the amount of
open space and the shade provided by the
rooted emergent species. While the floating
plants may invade livestock constructed
wetlands, they are not currently recom-
mended for the purposeful inclusion in these
systems. Lemna and water hyacinths would
have to be harvested to be most effective,
and such labor intensive systems would not
be welcome by most livestock producers.
Thus, no further discussion of this type plant
will be provided here.
Rooted plants for surface flow
constructed wetlands: Within the rooted
group are the submerged, floating, emergent
herbaceous, and emergent woody plants.
These are discussed briefly below, with
indications of their possible use in livestock
constructed wetlands.
2. Submerged aquatic plants may grow in
the water column of deeper pools within
wetlands. Through photosynthesis, they can
release large quantities of dissolved oxygen
directly into the water column and, in turn,
promote organic decomposition and
nitriflcation. Unlike some forms of algae,
submerged aquatic plants do not typically •
add to significant increases in suspended
solids. In most enriched wetlands where
floating plants cover the deep zones,
submerged aquatic plants will be shaded and
unable to compete effectively for light. Their
use in animal waste treatment wetlands
should be the subject of research.
2. Floating, rooted aquatic plants include
such species as pennywort (Hydrocotyle
spp.), water lilies (Nymphaea spp,),
spatterdock (Nuphar spp.), and pondweeds
(Potemogeton spp.)- The roots of these
plants can extend from 4 to 25 inches (10 to
60 cm) into the water column, depending on
the wastewater characteristics, and those
rooted in the bottom can be much longer.
The cover they provide can significantly
influence water temperature. By inhibiting
the growth of algae and reducing
temperatures, the floating rooted plants can
also influence dissolved oxygen
concentrations. Likewise, the cover provided
by these plants may also inhibit ammonia
volatilization.
Pennywort spreads between the emergent
plants at this swine waste treatment wetland.
Pennywort is a natural invader of the swine
wastewater constructed wetland project at
Sand Mountain, Alabama. It roots at the
edge and then grows out as a floating mat
over deeper water. At Sand Mountain it has
successfully filled open areas between the
emergents in the primary cells.
3. Emergent herbaceous plants are rooted in
the soil and have plant structures that emerge
or stand upright above the surface of the
water. The herbaceous nature of these plants
includes non-woody structures that allow the
plant to stand erect without the support of
the surrounding water. These plants have
1-15
-------
lenticels (small openings in the leaves and
stems) that allow air to move in and out;
vascular or aerenchymous tissue that allows
gaseous diffusion or air convection through
the length of the plant; and extra physio-
logical tolerance of chemical by-products
resulting from growth in the anaerobic soil
environment (Kadlec and Knight, 1996).
The emergent herbaceous plants are the only
ones currently recommended for planting in
constructed wetlands used for livestock
waste treatment. Although a wide variety of
plants have been used in municipal systems,
the selection becomes more limited for those
livestock systems with high concentrations of
BODS and ammonia.
The most common emergent herbaceous
aquatic plants in treatment wetlands,
including those for livestock wastes, are
cattails (Typha spp.), bulrush (Scirpus spp.),
and common reed (Phragmites spp.). Plants
such as duck-potato (Sagittaria spp.)', giant
cutgrass or American wild rice (Zizaniopsis
milicaea) and other varieties have also
performed well in livestock constructed
wetlands. Data on the level of treatment or
the biomass produced by these different
species in animal waste wetlands are limited.
(See Appendix B for typical species used.)
A variety of planted and naturally colonizing
herbaceous aquatic macrophytes might exist
in any given treatment wetland. In fact,
polytypic stands of vegetation are better than
monotypic stands for the wetland's ecolo-
gical balance. When monotypic stands of
cattail or bulrush have been studied, research
has indicated no clear advantage of using
specific plant species for reducing BOD5,
TSS, TN, or TP in surface flow treatment
wetlands (KMght, 1996).
4. Emergent woody plants are categorized as
shrubs, trees (canopy and subcanopy), or
woody vines. The distinguishing character-
istics of woody vegetation include its bark,
non-leafy vascular structures, decay-resistant
tissues and relatively long life. In general,
woody plants are larger than emergent
herbaceous wetland plants and may shade
out smaller species.
A variety of woody plants have been used in
municipal treatment wetlands. In the
southeast, the most common tree species
used in waste treatment include cypress
(Taxodium spp.), willow (Salix spp.), ash
(Fraxinus spp.), and gum (Nyssa spp.). In
the north, species of willow along with
spruce (Abies spp.), birch (Betula spp.), and
alder (Alnus serrulata) have been used.
Woody aquatic plants would probably not be
useful in constructed wetlands for livestock
wastewaters except in the tertiary phase and
where the system can be controlled to allow
alternate periods of wetting and drying.
Guntenspergen et al. (1989) indicate that
"few woody species survive in permanently
flooded soil." While species such as willows,
cypress, and blackgum can, indeed, survive
continuous flooding, they may not survive
the high organic and nitrogen loadings
typical of treated livestock wastes. More
information is needed on the various woody
species before recommending them for use
with animal waste wetlands.
The role of emergent herbaceous
wetland plants in the treatment
process: The herbaceous emergents have
unique features that help them survive in an
otherwise hostile environment. Many can
withstand continuous flooding and anoxic
soUs, and a few can thrive and proliferate in
1-16
-------
wastewaters with high pollutant loads.
Certain exotic species (elephant ear, canna
lilies, calla lilies, etc.) have done well in
domestic, on-site sewage treatment
wetlands, but many of these have failed in
the Ivestock wastewater environment. In
addition, some exotics cannot withstand the
harsh winters outside the lower to middle
South.
The primary function of the herbaceous
emergents is to facilitate waste treatment;
that is, they provide the means through
which the treatment mechanisms can occur.
As facilitators, the plants play several roles in
the treatment process. These roles are noted
below and are explained in terms of the
treatment mechanisms already noted.
1. As a source ofmicrobial substrate: The
wetland plants provide substrate for bacteria
and fungi, the source of biochemical
conversions of pollutants. This substrate is
important as a source of reduced carbon that
provides required energy for microbial
growth. It also provides a large surface area
upon which the microbial populations grow.
Reed et al. (1995) indicate that the micro-
organisms which populate the submerged
plant stems, fallen leaves, roots, and
rhkomes are responsible for much of the
treatment within the wetland. Kadlec and
Knight (1996) suggest that the submerged
substrate, comprised of a complex mixture of
plant litter in various stages of decompo-
sition and its highly productive biological
communities, is responsible for as much as
90 percent of the overall treatment within the
wetland. Thus, the principal function of the
emergent vegetation in most treatment
wetland systems is to provide the substrate
important in treatment (Kadlec and Knight,
1996). As the wetland's surface area
increases, so also does the substrate surface
area and, hence, the overall treatment
efficiency of the wetland, assuming complete
submergence of the litter and adequate
contact time (hydraulic retention time).
The primary function of the
herbaceous emergents is to
facilitate waste treatment
2, As a facilitator of nitrification /
denitriftcation: The process of converting
ammonia to nitrate and then to nitrogen gas
within a wetland depends upon-having both
aerobic and anaerobic conditions. In order
for ammonia to be converted to nitrate,
aerobic conditions are required for the
obligate nitrifying bacteria. Then for the
nitrate to be converted to nitrogen gas,
anaerobic bacteria are required. The unique
properties of the emergent macrophytes may
make this possible.
AM vascular plants are designed to transfer
oxygen from the surrounding air or water
into their roots via aerenchymous tissue
when conditions prevent normal O2 uptake
by the roots and rhizomes. In the aquatic
environment, direct oxygen transfer into the
roots is greatly restricted. When the surroun-
ding water contains oxygen demanding
organics, totally anoxic conditions may exist
and very little dissolved oxygen would be
available within the root zone. Under these
conditions, the plants draw atmospheric
oxygen into the above-water portions of the
plant through its lenticels and pass it to the
roots via aerenchymous tissue. The amount
of oxygen transported in this manner
typically ranges from 2.08 to 12 g O2/m2/d
(Brix and Schierup, 1990; and Armstrong et
1-17
-------
al., 1990), although higher values have been,
reported. In some cases, excess O2 becomes
available which exudes from the roots and
rhyzomes. This creates aerobic microsites
around the roots which can provide a limited
source of oxygen for the nitrifying bacteria
(Reed et al., 1995; Kadec and Knight, 1996).
Brix and Schierup (1990) reported a net
release of only 0.02 g OJnf/d through the
roots ofPhragmites australis.
Once ammonia is converted to NO3, either
within the water column or within these
microsites, the dissolved NO3 can diffuse
into the surrounding anerobic zone where
denitrifying bacteria convert it to nitrogen
gas (NjO or N-j). Information from
field-scale treatment wetlands is still scarce
on how much oxygen is transferred to the
root zone or how much nitrification or
denitrification occurs (Reed et al., 1995).
Wetland systems are so complex in terms of
types of plants, soils, and a host of other
related factors which couM influence oxygen
transfer and biological activity, that the loss
of N, however it occurs, is currently
explained in terms of general rate constants
based on influent and effluent sampling
rather than on kinetics of individual mierobial
processes (Kadlec and Knight, 1996).
3, As facilitators of soil adsorption: As
noted under Treatment Mechanisms, the
plants draw wastewater into the soil profile
to satisfy their normal water requirements.
The amount is determined by the trans-
piration rates of the plants. By drawing
wastewater into the soil and around the root
zone, the plants facilitate adsorption of
ionized pollutants onto soil particles and the
subsequent nitrification of ammonia and
denitrification of nitrates, as noted above.
4. As a user of nutrients: Plants utilize
nitrogen, phosphorus, potassium and the full
range of minor nutrients. The amount taken
up is usually small in relation to the total
pollutant load, and this process becomes
important only if the plants are harvested.
Otherwise, a high percentage of the nutrients
that are taken up return to the system during
plant senescence. The remaining minor
fraction may be lost as accretion of new soil
The loss of N, however it
occurs, is currently explained
in terms of general rate
constants developed from
sampling many wetlands.
5. As a filter: The plant stems and the litter
from the plants entrap solids. Thus, the
plants facilitate the breakdown of organic
solids by detaining this material and allowing
time for biochemical conversions to take
place. The plants also slow the movement of
water and promote settling.
6. As a source of shade: By shading the
water, plants help regulate water temper-
ature and decrease the light available for
algal growth. The reduction in algae
concentrations will provide an attendant
reduction in suspended solids concentrations
in the wetland effluent.
7. As a source of new soils and sediment: As
vegetation dies and as incoming sediment is
trapped, a layer of material gradually
develops in a process called accretion. The
accretion rate is usually less than 0.5 in (1.3
cm) per year. Some of the phosphorus,
1-18
-------
nondegradable suspended solids, and metals
are permanently trapped in this layer. The
effects of accretion should be considered in
designing embankments.
Planning for a SF Constructed
Wetland
System planners must be keenly aware that
the various elements selected for a waste
management system interact with one
another. If, for example, a structural element
is changed or added in the planning process,
such a change could affect the amount of
nutrients produced or the amount of water
that must be handled. Likewise, if it is found
that the land area for spreading wastes is
limited, the planner may have to modify the
treatment components to reduce nutrient
production. For this reason, an overall waste
management plan, and not an isolated
nutrient management or water management
plan, should be developed. Such a plan ties
the system together on paper, relating
planning, design, installation, and manage-
ment factors. The USDA Agricultural Waste
Management Field Handbook provides
excellent information on the many factors to
consider in developing an agricultural waste
management system (AWMS) plan (USDA,
1992).
A number of factors must be considered
when planning for a surface flow constructed
wetland. Listed below are some of the key
factors to consider with a brief explanation
of each. Professionally trained engineers, soil
scientists, agronomists and others should be
consulted on site-specific details and
methodologies.
D Pretreatment
Wastewaters from all confined animal
feeding operations must be treated in a
kgoon, waste storage pond, or settling tank
prior to being discharged to a constructed
wetland. This requirement is specified in
NRCS guidelines (USDA, 1991) and is
absolutely essential. Untreated wastewater
from animal confinement facilities will have
concentrations of solids, organics, and
nutrients that would kill most wetland
vegetation. In fact, the concentrations of
pollutants in the effluent from some
pretreatment units will stress or kill some
types of wetland vegetation.
Caution should be used with older
pretreatment units that have never had any
sludge removed. In these systems, waste-
water influent may pass over the top of the
sludge in a narrow band with very little
treatment occurring; thus, the effluent will be
unsuitable for discharge to the wetland. It is
imperative that sludge depth be determined
in these older systems and that samples of
the effluent be analyzed. A wetland should
not be installed until the pretreatment unit
has been renovated. The life of the
pretreatment unit, the anticipated sludge
buildup rate, and the future need for cleanout
should be factored into the AWMS plan.
D Wastewater characterization
Waste characterization can be accomplished
by estimating the pollutant loads or by
analyzing the supernatant (the surface liquid)
of the pretreatment unit. If this is a new
system and the pretreatment unit has not yet
been installed or is not fully operational, then
estimates will have to be used.
1-19
-------
If a lagoon or other pretreatment unit is in
place and nearly full, samples of the
supernatant should be taken and analyzed for
TKN, NEj-N, TP, BOD5, TSS, and pH.
Ideally, samples should be collected during
several months to reflect both warm and cold
season effects.
Estimates of wastewater strength can be
made using tables and other information in
the NRCS Agricultural Waste Management
Field Handbook (AWMFH) or in other
professional engineering publications, (See
Appendix C for sample tables.) In using
these methods, the planner should be sure to
evaluate or estimate the pollutants of
concern within only the supernatant (the
liquid portion which wM overflow into the
wetland) and not that in the entire
pretreatment unit (sludge plus supernatant).
Thus, an anaerobic lagoon may reduce the
nitrogen load by 80 percent, according to
some methods of estimating; however^much
of the remaining 20 percent is contained in
the settled sludge. In other words, only
about 10 percent of the original N may be
available for discharge to the wetland via the
supernatant.
Wastewater must also be characterized by
volume. This includes the volume of manure,
flushwater and other constituents. (See
Management of liquid wastewaters above
and Hydrologic and climatologic data,
below, for further discussion on volume
considerations.)
D Site evaluation
An on-site evaluation is essential to obtain
vital information on the physical suitability of
the site. Such factors as soils, depth to
bedrock, and land area should be investi-
gated, but the evaluation should also include
an estimate of the potential impacts of the
wetland on the surrounding area. In addition
to a visual inspection, testing, and sampling,
the planner should use soil maps, contour
maps, aerial photos, and other similar tools,
if available, to help with the assessment.
Some factors to consider in making a site
evaluation are discussed below;
1. Soils: Soil borings or backhoe pits should
be dug at several locations within the
boundaries of the proposed wetland site.
Borings or pits should extend to a depth of
at least one foot below the constructed
bottom of the wetland to identify permeable
seams and shallow bedrock and to generally
characterize soil type.
In order to reduce the potential for seepage,
soils should contain a relatively high fraction
of clayey material. Soils classified as clay,
sandy clay, sandy cky loam, or clay loam
would be suitable for the site. Clayey soils
will inhibit the root growth of nearly all
plants to some extent. However, plants such
as cattails, bulrushes, and reeds adapt to
these type soils, as noted under Vegetation.
If the soil in the top 12 to 15 inches (30.5 to
38 cm) is highly permeable (i.e., sandy), or if
a sand or gravel seam is located within this
layer, the surface material should be tempor-
arily removed and a compacted clay or
fabricated liner installed. Once the liner is
installed, the original material can be
replaced.
Since the rooting depth in surface flow
wetlands wffl typically be less than 12 inches
(30 cm) and approximately 80 percent of the
root mass for most emergent plants will be
within the top 6 inches (15 cm) of soil, the
1-20
-------
top of the liner should be 12 to 15 in (30 to
38 cm) below the intended surface of the
wetland. In other words, the soil/plant
medium which overlies the liner should be at
least 12 inches thick.
Whenever a liner is installed, care must be
taken to ensure that it ties in vertically at the
embankments, thus preventing lateral
movement under the embankments.
Sol permeability should be evaluated in light
of state restrictions on allowable seepage
rates. A typical allowable permeability rate is
1 x W5 cm/sec (0.028 ft/day). The specific
discharge is determined by use of Darcy's
Law as given in the following equation and
illustrated in Figure 3:
Q=k(h/d)A
where—
Q = discharge (seepage),(ft3/d)
k = hydraulic conductivity (ft/d)
h = vertical distance between the
maximum surface elevation of the
overlying liquid and the bottom of
the compacted soil liner (ft)
d = thickness of the soil liner (ft)
h/d = hydraulic gradient (ft/ft)
A = cross sectional area of flow (ft2)
With terms rearranged:
Q/A=k(h/d) or
v=k(h/d)
= specific discharge or seepage
per unit area (distance/time).
The hydraulic head (h) is relatively small for
constructed wetlands (usually less than 18
inches); therefore, the potential for seepage
is expected to be minimal, assuming a
moderately clayey soil is available or a well
compacted clay liner is installed. However, at
questionable sites (sandy soils, underlying
limestone rock, etc.), a detailed evaluation of
potential seepage should be conducted. The
planner is advised to review information
developed by NRCS on this topic, as needed
(USDA, 1993).
The soils investigation will also determine if
shallow bedrock is encountered. If bedrock
Wastewater
Figure 3. Relationship between h and d in
determining thickness of liner
consists of easily solubilized limestone or if
fractured sandstone is located close to the
proposed bottom elevation of the wetland, a
liner should be considered. The character-
istics of the soil and soil depth should be
carefully evaluated in this case.
2. Wastewater storage: Two types of storage
should be considered during planning: (1)
storage downstream of the wetland, used for
recycling, irrigation, and, possibly, winter
storage, and (2) winter storage in the
pretreatment unit. If the wetland will not
have a discharge, then a downstream storage
pond is essential If the wetland will not
function during the dormant season, then
wastewater must be stored during that
1-21
-------
period either in a downstream pond or in the
upstream pretreatment unit. If the upstream
unit is already in place but was not designed
to accommodate the volume of wastewater
needed for this system, it may be easier to
build a downstream storage pond than to add
additional storage to the pretreatment unit.
A water budget will be needed to determine
the required capacity of the storage units
whether upstream or downstream. See later
discussion on Hydrologic and Climatolog-
ical Data and on Water Budget.
3. Topography: The ky of the land has an
important impact on the number of cells that
may be needed and, hence, the overall cost
of construction. All wetland cells should
have level bottoms side-to-side and nearly
level bottoms in the lengthwise direction. If
the land has a considerable slope, it may be
necessary to install several cells in order to
maintain a relatively constant water depth.
With each new cell a new embankment is
needed which will occupy more room in the
overall system.
The wetland should accommodate topo-
graphy in such a way that, wherever
possible, earthwork cuts and fills can be
balanced during construction. A slight slope
in the direction of the outlet end of each cell
may be used to allow for complete drainage
of the cell for maintenance. However, the
same purpose can be achieved by installing a
deep zone at the end of the cell which can be
pumped to facilitate drainage. (See further
discussion under Bottom slopes/maximum
length.)
4. Land area: The wet area of the system, as
determined by appropriate design equations,
may be as little as half the total area required.
(See following section entitled Determining
wetted surface area.) If the land is sloping,
additional cells and embankments will be
needed. In most cases a storage pond will be
pkced downstream of the system to collect
the wetland effluent for recycling and
irrigation. This pond will need to be sized in
accordance with the design requirements of
the proposed irrigation system. If a discharge
will be permitted, additional space may be
needed for flow measuring weirs and a
sample collection station.
5. Surface water: The proximity of the
wetland to the nearest stream or waterbody
should be noted in the waste management
plan. The size and characteristics of the
stream will be important factors if a
discharge is planned. Any receiving stream
must have the capacity to assimikte
wastewaters discharged during low-flow
periods. State reguktory agencies must
determine if the stream has the necessary
capacity to receive wastewaters from the
wetland and provide the necessary
information on permits.
6. Groundwater: The site evaluation must
consider depth to groundwater and proximity
of the system to nearby wells. Allowable
distance to domestic wells will be specified
by the state reguktory agency.
If any wells are in close proximity to the site,
water samples should be collected prior to
installation of the wetlands and be evaluated
for fecal coliform and fecal streptococcus
bacteria, nitrates (NO3-N) and ammonia
nitrogen (NH4+NH3-N). Without
preconstruction sampling, there will be no
evidence of pre-existing contamination if it is
1-22
-------
later found that the wells are contaminated.
any doubt about the location.
If shallow ground water is noted, it is
suggested that at least one monitoring well
be installed down slope of the wetland or in
an area selected by a professional geologist.
State regulatory officials should be consulted
if the seasonal high-water table of such
ground water will be in close proximity to
the bottom of the wetland.
7. Floodplains: Two important questions
should be considered when planning the
installation of a constructed wetland within a
floodplain. First, will placement of the
structure restrict flow to the extent that
damage to upstream or cross-stream
properties could occur? And, second, can the
system be economically protected from a
relatively severe level of flooding, such as
the 50-year or 100-year, 24-hour storm?
State regulatory agencies may require certain
restrictions in this regard. In addition, Corps
of Engineer requirements may come into
play. Thus, both state and Federal regulatory
requirements as well as the overall economic
impact should be considered in designing to
protect from some prescribed storm event.
8. Fencing: Fencing around the site may be
required by state regulations, or it may be
needed if grazing animals could gain access
to the wetland. Cattle have been known to
enter a wetland, destroy vegetation and
damage embankments. Adding fencing will
add to the cost, but it will be essential in
some cases.
9. Jurisdictional wetlands: The site being
considered for a constructed wetland should
not be in a Jurisdictional wetland. A
professional opinion is essential if there is
10. Sociological factors: How will neighbors
react to the wetland? Since the animal
feeding facility and waste treatment facility
may already be in place, the addition of the
wetland should not be a problem; rather, it
should be viewed as a benefit. However,
distance to neighbors should be considered in
light of possible concerns about the type of
wildlife that might be attracted to the
wetland.
O Hydrologic and climatologic data
Monthly data on precipitation, pan
evaporation, and temperature are essential
for design and must be gathered during
planning. The rainfall and evaporation data
provide information needed to determine
hydraulic retention time in the wetland,, the
.-. si2»;of winter storage (if .used), in the ,
pretreatme'nt unit (lagoon, storage pond,
; etc.),, and,'the overai. mottthly;, water .budget,
•' whicliis especiallyiniportant Where land
applicatioh is ;used affef wetlands treatment.
\ -TFhe: VQlume of precipitation: water produced
each month includes that which falls directly
on the pretreatment unit and wetlands,
including embankments; and runoff from
roofs, open lots, and other areas draining
into the system. This information is
combined with data on manure and flush
water volumes to determine initial volume
input to the system.
Evaporation is deducted from the
precipitation input on open pretreatment
structures. This amount will be equivalent to
lake evaporation which is considered to be
70 to 80 percent of monthly pan evaporation.
In addition, evapotranspiration within the
1-23
-------
wetland Is also considered to be roughly
equivalent to lake evaporation.
Temperature data are used in equations to
size the wetland. If the wetland is being
designed for a discharge, the average
monthly ambient temperature for the coldest
period should be used for water temperature.
If the monthly average is below zero degrees
C and wastewater will be treated beneath an
ice cover, special precautions and design
considerations are necessary. Water level
must be raised to a depth equivalent to the
thickness of the anticipated ice cover plus the
expected depth of flow during the winter.
Following freezing, water level is lowered to
allow water flow under the ice. The embank-
ment height must be sized accordingly.
Treatment continues in the wetland even in
muter but at a slower rate.
In northern climates where long periods of
freezing weather can be expected, vege-
tation, ice cover, and snow can provide an
insulating effect for the water. Assuming
water continues to flow throughout the
period of ice cover, a design temperature of
35 - 39° F (2 - 4° C) would provide a
conservative design.
It should be noted that detailed guidance on
calculating temperatures can be found in
various literature (Kadlec and Knight, 1996;
Reed et. al, 1996). These take into account
such factors as temperature of the incoming
water, the length of the wetland, and open
areas. For large livestock facilities with high
volume and year-round flows in cold
climates, these resources should be consulted
for design of livestock wetlands. Otherwise,
the design guidance provided in this
publication will be adequate.
RECOMMENDATION: Despite the fact
that design guidance for constructed wet-
lands is available for cold weather systems, it
is recommended that wastewater from
livestock operations be stored in a lagoon or
waste storage pond during the cold months,
not only in northern climates, but also in all
areas where (a) land applcation will be the
preferred mode of operation versus
discharge and (b) vegetation at the land
application site will be dormant in winter. In
the case of discharge systems, this strategy
will eliminate the possibility of failure of a
wetland designed for winter treatment. For
nutrient matching (land application) systems,
this strategy will reduce wetland size,
eliminate operation and maintenance
problems throughout the winter, and,
generally, provide for ease of management.
If wastewater is stored during the winter
months and released to the wetland only
during the warmer season, and if a
discharge is planned, the average monthly
temperature for the coldest month during the
period of discharge is used in design.
However, if wastewater is released to the
wetland during only the warm season and if
the wetland is designed to reduce nutrients
to a specific level for land application
(nutrient matching), then the average
temperature over all months of the warm
season should be used in design.
1-24
-------
In cold climates knowledge of monthly
temperature data also allows the planner to
advise the user of the date that the water
level should be raised in anticipation of the
onset of freezing weather (assuming a winter
discharge is planned) or the date that
wastewater to the wetland should be
discontinued.
D Regulatory requirements
State water quality reguktors will determine
if the wetland can be permitted for discharge
under NPDES, state, or water conservation
district requirements. If permitting is allowed
and will be part of the system design, the
owner must be fully aware of all monitoring
requirements and the costs of obtaining and
maintaining the necessary permits. If the
system is not allowed to discharge, the
owner must plan on having a storage pond to
collect the wetland effluent for irrigation
and/or recycling as flush water. In this case,
the owner must manage the system to
prevent overflow and, thereby, avoid
violating regulations.
The planner should be familiar with other
regulations regarding natural or jurisdictional
wetlands, odor control, and setback
distances from property lines, neighbors'
houses, wells, streams, roads, public areas,
and other areas that may be governed by
regulation.
Design of SF Wetlands for
Livestock Wastes
It should be noted that design of surface flow
constructed wetlands for livestock waste
treatment is not an exact science. Likewise,
the methods for determining pollutant
reductions in animal waste lagoons or in
predicting wastewater nutrients available for
land application are based only on reasonable
estimates. Thus, the information presented
here provides the best available technology
based on research findings and allows the
designer to appropriately size a wetland to
meet, within reasonable limits of expectation,
the treatment goals.
D Determining wetted surface area
1. Methods available for sizing: In 1991, the
USDA-NRCSlformerly the SCS) published
its Technical Requirements on constructed
wetlands for treating agricultural wastes
(USDA-NRCS, 1991). This document,
entitled "Constructed Wetlands for
Agricultural Wastewater Treatment"
(CWAWT) was based on state-of-the-art
information at that time and has been the
principal design document for livestock
constructed wetlands since then.
It should be noted that the
design of constructed wetlands
for livestock waste treatment is
not an exact science.
The CWAWT document provides two
methods for design: the Presumptive Method
and the Field Test Method. The Presumptive
Method uses information developed by the
Tennessee Valley Authority (USDA-NRCS,
1991) and the Held Test Method uses
equations developed by Reed et aL(1988).
More recently, Kadlec and Knight (1996)
have developed a slightly different equation
that can be used to size livestock constructed
1-25
-------
wetlands; this is also a field test method,
meaning samples must be collected from the
pretreatment unit for use in the equations.
Each of these methods wffll be described and
compared.
The methods presented by NRCS have as
their goal treatment levels that meet or
exceed NPDES discharge requirements.
When NRCS initially prepared their
Technical Requirement on constructed
wetlands, they did so knowing that little
information was available on this type system
for animal wastes. Thus, they set treatment
goals related to discharge limits for BODS,
TSS and NH4-N to standardize design
procedures and to allow the agency to have a
basis for comparing the results from one
system to another. The establishment of
concentrations at or below the typical
NPDES discharge limits was not intended to
promote the discharge of wastewater but,
rather, to serve as a benchmark and to
promote consistency in design throughout
the country.
The NRCS guidance did indicate that
effluent couM be discharged only if
appropriate federal, state, and local permit
requirements were satisfied. Otherwise, the
wetland effluent must be collected in a
storage pond and held until it could be land
applied or recycled. No thought was given at
that time to detenmning the total nutrient
load desired at the final land application site,
then sizing the wetland so that nutrient levels
could be reduced just enough to meet the
needs on the land (nutrient matching). Only
after a number of systems were installed and
data gathered did it become apparent that
design could be based on nutrient needs at
the land application site and not necessarily
on discharge limits (Payne et al, 1996).
Using the wetland for nutrient matching is
discussed in the following sections.
a. Presumptive Method: This method
represents one of the original NRCS.design
approaches. It is used where data on the
wastewater characteristics of the lagoon or
other pretreatment unit are not available. In
this case, the designer makes estimates
(presumptions) about the amount of BOD5
or nutrients produced by the animals and the
amounts lost in the selected pretreatment
unit. Information is derived from tables such
as those presented in the NRCS AWMFH
(USDA-NRCS, 1992). The wetland is then
sized on the basis of 65 Ib BOD/ac/d (73 kg
BOD/ha/d). The goal in this method is to
reduce BOD5 concentrations to less than
30 mg/L, the anticipated allowable discharge
concentration. As noted, the NRCS was not
proposing discharges from the wetland but
was using current design technology during
the initial trials on animal waste constructed
wetlands. That technology, based on
municipal systems, sought treatment levels
below the allowable thresholds for discharge.
If this approach were used to design a
wetland for treating the lagoon discharge
from a 2,000 head swine facility, the designer
might use the following data:
BOD Produced: 2.08 Ib BOD/d/1000 Ib of
animal (Appendix C)
Avg. animal wgt: 180 Ib/hd for finishers
% BOD remaining after treatment: 25%
(recommended in original requirements for a
warm climate; USDA-NRCS, 1991)
1-26
-------
Areal loading for wetland sizing: 65 Ib/ac/d
The wetland size required to meet or exceed
the 30 mg/L wetland discharge guidance
would be calculated as follows:
(1) BOD available after treatment:
(2,000 hd.)(180 lb/hd)(2.08 Ib. BOD/1,000 lb)(0.25)
= 187 lb BOD/d
(2) Wetland size:
(187 lb BOD/d)/(65 lb BOD/ac/d) = 2.9 ac
After the initial sizing, it is necessary,
according to the original guidelines, to check
the hydraulic residence time in the wetland to
ensure that it is at least 12 days. The
equation is follows:
td = (SA)(d)(p)/Q
where—
td = hydraulic retention time (days)
S A = surface area of the wetland (ft2)
d = avg. water depth in the wetland (ft)
p = porosity, a figure which accounts for the
volume not occupied by the plants (i.e., 0.9
for cattails).
Q = flow rate (ftVday)
It has since been found that hydraulic
retention time (HRT) is a function of decay
rate constants for specific pollutants. In other
words, the rate constants for phosphorus,
BOD, and total nitrogen would be different
and, hence, the detention times to meet
specific discharge limits for these
constituents would be different. Thus, the all
purpose 12-day value used earlier is not valid
based on new information.
In addition, a recent review of actual data
from livestock wastewater constructed
wetlands (CH2M Hill and Payne
Engineering, 1997) indicates that the
presumptive method is overly optimistic in
predicting outflow concentrations for many
systems. Based on the review of published
treatment wetland data from a variety of
livestock wastewater management systems
throughout North America, at a presumed
loading of 65 lb BOD/ac/d, the average
effluent BOD concentration (that expected
to be achieved 50 percent of the time) is
about 70 mg/L. A loading of less than about
9 lb BOD/ac/d (10.1 kg/ha/d) would be
necessary to meet the 30 mg/L goal about 80
percent of the time.
The work of McCaskey and Hannah (Section
n, Case Histories) confirm this finding.
Based on more than five years of research on
a constructed wetland for swine lagoon
effluent, they suggest that the loading rates
required by the presumptive method are
greatly overstated. They contend that a BOD
loading rate of 5.9 Ib/ac/d is needed to meet
effluent discharge requirements most of the
time; however, in the winter months the
regulatory limits could not be met even at
these low loading rates.
Table 4 provides expected outflow concen-
trations for certain pollutants based on
estimated hydraulc loading rates (HLR).
This table has been presented by CH2M Hill
1-27
-------
(1997) and reflects an analysis of data from
actual livestock treatment wetlands (Knight
et al, 1996).
Table 5 summarizes treatment performance
for the various constructed wetlands
reported in the LWDB (Knight et al., 1996),
which is based on wetland systems for dairy,
beef cattle, swine, poultry, and aquacultural
sites located throughout North America.
Only the major constituents are reported in
this table.
Most of the systems identified in the LWDB
(Knight et aL, 1996) were designed using the
presumptive method. It has been found that,
in some cases, where low outflow
concentrations were regularly obtained,
actual loading rates were much less than the
65 Ib BOD/ac/d recommended in the original
Table 4. Estimated pollutant loadings to
achieve desired outflow concentrations 80
percent of the time.
Pollutant
BOD5
TSS
TN
TP
Desired Wetland Outflow
Concentration (mg/L)
20
30
50
100
200
Estimated pollutant loadings (Ib/ac/d)
5.5
6.0
3.0
2.5
9.0
11.0
6.0
6.0
18
27
10
10
39
71
26
26
80
180
53
53
presumptive design method. Based on this
information and on the data shown in Tables
4 and 5, it is evident that the original
assumptions used in the presumptive method
need to be modified in order to achieve the
desired outflow concentrations.
Table 5 . Summary of average performance
of animal wastewater treatment wetlands.*
Wastewater
Constituent
BODS
TSS
NB3+NH4-N
TN
IP
Inflow
Cone.
(mg/L)
263
585
122
254
24
Outflow
Cone.
(mg/L)
93
273
64
148
14
Avg.
Reduction
(%)
65
53
48
42
42
*Source: CH2M EHH (1997), as summarized from
Knight etal. (1996).
For the reasons indicated above, it is recom-
mended that the design guidance for the
original Presumptive Method not be used. It
should be noted that the original NRCS
guidance document (USDA, 1991) cautioned
that neither of the methods which they
presented had been thoroughly evaluated for
animal wastes systems over an extended
period of time and at a variety of locations. It
stated that "these methods are considered
state of the art and will likely be modified
and refined as additional systems are installed
and monitored as part of the demonstrations
associated with these technical
requirements."
Now, indeed, additional information is
available to allow modification of the original
design guidance, both for the Presumptive
Method and for the NRCS Field Test
Method. In this regard, a Modified
1-28
-------
Presumptive Method is provided in the
following section based on more recently
acquired data.
b. Modified Presumptive (MP) Method
Unlike the original presumptive method
which uses hydraulic retention time (HRT)
as a check, the Modified Presumptive
Method takes into account pollutant mass
loading or volume of water applied and
rektes the results to a data table derived
from an analysis of information gathered
from actual animal waste constructed
wetlands.
Table 6. Estimated pollutant loadings to
achieve desired wetland outflow concen-
trations su
Pollutants
BOD5
TSS
TN
TP
jercent 01 tne time.
Desired Wetland Outflow Cone.
(mg/L)
20
30
50
100
Estimated Pollutant Loadings (Ib/ac/d)
18
13
6.7
10
29
27
9.1
16
45
54
24
25
89
120
54
50
If a discharge is anticipated, the more
stringent loading rates in Table 4 should be
used. However, due consideration should be
given to the fact that discharge limits could
be exceeded at certain times.
If the system will be designed for nutrient
matching (see The value of constructed
wetlands), loading rates to achieve desired
outflow concentrations 50 percent of the
time are recommended. Table 6, based on
data from numerous livestock constructed
wetlands, provides the estimated loading
rates required to achieve this level of
treatment for several constituents. Thus, at a
given loading rate, the effluent concentration
shown in the table will represent the average
expected on an annual basis. Approximately
half the concentrations will be above the
listed value and half will be below. For the
nutrient matching approach, where treated
wastewater is collected and land applied,
some variation in nutrient application rates is
acceptable for most cropping systems; in
fact, it is normal
Based on this information, the Modified
Presumptive Method is presented here in the
form of an example. It is assumed that the
wetland effluent will be collected and then
land applied. In this case, Table 6 will be
essential for wetland sizing.
Given:
2,000 finishing swine
Average weight: 180 Ib.
Pretreatment of wastewater in an anaerobic
lagoon:
- surface dimensions of 400 x 400 ft.
- 20 percent of original as-excreted N
is available in lagoon effluent
- No surface runoff into the lagoon
Annual precipitation: 45 in
Annual evaporation: 38 in
Crop N required: 150 Ib N/ac/yr
Land available at application site: 40 ac
Wetland effluent stored in downstream
storage pond for 60 days between
irrigations.
Wastewater recycled during winter months.
1-29
-------
Determine size of constructed wetland to
match nutrient requirements on 40 acres:
Step 1: Determine total N per day and per
year after treatment losses:
a. As-excreted N:
(No. Anlmals)(Avg. Wt.)(N /d/1000 Ib. animal)
» IbN/day
( 2,000 finishers) (180 Ib/hd) ( 0.42 Ib N/d/1,000 Ibs )
« 151LbsN/day
b. Daily N in lagoon effluent after losses:
(151 Ib. N/day) ( 0.2 ) = 30.2 Ib. N/day in lagoon
effluent
c. Multiply answer from Step Ib by 365 to
find average annual N:
30.2 lbs/N/day x 365 days/yr = 1 1,023 Ibs N/yr
Step 2: Determine land area required if
pumping directly from lagoon:
4 _ N/yr available aftertrmt. [Stepl]
crop N/ac/yr rqd.
11,
150 IbN/ac/yr
Thus, the land available (40 ac) is less than
that required (73.5).
Step 3: Determine average N required per
year on the available acres:
(Acres available) (application rate)= annual N rqd.
(40 ac) (150 Ib/ac/yr) = 6,000 Ib N/yr required (net)
Step 4: Determine desired average daily N
required in the wetland effluent to satisfy the
irrigated crop N required:
NOTE: If wetland effluent will be stored for
more than 45 days between irrigations,
assume that 10 percent of the N will be lost.
Therefore, increase the net N required (Step
3) by 10 percent to determine the gross
(before storage losses) N desired in the
wetland effluent:
(N/yr. Step3)(loss adjustment) = (m rf.
365d/yr) * *
(6,000X1.1) =
(365)
Step 5: Determine volume of water
produced per day and per year
a. Flush water:
(Number of animals) (Flush volume/head [app. C])
= daily flush volume
(2,000 hd.)(15 gal/hd/day) = 30,000 gpd
b. Volume of animal waste (displacement of
lagoon water):
(No.nd)(mib/nd)(7ASgal/cf)
(l,QQOlbanimalwgt)
=(2,000 head)( 180 Ibs) (7.48 gal/cf)(1.0 cf/1,000 Ib)
= 2,693 gpd
1-30
-------
c. Precipitation minus evaporation on kgoon
surface:
(L)( w) (orecip. -evgp.M 7,48 sal/cf)
(12 in/ft)( 365 d/yr))
= rain volume (gal/d)
(400 ffi(400 ft)(45 - 38 in/vr)(7.48 sal/cf}
(12 infft)(365 d/yr)
= 1,913 gaUdavg.
d. Total average volume of water per day:
Items a + b + c = total gal/day
(30,000 gal) + (2,693 gal) + (1,913 gal)
- 34,606 gal/day average
Step 6: Apply average daily N in the wetland
effluent (Ib/day) in Step 4 to average daily
water volume (gal/day) in Step 5d, together
with a conversion factor, to find desired
average N concentration for the constructed
wetland effluent.
(18,1 lb. N/d)( 119.404)
(34,606)
(Avg,daifyN)(ll9,904)
-. s , .. •• — ,' =
(avg.dauywatervol.)
, ,,,
. (mgIL)
= 62.7 mgIL N (avg. daily)
Step 7: Apply result of Step 6 to Table 6 to
determine the estimated pollutant loading
rate to the constructed wetland.
Note that 62.7 mg/L is between 50 and 100
mg/L for the Desired Wetland Outflow
Concentration in the table; consequently, the
estimated pollutant loading rate will be
between 54 and 24 (Ib/ac/d). Extrapolate
between 50 and 100 as follows:
(62J 50) (54-24) =3l,6lb/ac/d
(100-50)
Step 8: Divide the average daily kgoon
effluent N (Step lb) by the estimated loading
rate (Step 7) to find the estimated area of the
constructed wetland:
Surfacearea of wetland =
(30.
(Step?)
(3l.6lbN/ac/d)
b. Field Test Method #1: The original
equations presented by Reed et al. (1988)
and used by NRCS allowed the user to first
solve for td and then insert this value into the
equation for determining the wetland surface
area. This method was state of the art at the
time NRCS adopted it for use in designing
agricultural waste treatment wetlands.
The area equation has been updated (Reed et
al., 1995), and only this new equation is
presented here. Thus, the equation for area
of the wetland is:
(KT)(d)(n)
where—
A = surface area of the constructed
wetland (m2)
Q = average flow through the
1-31
-------
wetland (nrVd)
Cj = influent concentration (mg/L)
C0 = effluent concentration (mg/L)
kj.= temperature dependent, first
order rate constant (d'1)
= k^S7"20, rate constant adjusted for
temperatures other than 20°C.
(km « 0.2187 @ 20° C; 6 = 1.048)
d = design depth of water in the
system (m); typically 0.1 - 0.46
depending on season and water
quality expectations
n = "porosity" of the wetland (0.65
0.75)
The porosity factor in this equation accounts
for the space occupied by plant stems and
litter within the water column. According to
Reed et aL, 25 to 35 percent of the water
column is filled by plant stems and litter.
Watson and Hobson (1989) reported fill
rates of 10% for cattails (Typha), 14% for
bulrush (Sdrpus validus), 2% for reeds'
(Phmgmites), 6% for woolgrass
(S. cyperinus), and 5% for rushes (Juncus).
Rogers (1995), based on field measurements,
reported rates of 10% for Sagittaria
lancifolia and 7% for Phmgmites austmlis.
However, it is necessary to use the values
prescribed by the originators of the
Method#l equation in order to achieve the
proper results for that model
c. Field Test Method #2: The equation for
this approach is as follows:
The value for Q is the average of inflow
and outflow (QJ. Since the total evaporative
loss for the wetland cannot be determined
until the wetland surface area is known, Q
cannot be initially known. The authors of
Method #1 suggest letting Qj = Qe for initial
design, but final system design should be
adjusted to account for all losses.
= ~(QlkT)la
(Ce -
where-
A = area of the constructed wetland
(m2)
Q = annual flow (mVyr)
kT= k209T"20, rate constant adjusted
for temperatures other than 20° C
(m/yr)
k20 = 14 for TN and 10 for NH3-N
(m/yr)
9 = 1.06 for TN and 1.05 for NH3
(dimensionless).
Q= inflow concentration (mg/L)
Ce= outflow concentration (mg/L)
C* = background concentration
(mg/L), assumed to be 3 for ammonia
andlOforTKN.
The equation was described initially for
treatment wetlands by Kadlec and Knight
(1996), and rate constants specific to
concentrated animal wastes were
summarized by CH2M Hill and Payne
Engineering (1997). Several rate constants,
recommended for use in the k-C* model of
Method #2, are shown in Table 7.
Water depth and hydraulic retention time are
not factored into the equation, but it is
assumed that water depth is sufficient to
cover all roots and rhizomes and that HRT
will be sufficient to provide the necessary
contact time for biological degradation of
pollutants. (See The role of emergent
1-32
-------
Table 7. Parameter values recommended for
use in the k-C* model for sizing animal waste
treatment wetlands*
Parameter
BOD5
TSS
NH3 + NH4-N
Total N
Total P
kM
(m/yr)
22
21
10
14
8
C'
(mg/L)
8
20
3
10
2
6
1.03
1.01
1.05
1.06
1.05
These values are preliminary and maybe revised as
additional data analyses are completed.
herbaceous plants in the treatment process
under Vegetation above.) Data analysis from
non-agricultural treatment wetlands indicates
that increasing water depth does not result in
a proportional increase of treatment
performance (Kadlec and Knight, 1996).
HRT does affect contact time between
wastewater and bacterial communities and
HRT does, in fact, have an influence on
treatment efficiency. A study at the Sand
Mountain project illustrated that as flow rate
was increased in one set of cells, effluent
concentrations of most pollutants more than
doubled (Payne et al., 1992). In other cells
flow rate was reduced and treatment
efficiency increased. In this study, contact
time was increased or decreased by simply
increasing or decreasing inflow rate into the
cells and not by increasing or decreasing
water depth.
2. Comparing methods:
Where actual data on pollutant concen-
trations are not available for a given site and
an estimate of wetland size is needed, the
planner may use the Modified Presumptive
Method. However, it should be understood
that the estimates or "presumptions" that are
made with this method can result in
considerable variation in the overall size.
In the example shown under the Modified
Presumptive (MP) Method, two assumptions
were made: (1) that the average animal
weight was 180 Ib and (2) that the fraction
of N remaining after lagoon treatment was
20%. If another designer chose 150 Ib as
the average weight and assumed only 10% of
N remaining, the wetland size would be
reduced from 1.0 to 0.3 ac. Table 8
illustrates the differences in wetland sizes
using 150 and 180 Ib average weights and
10 and 20 percent remaining N.
Table 8. Wetland sizes using the MP Method
for 2,000 top hogs with different estimated
average weights and N remaining after pre-
treatment
Avg. Wgt.
150
150
180
180
%N
remaining
10
20
10
20
Wetland area
(ac)
0.3
0.8
0.5
1.0
(m2)
1,215
3,239
2,024
1,048
It might be noted that the value selected for
N remaining could be a function of climate
just as BOD remaining is assumed to be
(USDA-NRCS, 1991). In other words, only
10 percent of the original N might be
remaining in warm climates, whereas 20
percent could be available in cooler climates.
The planner is advised to use the modified
presumptive approach with caution. The final
result should be a good approximation for
1-33
-------
planning purposes. However, the use of this
approach, like the one that preceded it,
should be evaluated over time and adjusted
as needed.
One of the Field Test Methods should
ultimately be used wherever possible. If the
animal confinement facility is not yet
installed, this may mean collecting samples
from the pretreatment unit of an identically
operated facility with a nearly identical
pretreatment unit and using that information
in the field test equations for the new facility.
Another option would to postpone construc-
tion of the wetland until samples can be
collected from the pretreatment unit of the
system for which the wetland will be used.
A comparison of the two field test methods
was made, using an example involving a
2,000-head swine finishing or top hog
operation. It is assumed that samples were
collected from an existing facility to help
design a waste managment system at a new
site. The would-be producer finds that he
has too little land for direct irrigation based
on information from the existing facility. If
the wastewater in the new lagoon were
pumped directly to the land, 73.6 ac (29.8
ha) would be needed to satisfy the nitrogen
requirement of a particular crop; however,
only 40 ac (16.2 ha) is available.
Using the nutrient matching approach, the
designer seeks to use a constructed wetland
to reduce the nitrogen level to 54 percent of
the original amount (40 / 73.6) so the
wastewater can be applied to the 40 available
acres. Based on data from the existing
lagoon, the nitrogen concentration in the
supernatant of the new lagoon is expected to
be 200 mg/L. This must be reduced to 108
mg/L or 54% of the original amount. Daily
flow rate (Q) for Field Test Method f 1 is
130 m3/d. Q for Method #2 is an annual
amount and is 47,450 m3/yr or 365 times the
daily average value used in Method #1.
The lagoon supernatant contains no NO3 and
ammonia constitutes 80 percent of TKN.
Table 9 illustrates the differences in wetland
size for given conditions of temperature (T),
porosity (n), and water depth (d). All metric
units have been used in this example.
Although Reed et al. (Method #1) suggest
using porosity values of 0.65 to 0.75, other
recent studies, noted above, indicate that
porosities in the range of 0.85 to 0.95 are
appropriate for the most popular plants that
would be used in wetlands for animal waste
treatment. However, in order for the
equation to provide the results achieved by
the model for certain municipal situations,
the developer's porosity value of 0.75 was
used in the comparison. Three values for
water depth have been used: 8 in (0.203 m),
10 in (0.2554 m), and 15 in (0.381 m).
It is noted that as water depth increases in
Method #1, wetland surface area decreases
and, in turn, the sites for microbial activity
also decrease. If Method #lcontinues to be
used for design, it would appear best to
initially minimize average depth in order to
maximize surface area and the sites where
most microbial activity will occur. That is,
water depth should be just high enough to
cover roots, rhizomes and pknt litter.
1-34
-------
Table 9. Comparison of constructed wetland surface areas using Method #1 and Method #2 for
ammonia concentrations of Q = 200 mg/L and C0 = 108 mg/L.
T(°C)
10
15
20
25
30
Area (m2); Method #1 for n = 0.75
water depth (dav)
0.203m
3840
3041
2402
1906
1503
0.245m
3182
2520
1991
1579
1246
0.381m
2046
1620
1280
1016
801
Area (m2)
Method #2
5150
4028
3141
2454
1927
It would appear that treatment performance
could be improved by simply raising the
water level and, thereby, increasing the time
of contact between the wastewater and the
microorganisms. While some increase in
efficiency may be obtained in this manner,
current data on animal waste treatment-
wetlands indicate that increasing water depth
will not result in a proportional increase in
treatment. In addition, data on municipal
systems (Kadlec and Knight, 1996) suggest
that the rate constants of Method #1 would
have to be reduced as water depth is changed
in order to get accurate results for animal
waste systems.
The general equations for both Methods #1
and #2 were developed for municipal
systems. However, the rate constants (kT and
k20) and theta values (0) applicable to
Method #2 were derived from the current
livestock wetlands database (CHjM Hill and
Payne Engineering, 1997). This method
assumes that treatment performance is
directly proportional to wetland surface area
and that increasing depth does not provide a
proportional increase to performance. Thus,
sizing for ammonia reductions based on
Method #2 provides a larger surface area
than Method #1 for all temperatures
evaluated as shown in Table 9.
3. Selecting a method (recommendations):
The planner should fully understand that
current wetland design criteria for animal
waste systems will provide a reasonable
prediction of treatment performance but that
predicted values may still be somewhat
different than actual values. If a wastewater
discharge is being considered, the most
conservative approach to design should be
used. If wastewater will be land applied after
treatment, the concern over reaching
presribed treatment levels all of the time is
much less critical With these thoughts in
mind, the following recommendations for
designing animal waste constructed wetlands
are provided:
(1) If adequate information on average daily
flow rates and effluent concentrations of
selected constituents from the pretreatment
unit are not available for the system, the
1-35
-------
Modified Presumptive Method can be used
to estimate wetland size. However, due
consideration should be given to the cautions
noted above when using any presumptive
method. If a discharge is anticipated, a field
test method should be used for final design.
(2) When using a field test method, it is
recommended that Method #2 be used
because:
(a) it is based on current livestock; data, and
(b) it provides a slightly larger surface area,
which adds a measure of safety
(3) When sizing for nutrient matching of
nitrogen, data for ammonia nitrogen should
be used rather than total nitrogen. Ammonia
nitrogen typically represents 70 to 80 percent
of the total nitrogen for most animal waste
pretreatment units. Some of the TN in the
wetland effluent will be lost during storage
and some during land application. Therefore,
the organic fraction (20 to 30 percent of
TN), which will convert to ammonia, is
expected to approximately balance that
which is lost during downstream storage and
application. In addition, using the given
values of k and 8 for TN without considering
conversions will result in a smaller wetland.
(4) If the wetland will have an approved
discharge, it is recommended that the size be
increased by 20 percent beyond that
recommended by the equation. It should be
noted that there will be variability around the
predicted outflow concentrations, regardless
of which equation is used. In other words,
the equations provide sizes that will ensure
treatment to at least the predicted
concentrations about 50 percent of the time;
consequently, increasing the size will add a
margin of safety.
(5) Dormant season storage of wastewater
and subsequent land application based on
nutrient matching is recommended over
discharge. This approach will allow for the
use of average warm season temperatures in
design, which will, in turn, reduce the size of
the wetland. It will also prevent the problems
associated with the permitting process and
with monitoring.
(6) If a discharge is being considered, it is
recommended that the designer refer to the
more detailed information provided by the
sources mentioned in this publication.
n Bottom slope/maximum length
The earlier design guidance by NRCS
indicated that a slope in the lengthwise
direction of the wetland would facilitate
drainage in case repairs or maintenance were
needed. Indeed, a slope can be built into the
wetland to accomplish this purpose;
however, due consideration should be given
to the rapid increase in depth that will occur
if the slope is even as flat as 0.5 percent, the
maximum value first recommended by
NRCS. For instance, if the initial water depth
at the upper end of a wetland cell is expected
to be 6 inches (15.2 cm), the water depth at
100 ft (30.5 m) from the inlet would be 12 in
(30.5 cm), and at 150 ft (45.7 m) it would be
15 inches (38 cm).
An acceptable alternative to this approach is
to have a flat bottom with a deep zone
across the downstream end. If the system
must be drained, the water can be pumped
from the deep zones to the land, to the
downstream holding pond, or to the
upstream pretreatment unit. The depth of the
1-36
-------
deep zones should be at least 3 ft (0.9 m)
deeper than the marsh. (See Figure 4.)
If a sloping wetland is used, either by choice
or by necessity resulting from site conditions,
the maximum length is dependent on the
allowable water depth associated with the
wetland plants involved. If a level-bottom
wetland is used, maximum length is not
important for most animal waste treatment
wetlands. If however, an exceptionally long,
level-bottom wetland is planned,
intermediate deep zones should be used, not
only to facilitate drainage, but also to allow
effective redistribution of flow.
D Hydraulic retention time
The original NRCS guidelines required a
minimum HRT of 12 days for livestock
constructed wetlands. This was thought to
be the time necessary to achieve the desired
treatment levels for BOD5 and ammonia.
Given the additional new data on animal
waste wetlands, it has been found that HRT
values needed to achieve recommended
discharge limits could be higher or lower
than 12 days depending on the constituent
involved.
However, it should be noted that accurately
estimating HRT is not a simple matter. In
this process it is necessary to determine the
as-built conditions in terms of bottom
elevations, slopes of bottom and sides,
width, and length. The designer must also
know the average water depth, the average
flow rate, and the volume of the water
column not occupied by plants (porosity).
Considerable error can be introduced into
any estimation of HRT by changes in as-built
conditions due to soil swelling, erosion of
embankments and sedimentation; the
uncertainty of actual porosity (see former
discussion under Field Test Method #1);
possible short circuiting due to uneven plant
development; the shallow depths involved;
and fluctuations in inflow rates.
There is no doubt that HRT affects treatment
performance. Therefore, if a discharge is
planned, the designer should still develop an
estimate of HRT, giving due consideration to
the possible inaccuracies that may be
involved. This may mean using conservative
values in making the estimates.
After the system has been installed, the HRT
should be checked using lithium or another
inert tracer to more accurately evaluate HRT
and to allow for adjustments in flow.
D Hydraulic loading rate
The hydraulic loading rate (HLR) is defined
as the inlet wastewater flow divided by the
wetland area, excluding embankments and
islands. It is typically reported in units of in/d
or ft/yr (cm/d or m/yr). HLR does not imply
that the water is physically distributed
uniformly over the surface area of the
treatment wetland. HLR is generally easier to
estimate than HRT, and it correlates more
closely with treatment performance because
of the reliance of most wetland processes on
surface area rather than on water depth.
CH2M Hill and Payne Engineering (1997)
reported an average HLR of 1.85 in/d for
treatment wetlands in the Livestock
Wastewater Treatment Wetland Database.
Table 5 summarizes the average constituent
outflow concentrations and reduction
efficiencies for those systems. Lower HLRs
are necessary to achieve lower pollutant
1-37
-------
Natural ground
4a. Steep iloping land: several cells needed to maintain level bottom
/Natural ground
Cell #2
4b. Moderately sloping land: fewer cells needed for level bottom
Deep zone (<3 ft):
pump to drain
Deep zone to
redistribute flow
4c. Celli with level bottoms and deep zone*
Figure 4: Effects of Topography and the Use of Deep Zones
1-38
-------
outflow concentrations from animal waste
treatment wetlands.
D Layout of the wetland
The layout of the system is' sometimes dic-
tated by site conditions. Shape of the site,
area available, and lay of the land could be
major constraints.
SF wetlands are often multi-eel systems.
Cells will typically be arranged in series,
depending on topography, and in parallel
(side-by-side). The parallel arrangement
allows two or more cells to receive effluent
at the same time; thus, if the inlet structure in
one cell plugs, the other cell(s) will keep
operating. In addition, the parallel arrange-
ment allows one set of cells to be closed for
maintenance while the others remain
operational The owner can also use the
parallel arrangement to rotate discharge
points or to use different treatment
strategies. (See Figure 5.)
An efficiently designed system will have
limited short-circuiting of wastewater
between inlets and outlets. In such a system,
the waste flow will have continuous contact
with all submerged surfaces most of the time.
A large, square single-celled system with one
inlet and one outlet would have dead areas in
the corners unless flow is evenly distributed
across the upper end. There will, of course,
be some dead zones in nearly all systems
caused by islands of roots, rhizomes, and
dead vegetation. However, the goal is to
provide flow across the entire width of each
cell as practically as possible.
Barring any significant site constraints, the
length-to-width ratio of the overall wetted
area of the system should be in the range of
1:1 to 4:1. Ratios of individual cells have
been as high as 20:1. Early designers of
wetlands encouraged a ratio of 10:1 to
ensure plug flow through the system. It was
found, however, that the longer the cells, the
greater the resistance to flow due to
vegetation, especially at densely vegetated
sites. At one municipal site having an aspect
ratio of 20:1, the flow was so restricted that
wastewater overflowed the embankment at
the inlet end of the system (Reed et at,
1995).
Proper L:W ratios may help prevent short-
circuiting of flows. Short-circuiting can also
be nodnimized by initially distributing flow
across the entire width of the first eel and
subsequently redistributing the flow. Flow
redistribution can be accomplished by adding
cells in series and discharging the wastewater
into each new cell through a distribution
header pipe, or with an inlet deep zone
across the width of the cell Adding cells in
series is a practice that may be necessary
whenever a cell has a bottom slope for
drainage or where the site itself is sloping.
Flow can be redistributed by constructing
deep, narrow channels across the direction of
flow. These deep zones can be placed at
mid-length or more often for very long cells.
Channels should be at least one meter deeper
than the constructed bottom of the wetland
cell to inhibit the growth of rooted
vegetation (Kadlec and Knight, 1996).
D Embankments
1. Design height: Although embankments at
most livestock wetland sites are usually all
the same height, a distinction can be made
between the outer embankments and those
which divide the system into cells. The outer
1-39
-------
Cross-sectional and plan views of Lagoon/Wetland/Pond
system with recycling and irrigation options
i
Set pipe outlet elev. at bottom
x-elev. for winter stg
Animal
housing
Pump for recycle
and irrigation
Lagoon
Constructed
Wetlands
Pond
1st Stage Cells 2nd Stage
Irrig
line
Recycle line
Figure 5: Typical layout of a lagoon / wetland / storage pond system for waste treatment
-------
embankments must be high enough to
protect the system from overtopping during
a specific design storm (i.e.5 25-yr, 24-hr).
These embankments must have an emer-
gency bypass set at such an elevation that
discharge will only occur when the design
storm has been exceeded.
Design height for the outer embankments
should be based on the following increments
of depth:
a. Normal design flow: Based on type
of vegetation; typically 8 to 12
inches (20 - 30 cm).
b. Accretion: Based on the design life
of the system; alow 1.0 in (2.5
cm) per year.
c. Design storm: Includes direct
precipitation on the wetland plus
runoff from embankments and, if
inflow is unrestricted, precipitation
on the pretreatment surface.
d. Ice cover: If the system will
operate in winter, allow depth
--~" equal to the ice thickness expected
during some design period (i.e.,
once in 25 years).
e. Freeboard: A safety factor of at
least 12 inches (30 cm) is recom-
mended.
f. Emergency bypass: As required by
the type of bypass.
Design height for the divider embankments
must include at least items a, b, and e, above.
2. Width of embankments: Outer embank-
ments should be at least 15 ft (4.6 m) at the
top to prevent burrowing animals from
draining the system to the surrounding area.
Inside slopes should be 2 horizontal to 1
vertical if this slope is part of an interior cell.
Outside slopes should be no steeper than 2:1
but will usually be shaped to fit the site.
Inside embankments should be wide enough
at the top for easy maintenance. Top widths
of 8 to 10 ft (2.4 - 3 m) are recommended so
grass can be mowed with tractor-driven
equipment and to reduce the potential for
animals burrowing through dikes. Narrower
dikes or embankments have been used, but
these must be cut with hand mowers, and
they are easily breached by muskrats.
D Liners
The bottoms of all cells and insides of the
outer embankments should be appropriately
lined to prevent seepage. The discussion
under Site Evaluation, L Soils should be
reviewed to determine when compacted soil
liners or manufactured liners should be used.
A properly sized orifice can control inflow
rate, but an upstream filter is needed to
prevent plugging.
D Inlet / Outlet Structures
1. Inlet structures: A variety of inlet control
structures have been used at livestock
constructed wetland sites. These include
simple overflow pipes with unregulated flow
between the pretreatment unit and the upper
wetland cells; pipes with orifice controls; and
valves. Some of these discharge directly at
the middle of the cell Others have a gated or
1-41
-------
slotted pipe which spans the width of the cell
to ensure even distribution of flow upon
entry. Other options for initial distribution
include deep zones across the width of the
inlet end and shallow dams with multiple
slots or weirs across the top.
If wastewater will be stored in the pretreat-
ment unit during winter, the invert elevation
of the effluent pipe leading to the wetland
should be in line with the bottom elevation
for winter storage (see Figure 5).
Some positive control is needed not only to
prevent discharge to the wetland during the
dormant season, but also to ensure a
controlled release throughout the in-use
period based on the water budget.
Plugging of control devices can be a
problem. A buildup of a crystalline substance
called struvite on the walls of piping systems
for animal wastes has been a problem in
some orifice control devices. Other devices
have been plugged with debris from the
pretreatment unit. Inlet screens or box
screens should be used around the inlet pipes
to prevent debris, including turtles, from
entering the inlet line. Inlet structures should
be observed daily for potential problems.
2. Outlet structures. The outlet structure is
used to maintain the proper water level in the
upstream cell and to control outlet flow rate.
Several types of outlet control structures
have been used. These include slotted pipes
laid across the bottom of cells or slotted
pipes buried in a shallow gravel trench at the
downstream end of a cell. These are con-
nected with a T section to a pipe that passes
through the embankment to the downstream
water-level control device. (See Figure 6.)
The water-level control section for this type
outlet is typically an elbow attached with a
swivel joint. The water level in the upstream
cell is controlled by the invert elevation of
the outlet pipe. The pipe can be tilted at the
swivel/elbow connection, allowing the invert
to be raised and lowered, which, in turn, sets
the water elevation upstream. Water can be
discharged directly at the invert (top photo
below) or another section can be added to
form a U, with another elbow and swivel
opposite the first elbow and swivel (lower
photo below). This section is then attached
to a slotted pipe which can distribute the
effluent across the head of the next cell.
This swivel pipe controls the water level in
the upstream cell.
,»•*
Two swivel elbows and a U are used here to
redistribute flow through a header pipe.
Some of the slotted pipe headers have had a
problem with plugging. For this reason other
1-42
-------
flashboard dam
Embankment In
upstream cell
4-
End view of dam
looking upstream
flashboard dam
PLAN VIEW
Figure 6: Flashboard Dam
1-43
-------
types of outlet control structures are being
explored. One approach is to install a deep
zone across the outlet end of the eel with a
flashboard dam set in the embankment to
control upstream water level. Treated boards
used for the dam can be removed or added
as needed to control water level (Fig. 6). In
addtion, various weir and drop box inlet
controls might be considered.
O Water budget
A water budget is needed for design of the
overall system and for management.
Designers who plan many of these systems
shouM consider developing a computer
spreadsheet. A sample spreadsheet for a
constructed wetland/land applcation
treatment system is presented in Table 10.
See also the following section on Operation
and Maintenance.
Operation and Maintenance
1. Operation: Normal annual operation of
the system will be dictated by the water
budget, by visual inspection, by wastewater
testing, and by common sense. Some of the
key operational requirements include:
a. Maintaining water levels in the wetland
cells as appropriate for the vegetation. In
cold climates where continuous winter
operation will be involved, increase water
levels as needed prior to the first freeze.
b. Control flows into the wetland in
accordance with requirements of the water
budget. Adjust as needed for periods of
drought, increasing inflow rates to ensure
that vegetation at the extremities of the
wetland are kept wet during dry times.
c. Ensure that water levels in the pre-
treatment unit and downstream storage pond
are lowered to appropriate levels in
preparation for winter storage.
d. Collect samples and measure flow rates
into and out of the wetland regukrly.
Determine treatment efficiencies and nutrient
mass loadings.
e. Sample wastewater in the downstream
storage pond prior to land application.
Determine fill rates of the pond to determine
total nutrient load available per, year.
f. Revise water budget as needed.
2. Maintenance: Regular maintenance of the
wetland system is essential and inevitable. If
frequent inspections are ignored, rodents can
destroy vegetation and embankments, pipes
can become plugged, wastewater can short
circuit through the cells, and the system can
become inoperational in a short time.
A short list of important maintenance items
is provided below. This is not intended to be
an all-inclusive list:
a. Inspect inlet and outlet structures for
plugging and damage on a daily basis.
b. Inspect embankments at least weekly for
damage and make repairs as needed. Control
rodent pests through removal or deterrents,
such as electric fences.
c. Mow embankments regularly to allow for
inspections and to enhance visual appeal
d. Inspect and repair fences as needed.
e. Inspect vegetation throughout the growing
1-44
-------
Table 10. Sample water balance spreadsheet for 2000 finisher swine with 400x400 ft lagoon and 26,400 ft2 constructed wetland (CW).
Climat
Precip.
Pan
Lake
Jan
5.6
3.2
12
feb
5.4
3.8
2.7
mar
6.0
4.0
2.8
apr
5.9
4.0
2.8
may
4.9
4.3
3.0
Jon
5.0
5.1
3.8
jul
4.5
5.6
3.9
aug
4.0
6.2
4.3
sep
3.8
4.9
3.4
oct
4.2
4.3
3.0
nov
5.0
4.0
2.8
dec
5.2
3.8
2.7
total
59.5
53.2
37.4
Items
Input
Manure
Precip.
lagoon
Precip.
CW
flush*
runoff
Total
Output
lap evap
CWevap
Mgt
(1000ft3)
Volume (1,000 ft?/mo)
Jan
11.2
74.7
12.30
0
4.7
102.9
jan
29.9
4.93
0
feb
10.1
72
11.90
0
4.5
98.5
feb
35.5
5.85
0
mar
11.2
80
1320
0
5.0
109.4
mar
373
6.16
S&llftfSllrtS:
0
apr
10.8
78.7
13.00
1203
4.9
227.7
apr
373
6.16
WiHiiSiiiXi*
210.19
may
11.2
65.3
10.80
1243
4.08
215.68
may
40.1
6.62
^•;-:-;-;":$:-t'?:?$fi%wS
210.19
jun
10.8
66.7
11.00
120.3
4.17
212.97
jun
47.6
7.85
210.19
jul
11.2
60
9.90
124.3
3.75
209.15
j«l
52.2
8.62
aug
11.2
53.3
8.80
124.3
333
200.93
aug
57.8
- 9.55
sep
10.8
50.7
8.40
120.3
3.16
19336
sep
45.7
7.55
oct
11.2
56
9.20
0
3.50
79.9
oct
40.1
6.62
nov
10.8
66.7
11.00
0
4.16
92.66
nov
37.3
6.16
dec
11.2
693
11.40
0
433
9623
dec
35.5
5.85
Met annual w. w. = total Input - total Output =
210.19
210.19
210.19
0
0
0
total
131.70
793.40
130.90
733.80
49.58
1839.38
total
4963
81.92
1261.16
1261.16
*flush= fresh flush water; if recycled wastewater from the CW is used, flush = 0. Hush at 15 gal/hd/d.
-------
season and replace plants that are not
performing as expected.
f. Inspect and repair pumps and piping
systems, if used.
Final Comments
The constructed wetland can be a very useful
tool in managing animal wastes. It is not
suitable for every operation, and, in fact, it
may be undesirable in many locations.
Constructed wetlands must be properly
planned, designed, constructed, and
managed. Failure in any of these areas could
result in failure of the system.
The constructed wetland is still a new
method for treating animal wastes. Much has
been learned, but much more remains to be
learned. As more systems are installed and
more information gathered, the design and
management techniques will be refined
further.
In the proper place and with proper
management, the constructed wetland can be
a valuable asset to the manager seeking an
economical and environmentally sensitive
way to treat animal wastes.
1-46
-------
References
Armstrong, W., J. Armstrong and P. M.
Beckett (1990); Measurement and Modeling
of Oxygen Release from Roots of
Phragmites australis; In Constructed
Wetlands in Water Pollution Control,
Oxford,UK.; Pergamom Press. (Eds. P. F.
Cooper and B. C. Findlater)
Billmore, S. K., P. Dass, and H. Hyas
(1994); Ammonia Volatilization through
Plant Species in Domestic Wastewater
Applied to an Agricultural Field and
Wetland; In Proc. of the 4th Intel. Conf. on
Wetland Systems for Water Pollution
Control; Guangzhou, China.
Brix, H. and H. Schierap (1990); Soil
Oxygenation in Constructed Reed Beds: the
Role of Macrophyte and Soil-Atmosphere
Interface Oxygen Transport; In Proceedings
of the International Conference on the Use
of Constructed Wetlands in Water Pollution
Control; Oxford, UK; Pergamom Press.(Eds.
P. F. Cooper and B. C. Findlater)
CH2M Hill (1997); Constructed Wetlands
and Wastewater Management for Confined
Feeding Operations; prepared for EPA's
Gulf of Mexico Program; published by
CH2M Hill, Gainesville, FL.
CH2M Hill and Payne Engineering (1997);
Constructed Wetlands for Livestock
Wastewater Management: Literature
Review, Database, and Research Synthesis,
prepared for the EPA's Gulf of Mexico
Program; pub. Alabama Soil and Water
Conservation Committee, Montgomery, AL.
Corbit, R. A, (1989); Standard Handbook of
Environmental Engineering; McGraw-Hill,
NY.
Freney, J. R., R. Leuning, J. R. Simpson, O.
T. Denmead, and W. A. Muirhead (1985);
Estimating Ammonia Volatilization from
Flooded Rice Fields by Simplified
Techniques; Soil Sci. Soc. Am. Jour., Vol.
49.
Gersberg, R. M., B. V. S. R. Lyons,
and C. R. Goldman (1985); Role of Aquatic
Plants in Wastewater Treatment by Artificial
Wetlands; Water Res., 20:363-367.
Guntenspergen, G. R., F. Steams and 3. k
Kadlec (1989); Wetland Vegetation; In
Constructed Wetlands for Wastewater
Treatment; ed. D. A. Hammer; Lewis
Publishers, Chelsea, MI
Gunterspergen, G. R., J. R. Keough and J.
Alien (1993); Wetland Systems and their
Response to Management; la Proc. of the
National Animal Waste Constructed
Wetlands Conference, Ft. Worth, TX; Texas
A & M University Press, TX.
Hemond, H. F^(1983); The Nitrogen Budget
of Thoreau's Bog; Ecology. 64(1).
Hughes, W. B., S. R. Kown and V. W. E.
Payne (1996); Economic Assessment of
Animal Waste Management Systems with a
Constructed Wetland Component; In Proc.
of the National Animal Waste Constructed
Wetlands Conference, Ft. Worth, TX; Texas
A & M University Press, TX.
Kadlec, R. H. and R. L. Knight (1996);
Treatment Wetlands, Lewis Publishers, Boca
Raton, FL.
Knight, R. L., V. Payne, R. E. Borer, R. A.
Clarke, Jr., and J. H. Pries (1996a);
Livestock Wastewater Treatment Wetland
1-47
-------
Database; In Proceedings of the 2nd NatL
Wkshp. for Constructed Wetlands for
Animal Waste Management; Texas A & M
University Press, TX.
Knight, R. L. (1996), unpublished data.
Lofgren, R. (1993); Creating a Wetlands
Wildlife Refuge from a Sewage Lagoon; la
Constructed Wetlands for Water Quality
Improvement; ed. G.A.Morshiri; Lewis
Publishers, Boca Raton, FL,
Miller, B. K., P. J. DuBowy and R. P.
Reaves (1996); Getting the Word out to
Producers: Extension Ideas, Potential
Approaches, and Activities; In Proc. of the
National Animal Waste Constructed
Wetlands Conference, Ft. Worth, TX; Texas
A & M University Press, TX.
Payne, V. W. E., (1992) Constructed
Wetland for Treating Swine Lagoon
Effluent; ASAE Paper No. 924526; ASAE,
St. Joseph, ML
Payne, V. W, E., R. L. Knight and S. R.
Kown (1996); A Holistic Approach to the
Design of Constructed Wetlands for Treating
Animal Wastes; In. Proc. of the National
Animal Waste Constructed Wetlands
Conference; Ft. Worth, TX; Texas A & M
University Press, TX
Reed, S. C, E. J. Middlebrooks and R. W.
Crites (1988); Natural Systems for Waste
Management and Treatment, McGraw-Hill,
NY.
Reed, S. C., R. W. Crites and E. J.
Middlebrooks (1995); Natural Systems for
Waste Management and Treatment (2nd
ed.), McGraw-Hill, NY.
Rogers, J. W., D. T. Hill, V. W. E. Payne
and S. R. Kown (1995); A Biological
Treatment Study of Constructed Wetlands
Treating Poultry Wastes; In Versatility of
Wetlands in the Agricultural Landscape,
ASAE and AWRA Proceedings, Tampa, FL.
USDA Natural Resources Conservation
Service, (1991); Constructed Wetlands for
Agricultural Wastewater Treatment,
Technical Requirements; Washington, DC.
USDA Natural Resources Conservation
Service (1993); Design and Construction
Guidelines for Considering Seepage from
Agricultural Waste Storage Ponds and
Treatment Lagoons, South Technical Center
Note 716, Ft. Worth, TX.
USDA Natural Resources Conservation
Service, (1992); Agricultural Waste
Management Field Handbook, Washington,
DC.
USEPA (1995), Guide Manual on NPDES
Regulations for Concentrated Animal
Feeding Operations, EPA 833-B-95-001,
Washington, DC.
USEPA (1976), Quality Criteria for Water,
Washington, DC
Watson, J. T. and J. K. Hobson (1989);
Hydraulic Design Considerations and
Control Structures for Constructed Wetlands
for Wastewater Treatment; In Constructed
Wetlands for Wastewater Treatment, ed.
D. K. Hammer, Lewis Publishers, Chelsea,
ML
1-48
-------
Section II
Case Histories on
Constructed Wetlands
for Treating Animal Wastes
Swine:
Duplin County, NC
Sand Mountain, AL
Dairy:
Kosciusko County, IN
Desoto County, MS
Maiden Valley, Ontario, Canada
Oregon State University, OR
Poultry:
Auburn University, AL
-------
-------
Constructed Wetland to Treat Swine Wastewater
Duplin County, North Carolina
F. Humenik, PhD, M. Rice, M. Cook, PhD, and S. Broome, PhD*
P. Hunt, PhD, and A. Szogi, PhD,** G. Stem and M. Sugg,* and G. SeaIf*H*
Introduction
This project was located at a swine facility
housing 2600 nursery pigs having an average
weight of 11.8 kg. The swine are housed in a
single building which is flushed six times per
day into a 0.24 ha anaerobic lagoon.
Wastewater from the kgoon is irrigated
directly onto nearby pasture and crop land.
The constructed wetland is operated as an
independent, no-discharge system with all
effluent being returned to the kgoon for land
application.
Aerial view of lagoons and wetland
The project was undertaken to address
concerns and to.answer questions about the
ability of wetland systems to (1) produce an
effluent that met discharge limits for nitrogen
and phosphorus and (2) remove high
percentages of nitrogen from wastewater. As
this description shows, these goals were
sometimes met individually but could not be
met at the same time. The performance of
the system depended on the loading rate at
which the effluent was discharged to the
wetland.
Regulatory Context
To discharge the treated effluent to a local
stream, the wetland system had to produce
effluent concentrations with nomore than 4
mg/L total nitrogen in the summer and 8
mg/L in the winter, as well as 2 mg/L total
phosphorus year-round.
Wetland Design
The wetland was designed according to
guidance on nitrogen loading rates. For
municipal wastewater wetlands, the
recommended loading rate was total Kjeldahl
nitrogen (TKN) or ammonia nitrogen
(NH3-N) at 18 kilograms per hectare per day
(kg/ha/day). For livestock wastewater
wetlands, recent guidelines varied from 10 to
15 kg/ha/day. While this system was being
designed, the Tennessee Valley Authority
(TVA) issued new criteria of less than 3
kg/ha/day for wetlands designed to meet
advanced discharge standards. As a result the
system was designed for a low TKN loading
rate of 3 kg/ha/day.
Six 3.6 x 33.5-meter wetland cells were
constructed in 1992 (Figure 1). They were
*North Carolina State University, Raleigh, NC; **Agricultural Research Service, Florence, SC;
4* USDA-Natural Resources Conservation Service, NC;* * Murphy Farms, Rose Hill, NC
-------
divided into three parallel systems of two
cells in series. Wetland system 1 contained
rushes and bulrushes; wetland system 2
contained bur-reed and cattails; and wetland
system 3 contained soybeans in saturated soil
culture and rice. Due to different operational
parameters for wetland system 3, a summary
of results was not available.
The cell bottoms and sidewalk were lined
with clay and then covered with 20 to 30 cm
of loamy sand soil. Lengthwise slopes were
0.2 percent or less. Water depth at the end of
the slope was maintained below 15 cm.
Monitoring
V-notch weirs and PDS-350 ultrasonic
open-channel flow meters were installed at
the inlet and outlet of each of the three
wetland systems. Five ISCO 3700 samplers
were installed; one sampler collected samples
of the wastewater influent and the other four
sampled the water at the end of each single
cell. The water sampler combined hourly
samples into composites. A CR7X data
logger with two multiplexers were installed
for hourly acquisition of flow, weather, and
soil redox potential data.
Operation and Performance
From May 1993 to June 1994, wastewater
was applied to the constructed wetland at a
rate of 3 kg/ha/day of TKN. Lagoon
wastewater was diluted about tenfold with
fresh water to meet the low TKN application
rate and to make up for evapotranspiration
during summer. As a result of the increased
dilution, TKN concentrations in the influent
wastewater were lower in the summer.
Wastewater input to the wetland was
continuous, and flow control valves in a
mixing tank were set to provide the desired
proportion of lagoon liquid and diluted
water. Effluent TKN ranged from about 30
to 50 mg/L total nitrogen for winter.
At the 30 mg/L loading rate, effluent TKN
was generally less than 8 mg/L. At the 50
mg/L loading rate, effluent TKN was
generally more than 10 mg/L. TKN removals
on a mass basis for the 3 kg/ha/day loading
rate were 96 and 91 percent for wetland
systems 1 and 2, respectively (see Table 1).
The effluent sometimes met local stream
discharge requirements of 4 mg/L total
nitrogen for summer and 8 mg/L total
nitrogen for winter.
Table 1. Effluent TKN and TP Concentrations in
Response to Different Mass Loading Rates.
TKN
Loading
(kg/ha/d)
3
10
Effluent TKN
mg/L
<8
10-20
%
Removal
91-96
73
Effluent TP
mg/L
7
10-20
%
Removal
73
10-17
Effluent total phosphorus averaged about 7
mg/L for the TKN loading rate of 3
kg/ha/day. In general, effluent total
phosphorus concentrations exceeded the
discharge allowance of 2 mg/L year-round.
Total phosphorus removal on a mass basis
was about 73 percent.
From June 1994 to January 1996, the TKN
loading rate was increased to 10 kg/ha/day
with the new goal being maximum nitrogen
removal rather than meeting stream
discharge requirements. After increasing the
TKN loading rate, effluent TKN
concentrations generally exceeded local
stream discharge requirements. However, at
the higher loading rate, both wetland systems
stiU removed more than 73 percent of TKN
II-2
-------
on a mass basis.
At the higher TKN loading rate, effluent
total phosphorus ranged from 10 to 20 mg/L
Only 10 percent and 17 percent of the total
phosphorus was removed by wetland
systems 1 and 2, respectively. Total
phosphorus removals on a mass basis
decreased significantly with time and higher
TKN loading rates.
Effluent total organic carbon concentrations
varied widely for both TKN loading rates.
The wetland systems did not appear to affect
total organic carbon concentrations (TOC)
and, in some cases, TOC increased.
Conclusions
At the loading rate of 3 kg/ha/day of TKN,
the wetland discharge met nitrogen criteria
during some time periods. The discharge did
not meet phosphorus criteria, except
temporarily in wetland system 2 before TKN
loading was increased.
At the loading rate of 10 kg/ha/day of TKN,
the wetland discharge often exceeded the
nitrogen criteria and consistently exceeded
the phosphorus criteria. However, the
wetland did meet the secondary goal of high
nitrogen removal with removal efficiencies of
about 73 percent.
While wetlands can significantly reduce
nitrogen mass loading, they do not appear to
be a viable treatment method to achieve
stream discharge since the procedure of
diluting livestock wastewater to obtain
constructed wetland loading rates for
advanced discharge standards is not -
consistent with basic principles of
wastewater volume reduction and pollution
prevention.
The wetland system is Very well monitored.
To further evaluate the potential for nitrogen
removal at higher loading rates, the TKN
loading rate has been increased to
15 kg/ha/day. In addition, researchers are
evaluating model unit processes that could
improve treatment, such as overland flow,
media filter, aerated lagoon, and unaerated
lagoon. The goal of the current evaluation is
to identify a treatment train of aerobic and
anaerobic unit processes that provide
maximum removal of phosphorus and
nitrogen. The ultimate objective is to
incorporate wetland systems into livestock
wastewater management programs that
reduce costs and land requirements to swine
producers.
n-3
-------
Fresh
Water
2,600 Pig
Nursery
Anaerobic
Lagoon
Wetland Cells
System 1
System 2
System3
SU Sampling Station
HI Flow Meter
Fresh Water
**«N **<•* * 4 **t
6
Mixing Tanks
DDDDDn
DDDDDD
DDDDDD
Microcosms
-------
Performance of a Full Scale Constructed Wetland
Treating Swine Lagoon Effluent in Northern Alabama
T. A. MeCaskey, PhD., and T. C. Hannah*
A full scale constructed wetland was built in
the fall of 1988 at the Sand Mountain
Agricultural Substation in CrossviUe,
Alabama. The wetland size was based on
BOD5 design criteria(l kg BOD/ISO m2/day)
for a 500 pig-per-year farrow-to-finish
production facility (Hammer et aL, 1993).
The waste treatment system consists of three
anaerobic lagoons (two primary and one
secondary), a 0.1 hectare shallow mixing
pond, and five series of dual eel wetlands
(Figure 1). Each wetland cell was excavated
1.5 meters deep, 52 meters long, and 7.8
meters wide with an earthen bottom sloped
lengthwise less than one percent. The five
cells in the upper tier and five cells in the
lower tier have a total surface area of 0.405
hectares. No plastic liner or clay backfill was
used to seal the bottom of the cells.
Aerial view of the wetland system at the
Sand Mountain Experiment Station
In the spring of 1989, emergent aquatic
plants were planted in the wetland cells.
Species included broadleaf cattail (Typha
latifolid), soft stem bulrush (Scripus
validus), and rush (Juncus effusus).
Common reed (Phragmites amtralis), giant
cutgrass (Zizaniopsis milliacea), and
narrowleaf cattail (Typha angustifolia) were
planted in 1990. The wetland cells were kept
moist with pond water for two growing
seasons to allow the plants to become
established before wastewater was
introduced into the wetland cells.
Table 1. Treatment efficiency of a two-stage
constructed wetland treating swine lagoon
effluent over a 55-month period.8
Pollutant
TKN
NOrN
NH4-N
>jp
BOD5
COD
TSS
Lagoon
Effluent
155.3
<1
126.0
•51.1
146.3
554.6
241.5
Wetland
Inflow
(mg/L)
73.7
<1
55.6
28.4
76.6
319.9
135.7
Outflow
(mg/L)
12.2
<1
8.6
6.8
7.9
64,2
15.5
%
Change
83.4
<1
84.5
76.1
89.7
79.6
88;e
a. Lagoon effluent plus dilution water from a farm pond
Manure from a swine farrowing house,
nursery, and finishing house were flushed
into two primary lagoons, which overflowed
into a secondary lagoon. Effluent from the
secondary kgoon was combined with pond
water (2.7 parts to 1 part) in a 0.1 hectare
mixing pond to reduce the ammonia content
of the kgoon effluent below 60 mg/L.
* Professor and graduate research assistant, respectively; Department of Animal and Dairy
Sciences, Auburn University, Alabama
-------
Effluent from the mixing pond was
distributed equally into each of the five cells
in the upper tier of wetland cells (each 0.4
hectares). Effluent from each upper cell
flowed into a corresponding lower cell.
Based on the mean hydraulic flow rates over
a 55-month period, the theoretical hydraulic
retention time for both tiers of cells was 18
days, and the mean BOD5 loading rate was
5.9 kg BODs/ha/day. How rates were
monitored daily excluding weekends and
holidays, and wastewater samples were
collected and analyzed biweekly. Data
collected over a 55-month period (Table 1)
indicate that the constructed wetlands were
highly efficient at treating swine lagoon
effluent. Total Kjeldahl nitrogen (TKN)
content of the wetland inflow was reduced
from 73.7 mg/L to 12.2 mg/L after
treatment, an 83.4% reduction. Ammonia
nitrogen represented 75.4% of the TKN in
the wetland inflow, and the NH4-N was
reduced through wetland treatment from
55.6 to 8.6 mg/L, a reduction of 84.5%.
The influent to the wetland, a combination of
lagoon effluent and pond water, was
essentially anaerobic. Since no dissolved
oxygen was present in the influent
wastewater, ammonia was not readily
converted to nitrate; thus, NO3-N
concentrations in the wetland effluent were
always less than 1 mg/L. Total phosphorus
(TP), BODS, COD, and total suspended
solids (TSS) were reduced 76.1%, 89.7%,
79.1%, and 88.6%, respectively.
Most of the treatment occurred in the upper
tier of wetland cells (1A through 5 A in figure
1), with treatment results shown in Table 2.
The USDA-NRCS (1991) guidelines
recommended that effluent concentrations
from animal waste constructed wetlands be
<30 mg BOD5/L , <30 mg TSS/L, and <10
mg NH4-N/L. Treatment by the upper tier of
cells was sufficient to meet the effluent
criteria for BOD5 and TSS, but both the
upper tier and lower tier of cells were
required to reduce NH4-N to the acceptable
levels.
The BOD5 loading rate for the entire
wetland, including the upper and lower tiers
of cells, was 5.9 kg BOD5/ha/day. This
loading rate is approximately 61 kg
BOD/ha/day less than the rate suggested by
Hammer et al. (1993) and USDA-NRCS
(1991). Based on average concentrations of
analytes in the wetland effluent during the
55-month study, the 5.9 kg BOD5/ha/day
loading rate met the effluent criteria
suggested by USDA-NRCS (1991).
However, during the winter and early spring
months, when heavy rainfalls occur in
Alabama, even this BOD5 loading rate was
too high, and on several occasions NH4-N
concentrations in the wetland effluent were
in excess of the 10 mg/L limit suggested by
USDA-NRCS (1991). Although the wetland
Sagittaria thrives in treated swine effluent
effluent cannot be legally discharged and
must be recycled, minimum treatment
efficiencies for wetlands treating animal
manure lagoon effluents should be
H-6
-------
Table 2. Wastewater treatment efficiency occurring in u
Pollutant
TKN
NO3-N
NH4-N
TP
BOD5
COD
TSS
Upper Tier
Inflow
mg/L
73.7
<1
55.6
28.4
76.6
319.9
135.7
Outflow
mg£L
27.1
<1
20.7
12.7
16.8
107.7
19.1
Decrease
%.
63.2
—
62.8
55.3 .
78.1
66.3
85.9
sper and lower tiers of wetland cells.
Lower Tier
Inflow
mg/L
27.1
<1
20.7
12.7
16.8
107.7
19.1
Outflow
mg/L
12.2
<1
8.6
6.8
7.9
64.2
15.5
Decrease
• %.
55.0
—
58.5
46.5
53.0
40.4
18.8
mandatory because there can be no
guarantee that wetland effluents will be
totally contained during high rainfall events
that occur during the winter months.
References
Hammer, D. A., B. P. Pullen, T. A.
McCaskey, J. T. Bason, and V. W. E. Payne.
(1993). Treating livestock wastewaters with
constructed wetlands. In: Constructed
wetlands far water quality improvement. G.
A. Moshiri, ed., Lewis Publishers, CRC
Press, Inc., pp. 343-347.
U. S. Department of Agriculture, Natural
Resource Conservation Service. (1991).
Technical requirements for constructed
wetlands for agricultural wastewater
treatment. National Bulletin No. 210-1-17.
Washington, DC.
n-?
-------
G
O
O
s?
I
$
1
o
M
1-i
CM
Secondary
lagoon
d
o
DO
Farm pond
Figure 1. Diagram of the Sand Mountain Constructed Wetland
System at the Agricultural Experiment Station, Crossville, AL
II-8
-------
Tom Brothers' Dairy Constructed Wetland
Richard P. Reaves, PhD., and Paul J. DuBowy, PhD.*
Introduction Tippecanoe Lake.
The Tom Brothers' dairy, located in
Kosciusko County, Indiana, is a family
operation owned by Garry and Max Tom.
The dairy milks about 70 cows and is
therefore classified as a nonpoint source
facility rather than a point source subject to
the more stringent regulations of the Indiana
Confined Feeding Law.
Like many small dairies in northern Indiana
the Tom Brothers' dairy is located upslope of
a lake formed by glacial recession. The lake
in this case is Tippecanoe, which has been
developed for residential and recreational
use. In the early 1990's there was concern
that the Tom Brothers' dairy was adversely
impacting water quality in the lake In
summer, the lake would be closed to
swimming because of elevated fecal coliform
levels in a portion of the lake near the dairy.
Although Garry and Max Tom believed their
operation was not the cause of the high
bacterial levels in the lake, they chose to
participate in a constructed wetland
demonstration project with the USDA Soil
Conservation Service (now the Natural
Resources Conservation Service), the
Indiana Department of Environmental
Management, and Purdue University. The
project would augment Tom Brothers'
existing waste management system;
demonstrate a relatively new waste
management option; and determine the
degree to which the dairy might be
contributing to water quality problems in
The waste management system
Waste at the dairy is scraped daily from the
barns to a stack pad. The stacked manure is
removed regularly and land applied to notill
crops on the farm. Barn wastewater (750 L
d"1) is delivered to a septic manure pit
located beneath the stack pad where solids
are settled and separated from the
wastewater. Liquid from the stack pad drains
into the pit through slots beneath the pad.
Wetland as viewed from adjacent silo
(Courtesty Brian Miller)
The constructed wetland consists of two
cells in series (see Figure 1). The first cell is
a rectangle (64.6 x 14 m). The second cell is
horseshoe-shaped, with the two arms each
being 32.3 x 14 m and the upper end or
crossover being 9 x 6.1 m. The bottom slope
in both cells is 0.25%, resulting in a depth
differential of 16 cm from inlet to outlet.
Cell 1 and the first half of Cell 2 are lined
with plastic to prevent groundwater
Biologist, SD/International, Cincinnati, OH; and Associate Professor, Dept. of Wildlife and
Fisheries Sciences, Texas A&M University, TX, respectively.
II-9
-------
contamination. Soils beneath the last half of
Cell 2 have sufficiently low conductivity to
preclude the need for a liner.
Both cells were hand planted with broadleaf
cattail (Typha latifolia) obtained from
roadside ditches within the county.
Smartweed (Pofygonum spp.) and water
grass (Echinochloa walteri) volunteered in
the system. The first cell was a near
monoculture of cattails, and the second cell
became a mix of cattails and water grass as
predominant plants. Smartweed was a
marginal plant in each cell, growing around
the edges and in shallow areas. Overflow
from the manure settling pit discharges by
gravity to the upstream end of the first
wetland cell The effluent is distributed across
the width of the cell through a slotted
horizontal pipe. In addition to the pit
discharge, yard runoff is diverted around the
stack pad and delivered to Cell 1 at a point
approximately 60% of the distance from the
upper end; this discharge enters the side of
the cell and is not distributed across the cell.
Effluent from Cell 1 flows into the upstream
arm of Cell 2 and is redistributed across the
width of the cell through a slotted pipe, in
the same manner as delivery of the pit
effluent to Cell 1. Effluent from Cell 2 enters
a holding pond (35 x 35 m surface water
dimensions and depth of 1.8 m when full).
During periods of overflow, the holding
pond discharges into an infiltration strip for
final disposal
Results and Discussion
The wetland became operational in the
spring of 1994 and was monitored through
1995. Water samples were collected from the
inlet to Cell 1, from the point where the yard
runoff entered Cell 1, from the inlet and
outlet of Cell 2, and from the holding pond,
the infiltration area, and a roadside ditch
downhill of the infiltration area. The ditch
received seepage water from the a
subsurface tile beneath the infiltration area as
well as road runoff. Samples were collected
at least twice monthly during the growing
season. On some of the sample collection
dates, the wetland cells had no standing
water. On those dates a "no flow" reading
was noted. Samples were collected from the
holding pond only on those sampling dates
when flow was occurring in any part of the
wetland system.
Samples were evaluated for five-day
carbonaceous oxygen demand (CBODS),
total suspended solids (TSS), total Kjeldahl
nitrogen (TKN), total phosphorus (TP),
reactive phosphorus (PO4-P), ammonia
nitrogen (NH4-N), nitrite nitrogen (NO2-N),
and nitrate (NO3-N). Fecal coliform bacteria
were evaluated at the outflow of each
wetland cell and in the holding pond. They
were not evaluated at the initial inlet points
to Cell 1 because high suspended solids
concentrations created problems in
conducting the tests. Beginning in July 1994,
pH, conductivity, temperature, and dissolved
oxygen were added to the analyses.
No data are presented for 1994 because the
pit had been pumped out the previous fall
and very little waste discharged to the
wetland. The wastewater entering Cell 1 was
not typical of liquid dairy waste. During kte
summer Cell 1 went dry for lack of both
rainwater and wastewater, so there was no
flow into either wetland cell
In 1995, wastewater from the pit and runoff
from the lot entered the wetland, and the
n-io
-------
inflow concentrations were like those of a
typical dairy. In early spring, wetland
performance was not as effective as later in
the season. This was, in part, due to the
heavy flush of waste into the system from
spring rains, the cool temperatures which
slowed microbial growth, and the immaturity
of wetland plants early in the season. After
the spring rains, the weather turned dry, and
flow through the system was greatly
reduced. The cells were virtually dry by
September, and inflow rates did not return to
normal until kte October. The low rainfall
levels from kte spring through summer
resulted in increased detention times and
improved performance.
Table 1 shows the average annual
concentrations of various constituents and
the percent difference between the inflow
and outflow of Cell 1 and between the inflow
to Cell 1 and the outflow of Cell 2. It should
be noted that the term "percent difference" is
used rather than "percent removal efficiency"
because the change in concentrations is
affected by dilution by direct precipitation on
the wetland cells and also by the fact that
runoff water entered Cell 1 at a point along
Table 1. Treatment efficiency for a two-cell wetland treating runoff from the Tom Brother's Dairy
in Indiana.
Constituent
TKN (mg/L)
NH4-N (mg/L)
PO4 (mg/L)
TP(mg/L)
CBOD5 (mg/L)
TSS (mg/L)
Cell 1 Inlet
Concentration
215.3
199.4
47,3
25.3
910.3
483.4
Cell 1 Outlet
Cone. (% diff.)
113.1
99.8
28.9
10.8
155.6
113.2
(47)
(50)
(39)
(57)
(83)
(77)
Cell 2 Outlet
Cone. (% diff.)
30.4
21.6
10.0
4.2
67.6
30.7
(86)
(89)
(79)
(83)
(93)
(94)
its width, as noted above. Thus, there was,
indeed, an overall reduction in concen-
trations but the differences do not necessarily
reflect only the percent reductions due to
treatment. A mass balance was not
conducted because a rain gage was not
installed and information on seasonal
evapotranspiration rates were not available
for the various plants for this region.
Table 2 presents data on samples from the
holding pond, the infiltration area, and the
ditch. Some increases in concentrations
occurred for nearly all constituents. This is
probably the result of algal growth in the
pond. It should also noted that, although
only two samples were collected from the
ditch, the concentrations of all potential
pollutants were low.
11-11
-------
Table 2. Average concentrations of wastewater constituents in the holding pond, infiltration area,
and ditch (1995).
Constituent
TKN (mg/L)
NH4-N (mg/L)
PO4 (mg/L)
TP(mg/L)
CBOD5 (mg/L)
TSS (mg/L)
Holding Pond
22.8
10.0
3.1
4.3
29.0
61.0
Infiltration Area
20.2
<0.02
1.4
5.0
71.6
30.0
Ditch
1.4
<0.02
<0.02
<0.02
9.3
16.5
Summary
It became evident through this study that
some storage capacity is needed upstream of
the wetland to allow wastewater to be held
during the dormant season and to alow
proper timing of the effluent to the wetland
in conjunction with wet and dry periods.
Excess water can be used to dilute the waste
and to maintain flows during dry seasons.
The study at the Tom Brothers' Dairy
indicates that high levels of pollutant removal
can be achieved. It should also be noted that
a large measure of the success of this project
is due to the good management of the Tom
brothers. A constructed wetland was
installed on a similar sized dairy in northern
Indiana, and it failed for lack of good
management. Fences must be maintained and
water levels managed, among other things.
In other words, some work and commitment
are required on the part of the farmer if the
wetland is to be successfuL
As result of this study, it was determined that
the pollution in the downstream lake was not
attributable to the Tom Brothers' Dairy.
References:
1. Reaves, R. P, and P. J. DuBowy (1996);
Performance evaluation of the Tom
Brothers' Dairy waste treatment wetland',
Proc. of the 2nd Natl. Wkshp. For
Constructed Wetlands for Animal Waste
Management, Ft. Worth, TX; Texas A&M
Press, TX.
2. Reaves, R. P. (1995); Evaluation of free
water surface wetlands for treatment of
livestock wastes in Indiana; Ph.D.
Dissertation. Purdue University, West
Lafayette, IN.
3. Reaves, R P. P. J. DuBowy, D. D. Jones,
and A. L. Sutton (1995); Constructed
wetlands treatment of animal waste in
Indiana: management implications, In Clean
Water Clean Environment. 21st Century,
Vol. H: Nutrients. ASAE, St. Joseph, MI.
H-12
-------
Cropland
Yard runoff collection
Drying
pad -
Manure pit
Runoff diversion
Cell 2
Holding
pond
Flow control Switching valve
Infiltration
Area
Figure 1. Generalized layout of waste treatment system at Tom Brothers' Dairy
n-is
-------
A Constructed Bulrush Wetland for Treatment of Cattle Waste
Charles M. Cooper, PhD.* and Sam Testa III**
Introduction
Processing and disposing of concentrated
on-farm animal waste, a major source of
water quality deterioration, is a concern of
the Natural Resources Conservation Service
(NRCS) and regulatory agencies. Several
projects for evaluating the ability of
constructed wetlands to process animal
waste have been initiated across the United
States. As a result, optimal design criteria for
such future animal waste management
systems may be forthcoming. The Mississippi
NRCS and the Agricultural Research Service
(ARS) National Sedimentation Laboratory in
Oxford, Mississippi, cooperated on an
on-farm dairy waste treatment project which
used a constructed bulrush wetland for
processing. Herein we present findings from
three years of operation and make
suggestions for future design criteria for such
systems.
Development of the Study Site
The Alan Scott dairy farm is located in
DeSoto County, in extreme northern
Mississippi, approximately 5 mi ESE of
Hernando. During the study period an
average of 80 (60 to 100) Holstein cattle
(approximate weights of 1000 to 1200
pounds each) were milked twice daily in a
concentrated animal feeding operation
where they were held for approximately 6
hours per day. Total runoff area for the
ranking parlor and concrete loafing area
where animals were confined during milking
was 351.5 m2. Total waste production in this
area was estimated at 10,336 liters per day.
Wastes drained through 15.24 cm (6 in)
diameter PVC pipe to a 42 m x 52 m settling
lagoon. The lagoon received input from
milking equipment and tank cleanings,
milking barn washing, loading area runoff,
and rainfall. Export from the lagoon, drawn
Cell 2 with walkways (Summer 1991)
from approximately 0.3 m below the water
surface, traveled through 7.62 cm (3 in)
diameter PVC pipe to three parallel con-
structed wetland cells, each 6 m wide and 24
m long (Figure 1). Wastewater entered the
eels through a horizontal, perforated section
of pipe which spanned the width of the cell
to prevent short-circuiting or channel flow.
The pipe was elevated 20 cm above the
water surface to prevent settling of solids
and to allow for easier periodic cleaning.
Land slope was such that only part of the
bottom of the cells was excavated; the
remainder of the bottom and levees was built
from soil excavated from the lagoon to
create an approximate 0% slope system.
Construction occurred in April, 1990, and
*Research Leader and Supervisory Ecologist, **Biologist; USD A-ARS, National Sedimentation
Laboratory, Oxford, MS.
-------
constructed wetland cells were planted
immediately in bulrush (Scirpus validus) at
0.3 m intervals with rhizome cuttings
purchased from a wildlife supply company.
Subsequent rains, supplemented with water
pumped from the lagoon and weU water,
maintained standing water in the cells for the
remainder of the year. Water level in the
lagoon increased slowly because of high
evaporation rates and lateral seepage
through levees until the basin sealed. An
insufficient amount of water accumulated in
the lagoon to allow a gravity fed water
supply to the cells during 1990. Bulrush
growth in the cells was rapid. By September,
1990, the cells were covered by a uniformly
dense monoculture with the majority of
culms supporting flowering/seeding heads.
Natural senescence occurred in November
and December. Re-emergence of bulrushes
from rhizomes occurred in February, 1991,
through the litter created by the previous
year's growth. Duckweed (Spirodela pply-
rhiza) spread to cover nearly all available
water surface by May, 1991. In April, gravity
flow from the lagoon to the wetland cells
began functioning. Discharges to cells were
calibrated to yield 3.0 L/min using in-line
valves. Water depth in the cells was a
maximum of 0.3 m.
Rapid water level decline in the anaerobic
lagoon during summer, 1991, prompted a
reduction of cell inflow rates to 0.5 L/min,
but settling of solids in pipes and valves
generally resulted in lower rates. Standpipes
were fitted with threaded end caps con-
taining orifices sized to achieve desired flow
rates. Original valves were opened fully to
prevent occlusion. Because of variations in
flow a 4,000 L constant head tank was
placed on the lagoon levee and connected to
the supply pipe that led to the cells. A timer-
controlled electric pump maintained water in
the tank, creating constant hydraulic head
and, thus, producing constant inflow. Using
this method, a cell inflow rate of 1.0 L/min.
was implemented, and the frequency of
remedial action was greatly decreased. Also
during the summer, 1991, another cell, Cell
4, was constructed in series with Cell 1 to
allow greater loading capacity and
assessment of further treatment (Fig. 1).
Mats of decaying vegetation from previous
years growth increased during 1992-1994,
and caused sparser and clumpy emergence of
the bulrush within the treatment cells.
Growth of volunteer vegetation such as
Smartweed (Polygonum sp.) and cutgrass
(Leersia sp.) increased also. Removal of the
decaying vegetative mat from affected areas
with hand tools restored bulrush growth.
Methods
Eighteen parameters were monitored at
biweekly intervals from May, 1991, through
April, 1994. Total rainfall for the two week
period prior to sampling and lagoon water
column depth were recorded. Lagoon
samples were taken from the outflow control
platform at a depth of 0.3 m below water
surface. Flow rate, temperature, conducti-
vity, dissolved oxygen, pH,, total solids,
dissolved solids, suspended solids, filterable
ortho- phosphorus, total phosphorus,
ammonia nitrogen, nitrate nitrogen, total
chlorophyll, sediment redox potential 5-day
carbonaceous biochemical oxygen demand,
and total coliforms were measured at cell
inflow and outflow. Chemical oxygen
demand was determined at all sampling sites
quarterly.
Early in the project 3 walkways were con-
n-is
-------
Rainwater inlet-pipe-W^
Inflow
LAGOON
Direction of Flow
40m x 52m
PITCH
°Utf
2
o o
o3
Outflow/Inflow
Inflow
CELL#1
a
Inflow
CELL #3
f
Outflow
CELL #4
fi Outflow
CELL #2
^"^ Walkway
fi Outflow
6m x 24m
each
Valves 1, 2, and 3
• Main
Valve
Figure 1. Layout of lagoonXwetland cell system at Hernando
Wetland on Alan Scott Farm., DeSoto County, Mississippi, USA.
40%
60%
-20% 0% 20%
Figure 2. Mean percent reduction for each parameter.
11-16
80%
100%
-------
stracted equal distances apart in Cell 2
(Fig.l) so that in-cell measurements could be
taken at intervals without disturbing the cell.
Results from in-cell processing and a more
detailed discussion of the project were given
earlier by Cooper et aL (1995).
Water quality parameters were measured
according to APHA (1989) guidelines. Cells
did not continually discharge due to
evapotranspiration, ground infiltration and
seepage. When such times coincided with
sampling visits, water quality samples and
measurements from non-discharging outflow
stations were taken by tilting the standpipe
until flow occurred. For computing loading
rates of pollutants, hydraulic load on
individual cells was 1 cm/day (1440 L over
144 m2 per day).
For purposes of computing seasonal values
for wetland performance, seasons were
assigned months as follows: SPRING =
February, March, April; SUMMER- May,
June, July; FALL - August, September,
October; WINTER - November, December,
January. The analysis period for which the
following summaries are presented began
May 1991 (beginning of Summer season)
and ended with May 1994 (end of Spring
season). Data from monitoring of the original
three parallel wetland cells follows directly.
Results that include Cell 4 in series with Cell
1 are detailed in a later section.
Results and Discussion
Measured parameters varied with season and
as the system matured. Overall reductions of
individual parameters, calculated from mean
inflow and outflow measurements over the
duration of the study, can be compared using
Figure 2. Warm season (summer and fall)
and cool season (winter and spring)
information is given in Table 1. Average
rainfall for the study area is 127 cm/yr.
Precipitation was 101 cm the first sampling
year (May 1991-May 1992), 132 cm the
second year, and 162 cm the third year.
Increasing rainfall amounts through the study
period could affect interpretations of
seasonal and long-term wetland functioning.
Individual rainfall events resulted in
temporarily increased discharge from the
cells, turbulence, and dilution. Fluctuations
in rainfall, variability in waste production,
and weather conditions also influenced water
depth in the anaerobic lagoon.
Inflow rates to the cells for most of the study
period were targeted at 1.0 liters per minute.
Actual inflows fell between 0.75 and 1.25
L/min at 84% of our sampling visits. As
noted in the methods section above, the
wetland cells did not discharge continually.
There was zero discharge from the system at
43% of visits to the site. Outflow was
observed at 103 out of 181 sampling visits
(57% frequency). Of these 103 discharge
observations, 57 (55%) were at a rate of less
than 0.75 L/min, and 83 (81%) at less than
1.25 L/min. Discharges in excess of 1.0
L/min were always associated with rainfall
events except during the initial high inflow
phase of the project. Average warm season
inflow was 1,354 L/day, and warm season
outflow was 576 L/day. Average cool season
inflow was 1,368 L/day, with outflow of 749
L/day. When no outflow occurred, water
samples and measurements were taken after
tilting the outflow standpipe until discharge
resulted, and the pipe was flushed. This
allowed within-eell reduction efficiencies to
be calculated without the necessity of
outflow.
H-17
-------
Table 1. Warm and cool season parameter means and percent difference due to constructed
wetland processing.
Warm(W)
Cool (C)
W
C
W
C
W
C
W
C
W
C
W
C
W
C
W
C
W
C
W
C
Parameter
CBOD5
CBOD5
COD
COD
NH3+NH4-N
NH3+NH4-N
TOTAL P
TOTAL P
PO4-P
P04-P
TSS
TSS
DO
DO
Fecal CoMforms
Fecal CoHforms
PH
PH
Water Temp.
Water Temp.
Units
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
No./100 mL
No./100mL
Std. Units
Std. Units
deg.C
deg.C
Influent
29.99
31.08
244
343
5.54
8.02
12.66
20.83
8.59
11.2
128
111
2.87
4.77
9,970
19,968
6.86
7.2
22.46
11.23
Effluent
9.51
5.07
122
80
1.8
1.55
7.19
7.67
4.96
6.4
53
33
1.43
2.58
1,136
596
6.15
6.52
20.43
9.43
% Change
68
84
50
77
68
81
43
63
42
43
59
70
50
46
89
97
9
9
9
16
Temperatures at outflows from the cells
were 10,9% lower than at inflow stations
because of shading by wetland plants and the
shallow depth of water within the wetland
cells. This was most evident during winter
when outflow water temperatures averaged
21% lower than inflow temperatures.
Summer temperature reductions averaged
8%. Inflow station extremes ranged from 5.4
to 30.3° C (mean = 17.8). Outflows ranged
n-18
-------
from 1.5 to 27.3° C (mean = 15.9).
Conductivity decreased 28.5% with passage
through the constructed wetland cells.
Greatest reductions occurred during the
winter and spring seasons. Reductions
increased each year of the study to a peak
during spring of 1994 at 44%. Conductivity
varied from 28 to 773 jimhos/cm at inflows
(mean = 343). Outflow conductivity ranged
from 103 to 785 umhos/em (mean = 245).
Dissolved oxygen concentrations decreased
by nearly half (48.8%) when passed through
the wetland cells. Increases were measured
only during the initial three months of
operation. Reduced oxygen levels were
attributable to biochemical oxygen demand,
bacterial consumption (and nitrification), and
duckweed which quickly colonized open
water surface. Measurements during the
study period ranged from 0.03 to 14.2 (mean
= 3.6) mg/L for inflow stations, and from
0.03 to 7.3 (mean = 1.9) mg/L at outflo'ws in
the reducing environment.
A small (9.9%) decrease in pH was observed
for water flowing through the wetland cells.
Inflow values for pH ranged from 5.7 to 8.5
(mean = 7.0). Outflow values ranged from
5.7 to 7,4 (mean = 6.3). Seasonal reduction
percentages for pH were fairly uniform
throughout most of the study. Redox
potential in the wetland cells increased an
average of 137%. Inflow station
measurements ranged from (-)259 to (+)311
mV with a mean value of (-)48.39 mV.
Outflow measurements ranged from (-)270
to (+)395 mV with a mean of (+)18 mV.
Research by Rogers et al (1991) suggested
that increased redox potential in wetland
waste treatment systems is due to plant
presence, while a decrease in redox occurs in
implanted systems. Though mean values at
our site showed an overall increase, values
varied widely during the study.
Dissolved solids removal was low (21.8%),
while suspended solids removal was
relatively high at 60.5%. Total solids were
reduced by 31.6% during the three year
evaluation (Figure 2). Suspended solids
reduction in the wetland cells exhibited
distinct seasonal changes linked to plant
growth/senescence and plant biomass
accumulation/decay. Since much of the
suspended solids contained in the waste
settled in the lagoon, dissolved solids were
the major component entering the wetland.
Dissolved solids concentrations during the
study varied from 72 to 573 mg/L (mean =
364) at inflow stations, and from 60 to 554
mg/L (mean = 285) at outflows. Suspended
solids ranged from 0.0 to 466 mg/L (mean =
122) at inflows and 0.0 to 332 mg/L at
outflows (mean = 49). Total solids at
inflows) ranged from 176 to 749 mg/L
(mean = 484), and at outflows from 149 to
605 mg/L (mean = 331).
Total phosphorus (TP) removal averaged
53.2% for the three year study period.
Removal efficiencies climbed from slightly
above 50% during the initial season of
operation to near 85% after 9 months of
system operation. Trap efficiency declined
over the next 6 months to 44% removal
(summer, 1992). Thereafter, phosphorus
removal efficiencies remained moderate.
Inflow TP concentrations ranged from a
minimum of 1.3 mg/L to a maximum of 69.0
mg/L (mean = 15.9). Outflow concentrations
ranged from 0.2 to 22.8 mg/L (mean = 7.4).
Principal phosphorus removal mechanisms
were probably precipitation and adsorption
to sediments. Plant uptake accounted for
n-19
-------
some removal If plant removal had been a
major uptake mechanism, reduction
efficiency would not have declined drastically
during the study. Spangle, et aL (1976)
found 30 to 66 percent of the total
phosphorus in bulrush wetland cells was
associated with substrate. Phosphorus is
immobilized in organic materials and
saturation is reached rapidly (Hammer and
Kadlec, 1983). Dolan et al. (1981) discussed
phosphorous dynamics in a Florida marsh
receiving treated wastewater, and Jones and
Lee (1980) evaluated wetlands based
phosphorus control for eutrophic waters.
Filterable ortho-phosphorus (FOP) removal
efficiency averaged 42.4%, somewhat lower
than that for total phosphorus. FOP trapping
by the system was near 70% during the initial
season of operation and peaked at 85%
during the second season. Trapping
efficiency declined nearly linearly from that
point during the next 12 months to 31% in
fall 1992, and averaged 37% afterward.
Inflow concentrations varied from 0.9 to 24
mg/L (mean =9.6). Outflows had a low of
0.1 mg/L and high of 15.5 mg/L (mean =
5.5).
Ammonia nitrogen reduction by the wetland
system averaged 81.6% overall Reductions
exceeded 90% during the first year of
operation, then declined to an average of
81% for the next 5 seasons of the study
(summer 1992 through summer 1993).
Removal efficiency exceeded 90% again in
fall 1993. Reduction then declined to 57%
during the next season (winter 1993) and
was 65% the final season of the study period
(spring 1994). Ammonia nitrogen
concentrations entering the wetland cells
varied from a low of 0.1 mg/L to a high of
30.8 mg/L (mean = 7.0). Minimum
outflowing concentrations reached
undetectable levels (<0.01 mg/L) while the
maximum outflow concentration measured
was 10.8 mg/L (mean = 0.1).
Nitrate nitrogen concentrations entering and
leaving the wetland treatment system were
low (means = 0.09 and 0.1 mg/L,
respectively). Concentrations indicated a net
export of nitrate 14.4% higher than inflow,
though actual concentrations were nearly
negligible. Inflow and outflow
concentrations were not expressive of the
massive ammonia nitrogen to nitrate nitrogen
conversion that occurred within the cells.
Export of nitrates was influenced almost
totally by that transformation. Seasonal
nitrate-N processing began with a 28%
average reduction for the first two seasons of
treatment, followed by two seasons of 270%
export. The system then fluctuated
between net reduction and net export
during the following 15 months
(summer 1992 through summer 1993,
averaging 5% reduction overall), before
exhibiting > 50% reduction in nitrate-N for
the last three seasons of the study (fall 1993
through spring 1994). Inflow nitrate nitrogen
ranged from undetectable (<0.01 mg/L) to
0.9 mg/L. Outflow concentrations also
ranged from undetectable levels to 3.3
mg/L.
Five-day carbonaceous biochemical oxygen
demand (BOD5) was reduced consistently by
about 80% following the first season of
operation in which there was only a 42%
reduction (overall 74.6% reduction). BOD
for inflow stations averaged 35.1 mg/L
(minimum = 9.7, maximum = 80), while
outflow stations averaged only 8.9 mg/L
(minimum = 0.3, maximum = 48).
H-20
-------
Total chlorophyll was also reduced by about
75% (78.8%), though with more seasonal
fluctuation than seen for BOD. Mean inflow
concentration was 306 mg/L, with a
minimum of <0.01 and a maximum of 1,505
mg/L total chlorophyll. Outflows had a mean
of 64 mg/L, a minimum of 1 mg/L and a
maximum of 759 mg/L. Inflowing
chlorophyll was reduced because of settling
and flocculation. In-cell production was
minimal because of plant shading.
Coliform bacteria were abundant in
pre-treatment lagoon wastewater; yet our
data showed there was an 89% reduction in
total coliforms with passage through the
wetland cells. Inflow concentrations had a
mean of 14,525 colony forming units
(CFUyiOO ml, with a minimum of 40 and a
maximum of 101,000 CFU/100 ml. Outflow
mean concentration was 1,585 CFU/100 ml,
with a minimum observed of 20 and a
maximum of 19,700 CFU/100 ml. (TMs
information excludes individual tests where
sample dilution resulted in extinction of
coliform bacteria and resultant lack of colony
forming units.)
Chemical oxygen demand, the oxygen
equivalent of the organic matter that can be
oxidized by a strong chemical oxidant, was
measured on a quarterly schedule. Average
inflow demand was 263 mg/L, while outflow
demand was only 96 mg/L, resulting in a
mean reduction in COD of 63% with passage
through the wetland cells.
Results from Addition of Cell 4
A single additional cell of the same
dimensions as an original cell was added in
series to Cell 1 during Summer 1991 (Fig.
1). This cell, Cell 4, received effluent from
Cell 1 only, and served as an experimental
polishing cell. The cell was planted in Spring
1992 (late April/early May) with rhizome
cuttings at one meter intervals. By early
June, 1992, plantings had expanded as
healthy spreading clumps. By early August
they formed a nearly continuous stand within
the cell.
When compared to mean changes of
parameters in the original three cells, Cell 4,
acting as an additional treatment cell,
produced the following notable changes in
water quality. Conductivity produced an
added 23% reduction for a total reduction of
over 51% from inflow values at Cell 1.
Dissolved oxygen concentrations in Cell 4
increased relative to CeU 1 outflow, yet did
not equal concentrations entering the
wetland cells from the lagoon. This increase
in oxygen resulted in an overall 28%
reduction of DO, as opposed to a 48%
average reduction that occurred in the three
original cells. Likewise, pH values increased
as water flowed through Cell 4, as opposed
to a decrease in the original cells, but again
not reaching original inflow values (5%
decrease with Cell 4 included vs. 10%
average decrease in original cells alone).
Total solids decreased an additional 20%, for
a total reduction of >51% because of
dissolved solids trapping. There was little
change in suspended solids concentrations in
CeU 4.
Filterable ortho-phosphorus concentrations
declined an added 37%, and total
phosphorus declined an additional 23% in
Cell 4. This added trapping was similar to
initial phosphorus trapping in the first three
cells. Ammonia-nitrogen concentrations
11-21
-------
declined an additional 13% over the average
original cell reduction of 82%.
Nitrate-nitrogen concentrations at outflow
from Cell 4 were 52% lower than inflow at
Cell 1, negating the increase of
nitrate-nitrogen observed in the original cells
(though, again, actual concentrations were
wry low, with mean values of 0.10 mg/L or
less for all inflow and outflow stations).
Total chlorophyll concentrations and BOD
were 9% less when Cell 4 was used.
Coliform bacteria concentrations were
relatively unchanged.
Observations and Recommendations
A. Biomass Removal: Bulrush vegetative
growth generally matted after senescence. In
many cases mats assisted in forming
anaerobic conditions and were impenetrable,
preventing renewed spring growth.
Harvesting of biomass enhanced growth in
Cell 1 and eliminated the matting problem
for the short term. However, bulrushes
should not be recommended unless some
harvesting method is planned. An annual
harvest would also reduce the phosphorus
which is temporarily bound in plants. Natural
wetland studies reveal seasonal export of
phosphorus in spring after plant material
decays (Spangle, et aL, 1976). Our study
exhibited a decline in total phosphorus
trapping each spring. Spangle, et al. (1976)
also found that a single fall harvest netted
greater biomass than several periodic
cuttings over the growth season. Gersberg,
et al.(1983) recommended annual harvest
for improved cell productivity. They also
mulched wetlands to add carbon to assist in
nitrogen removal With some species such as
cattails, burning excess biomass during the
simulated dry part of a hydro period may be
feasible.
B. Plant spacing: Some authors (Gearheart,
1992) recommend high initial planting
density. We initially used a 0.3 m setting on
staggered rows for our first plantings. When
we planted our 4th cell and
replanted bare spots, we used a 1.0 m setting
on staggered rows. The 1.0 m setting was
satisfactory with Sdrpus validus because of
vigorous rhizome growth.
C. In-cell processing and cell dimensions:
Results showed that some contaminants
were processed in a linear fashion and that
processing was dependent upon cell length.
Processing for others was mainly in the first
third of the cell. Cell design should be
targeted for principal contaminants, and cell
size should depend on the worst case
processing efficiency. Over designing cells so
that outflow is not continuous but is seasonal
allows for some resemblance of a natural
wetland hydro period.
D. Maintenance: Properly designed
inflow/outflow piping requires routine
flushing. Levee maintenance is essential.
Multiple cells are highly desired so that a
single cell can be isolated for maintenance.
With bulrushes, plant biomass becomes a
significant problem within three years. Plants
must be harvested or biomass otherwise
removed. We conducted two biomass
removals. Both exhibited some degree of
success although each had difficulties.
Mechanical removal of dead material
successfully allowed new shoots to sprout,
but it was kbor intensive. We also drained
one cell and burned dead material. This
method required constructing a sump and
pumping water from the cell for 4 days.
H-22
-------
Burning may be more successful with species
like cattails which produce clumpy, easily
burnable biomass.
E. Cost: Constructed wetlands represent a
low initial cost/low maintenance method for
treating some animal wastes.
Conclusions
Three parallel wetland cells, planted to giant
bulrush, were evaluated for 36 months while
receiving wastewater via a primary settling
lagoon from a <100 cow dairy operation.
Results from these three original cells
showed best reductions in coliforms, BOD,
chlorophyll and ammonia nitrogen, the
potential contaminants of concern. Study
length allowed for some evaluation of cell
maturation. Initial phosphorus and nitrogen
processing was quite high. However, when
cells became loaded with these nutrients,
efficiencies stabilized at lower rates.
Seasonal variations were evident, but cells
were functional continuously. Effectively
doubling the cell length by adding a fourth
cell allowed for 24 months of additional
comparison. Greatest changes from
additional length were in nutrient removal
and dissolved oxygen improvement. Cell size
should be based on most effective treatment
of targeted contaminants. The major
negative factor associated with bulrushes
was build up of biomass. Removal of
decaying biomass was essential for annual
plant emergence from rhizomes. Constructed
wetland cells represent an alternate method
of processing some agricultural wastes, but,
as our study showed, individual cell
variability and seasonal/long-term trends
make operation challenging.
Acknowledgments
This report was prepared as part of a
cooperative effort between the Agricultural
Research Service (ARS) at the National
Sedimentation Laboratory, Oxford,
Mississippi and the Natural Resources
Conservation Service (NRCS), Jackson,
Mississippi Research was accomplished as
part of the Demonstration Erosion Control
(DEC) Project in the Yazoo Basin with
partial funding provided by NRCS. NRCS
technical assistance was provided by Lon
Strong, Ross Ulmer and Jimmy Wilson.
Numerous New Albany Area and DeSoto
County District staff also helped with
construction aspects. Alan Scott, the farm
cooperator was helpful and provided a
willing opportunity for innovative
experimentation. The authors wish to thank
these people and the following ARS and
University of Mississippi personnel: Patrick
McCoy, Jennifer Bowen, Terry Welch, Betty
Hall, Belynda Garraway and A. T. Mikell.
Literature Cited
Amer. Public Health Assoc. (1989);
Standard methods for the examination of
water and wastewater. APHA, Washington,
D. C.
Cooper, C. M., S. Testa HI, S. S. Knight,
and J. J. Bowen. 1995. Assessment of a
constructed bulrush wetland for treatment of
cattle waste: 1991-1994; USDA,
Agricultural Research Service, National
Sedimentation Laboratory Research Report
No. 4. 50 pp.
Dolan, T. I, S. E. Bayley, J. Zolteck, Jr.,
and A. J. Hermann (1981); Phosphorus
11-23
-------
dynamics of a Florida freshwater marsh
receiving treated waste-water. J. Appl, EcoL
18:205-219.
Gearheart, R, A. (1992); Use of constructed
wetlands to treat domestic wastewater, city
ofArcata, California. Wat. Sci, and Tech.
26:1625-1637.
Gersberg, R. M., B. V. Elkins and C. R.
Goldman (1983); Nitrogen removal in
artificial wetlands. Water Research 17:
1009-1014.
Hammer, D. and R. H. Kadlec (1983);
Design Principles for Wetland Treatment
Cells. EPA 600/S2-83-026.
Jones, R. A. and G. F. Lee (1980); An
approach for the evaluation of efficiency of
wetlands-based phosphorus control
programs for eutrophication related water
quality improvement in downstream water
bodies. Water Air Soil Pollution.
14:359-378.
Peverly, J. H. (1985); Element accumulation
and release by macrophytes in a wetland
stream. J. Environ. Qual. 14:137-143. '
Spangler, F. L., W. E. Sloey, and C. W.
Fetter (1976); Experimental use of emergent
vegetation for the biological treatment of
municipal wastewater in Wisconsin In
Biological Control of Water Pollution, edited
by Tourbier, J. and R. W. Pierson, Jr., pp.
Univ. of Pennsylvania Press, Philadelphia,
PA.
H-24
-------
Essex Treatment Wetland, Essex, Ontario, Canada
Paul Hermans and John Pries*
Background
The Essex treatment wetland was
constructed on the Malder Valley farm in fall
1993 to treat barnyard runoff and milkhouse
washwater wastes from a dairy operation. It
is one of seven sites in Ontario where
research is being conducted to determine the
feasibility and treatment effectiveness of
constructed wetlands for reducing
contaminants in high strength wastewater
from animal operations. At the same time,
these projects will evaluate the effectiveness
of dissipating the water through natural
processes such as evaporation and
transpiration without discharging offsite.
Four of the systems, including the system in
Essex, were constructed using similar
designs to allow for easier comparison of the
monitoring and performance data. The Essex
design consists of a holding pond (see Figure
1) followed by a serpentine wetland treat-
ment cell that discharges into a final holding
pond. Capital and operating costs and system
maintenance requirements are being tracked
over time. Source controls to reduce the
contaminant loading to the treatment wet-
land include a covered manure storage that
was constructed to reduce rainwater runoff
from the manure and an exercise yard that
was paved and curbed with concrete and
sloped to drain to a central catch basin.
Treatment Wetland Design
Before installing the Essex treatment
wetland, all liquids drained off the barnyard
facility directly to Woltz Creek. The
milkhouse wastewater entered the creek
through a tile drain. Barnyard runoff and
approximately 200 gallons per day of
milkhouse washwater are directed to a sump
and then pumped to a 50,000 cubic foot (ft3)
sedimentation basin/facultative pond. The
pond was designed to pretreat the waste-
water by providing anaerobic conditions and
allowing solds to settle, thus reducing the
solids, BOD5, nitrogen, and total phosphorus
loading to the wetland. It also provides
storage during the nondischarge period of
approximately 6 months. The pond was sized
for a 100-year storm combined with wash-
water produced on a daily basis. Removal of
sediment from the pond is possible when
required with standard liquid manure
handling equipment or a backhoe.
The single wetland cell at the Malder Valley
farm has a surface area of about 600 m2
(0.15 ac) and is serpentine in shape with an
aspect ratio of about 24:1. During the
growing season, stored wastewater is dis-
charged at a controlled rate to the wetland
cell using an inground weir structure. This
weir also controls the liquid level in the
sedimentation basin. The wastewater flows
through shallow zones vegetated with cattail
(Typha latifotta), water plantain (Alisma
triviale), arrowhead (Sagittaria latifolia),
flowering rush (Butomus unbellatus),
softstem bulrush (Scirpus validus), and
duckweed (Lemna spp.) that are separated
by deep zones vegetated with duckweed,
bur-reed (Sparganium eurycarpum), horn-
wort (Ceratophyllum demersuni), and sedge
(Carex spp.). The vegetation was trans-
*Professional Agrologist, Essex Region Conservation Authority, Essex, Ontario, CN, and
Certified Engineering Technician, CH2M Gome & Storrie, Ltd., Ontario, CN., respectively.
11-25
-------
planted to the wetland cell from road-side
ditches in spring 1994. Monitoring equip-
ment was installed in fall 1993 and spring
1994.
The treated effluent discharges to a polishing
pond for final treatment. Water loss from the
final pond is due, in part, to evapotrans-
piration. There is a net precipitation gain in
Ontario, and the system will likely have some
discharge during part of the year. The wet-
land is designed with additional freeboard
capacity in the event of a severe rainstorm. If
the wetland cannot handle a major rain event
or the final holding pond is filled to capacity
during the rainy season, the treated water is
spray irrigated onto a grassed waterway
located in the farmer's pasture. If these
systems are proven to be effective for the
removal of contaminants and are cost-
effective, the Ontario Ministry of the
Environment and Energy (MOEE) may
require a certificate of approval under the
Water Resources Act to permit discharge
into a watercourse. A plan view of this
system can be seen in Figure 1.
A clay soil overburden at the site negated the
requirement for a liner. After excavation, the
native soil was compacted to reduce the
potential for the wastewater percolating into
the subsoil.
Wildlife
Wetland wildlife observed during summer
1994 include insects such as predaceous
diving beetle larvae, adult dragonflies and
damselflies, water boatmen, dragonfly and
damselfly larvae, backswimmers, midges,
riffle beetles, and whirligig beetles;
amphibians including the young of the year
American toads; and mammals including
muskrats. The muskrats severely damaged
the wetland vegetation in the first year, and
their burrowing activities caused short-
circuiting between sections of the serpentine
wetland path. Muskrats have been controlled
by trapping them and by reducing the
wetland water level in the fall (to expose the
entrance to the muskrat home) and main-
taining the low water level during winter.
Project Involvement
This project was a cooperative effort. The
landowner provided the property, partial
capital funding, and has the responsibility
for the ongoing maintenance. The Essex
Region Conservation Authority designed the
treatment wetland system, oversaw the
construction, implemented the monitoring
program, coordinates the research, and
provided materials for and installed the
monitoring equipment. Funding for the
wetland construction was provided by
Agriculture Canada's Rural Conservation
Clubs Program and Harrow Research Station
Environmental Assessment Review. The
MOEE provided technical support for the
data logger monitoring, staff gauges, rain
gauge, and redox meter. The covered
manure storage was funded through the
MOEE program Clean Up Rural Beaches
(CURB), and Canada Trust funded the
groundwater monitoring equipment (water
level meter, supplies, and materials for seven
piezometers). Ontario Ministry of
Agriculture and Food (OMAF) provided
support for the groundwater investigations.
Centralia College funded and conducted a
soils and groundwater assessment.
The Association of Conservation Authorities
of Ontario provided a facffitation/research
role among the various partners to establish
treatment wetlands in several Ontario
Regions. The association was also actively
H-26
-------
involved in the design and construction
supervision of the systems and are currently
monitoring the systems.
In January 1996, the funding for these
projects was severely cut in many locations.
In the future, these projects may not receive
the level of effort required to continue to
gather long-term monitoring data as these
systems mature. The project managers
believe that tracking these systems is
extremely important, and they are searching
for innovative sources of funding and
analytical support that will allow them to
continue to collect performance data.
Monitoring Program
An intensive monitoring program is
underway with dataloggers recording water
levels in the storage pond and wetland, water
and ground temperatures, and rainfall; and
frequent sampling of the water quality
throughout the treatment wetland system, A
total of four deep piezometers and three
shallow piezometers were installed around
the site to monitor potential groundwater
contamination. Sampling frequency for
groundwater monitoring is one sample per
sampling location during each of the months
of January, April, July, and October. Results
of the groundwater samples collected to date
have not been reported.
Operational performance
Table 1 summarizes monitoring data
collected during the first 9 months of
operation, April to December 1994.
Monitoring data from May to November
1995 are presented in Table 2. The data are
typical of early operating results reported by
others.
Future Direction
The Essex Region Conservation Authority
will be completing a summary report of the
Essex project in 1997. Each of the Conser-
vation Authorities involved in a treatment
wetland will prepare similar summary reports
of their treatment wetland systems. A
committee has been established to secure
further funding for long-term data collection,
summarize the effectiveness of the Ontario
systems, consider the applicability of
treatment wetlands for high strength
livestock wastewater, and recommend
improvements for future installations. The
committee, called the Agriculture Waste-
water Treatment Group, is made up of the
Conservation Authority representatives and
the Ontario Ministry of the Environment and
Energy.
Conclusion
The Essex treatment wetland system has
significantly reduced the contaminant loading
from the farm to Woltz Creek. The
Conservation Authority and the Ontario
Ministry of the Environment and Energy are
aware of the potential this treatment
technology provides to the livestock farming
community and are committed to further
research. The owner of the farm has seen the
positive environmental effects of the
treatment wetland and has become a strong
proponent of this technology to friends,
neighbors, and the news medk.
11-27
-------
Table 1. Treatment wetland monitoring data for a dairy in Exxex, Ontario, Canada; April to
December, 1994.
Parameter
BODj
TSS
NOj-N
TKN
TP
Dissolved P
Conductivity
Chloride
Fecal
conforms
E. coli
Units
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
pmhos/cm2
mg/L
No^lOOmL
No./100mL
Avg. wetland inflow
concentration
357
1,596
0.19
119
25
11.5
3,091
293
1,030,000
220,600
Avg. wetland outflow
concentration
202
48
0.12
17.5
3.9
2.3
1,225
182.5
11,999
11,343
% concentration
reduction
43
97
37
85
84
80
60
38
99
95
Table 2. Treatment wetland monitoring data for a dairy in Exxex, Ontario, Canada; Geometric
mean concentrations for May to November 1995.
Parameter
BOD,
NHrN
Total PO4
TSS
E.coli
Units
mg/L
mg/L
mg/L
mg/L
NoJIOO mL
Transfer pump
to storage pond
487
50
26
332
149,267
Wetland inflow
concentration
68
12
12
151
1,208
Wetland outflow
concentration
26
2.4
3,7
104
409
% Concentration
reduction
62
80
69
66
66
H-28
-------
Transfer Pump
O
O
Legend
O Trees
A Data Logger
C Bain Gauge
/ Sonde and
/ Temperature
J Probe
Figure 1: Plan View of Essex Treatment Wetland System
n-29
-------
Oregon State University Dairy Wetland
James A. Moore, PhD. and Steven F. Niswander*
Background
Six wetland ponds were constructed at the
Oregon State University campus dairy and
began receiving wastewater in the fall of
1993. All of the ponds were loaded at the
same rate with diluted dairy wastewater.
This diluted wastewater was still "high"
strength, with biochemical oxygen demand
and ammonia concentrations of
approximately 700 and 130 mg/L,
respectively. This is much higher than
recommended loading rates for domestic
wastewater (US EPA, 1988). However, we
were interested in looking at the highest
mass removal we could achieve. This would
then reduce the amount of manure nutrients
that would have to be land applied. This
objective is slightly different than most
domestic wastewater wetlands that are
designed to achieve some minimum outlet
concentration. Samples of the influent and
effluent were collected from all wetland cells
twice a month. Analyses of thirteen water
quality parameters were conducted. These
included chemical oxygen demand,
biochemical oxygen demand, total solids,
total suspended solids, total phosphorus,
orthophosphorus, total KjeldaM nitrogen,
ammonia, fecal coliforms, pH, conductivity,
dissolved oxygen, and temperature.
Site description
The Animal Sciences Department at the
Oregon State University campus operates a
working dairy with 140 mature milking cows
and an accompanying complement of young
stock and calves. The manure is handled with
a recycling flush system. The solids are
removed using a stationary screen separator,
and the liquid manure is held in an above
ground storage tank. A second smaller
storage tank is used for dilution of the
wastewater before pumping it to the six
wetland cells (Figure 1). The ponds were
loaded at approximately 3.9 cm/day (1.5
in/day) to achieve a 7.7 day retention time.
Each wetland cell is 26.7 m (87.6 ft) long,
5.5 m (18 ft) wide, and 0.3 m (1 ft) deep.
Cells 4 and 9 have deep (1m) center sections
while the others have a flat bottom, sloping
0.5% toward the outlet (Fig. 1). Two
treatment cells were planted with cattail
(Typha latifolia) and four were planted with
hardstem bulrush (Scirpus acutus) in 1992 as
shown in Table 2. Nursery rhizome stock
was planted in a 1.0 x 0.6 m (3.3 x 2.0 ft)
Table 2. Principal types of vegetation in each
of the cells.
Cell Number
4
5
6
7
8
9
Type plants
Cattail
Cattail/grass
Bulrush/grass/cattails
Grass
Bulrush/other species
Bulrush
* Professor and Head, Bioresources Engineering Department, and graduate research assistant,
respectively; Oregon State University, Corvallis, OR.
11-30
-------
pattern. After the plants were established and
the pond filled, nutria (Myocaster coypus)
destroyed most of the plants, A welded wire
fence and electric fence wire were con-
structed around the research site to limit
nutria access. The ponds were than replanted
in the spring of 1993. There has been no
noticeable damage from nutria since the
installation of the fence. By the fall of 1994,
most of the ponds had become a mix of
original species and invader species (Table
2),
The wetland eels are constructed in an area
with Amity silty clay and Bashaw clay loam
soils. Soil depth averages 60 cm (24 in)
throughout the site. Soil profiles show a
poorly drained mottled clay layer at 60 cm
(24 in). Cell bottoms are just above the
surface of this clay layer, except for the
center sections of cells 4 and 9, where the
deep sections enter about 60 cm of the clay
layer. Topsoil was not used in the bottom of
cells as the Amity and Bashaw soils were
adequate for establishing wetland plants.
Bottoms of cells are compacted Bashaw clay
with an estimated hydraulic conductivity of
less than 1 x 10"7 m/sec. The above ground
and above water level berms are compacted
Amity soil with a conductivity of 1 x 10"6
m/sec, which is within the recommended
conductivity for wetland systems (US EPA,
1988).
Results and Conclusions
The wetland reduced the concentrations of
all parameters by an average of 65%. The
lowest reduction was 48% for both total
phosphorus and ammonia. The greatest
reduction was 94% for fecal eoliforms. Table
3 is a summary of the average percent
reductions for all of the parameters.
The deep center sections in ponds 4 and 9
did not show any significant impact on the
treatment efficiency of the water quality
parameters evaluated. While treatment
differs by plant species, the treatment
differences in wetland cells with mixed plant
populations appears to be small.
The phosphorus removal needs to be studied
for a longer time to confirm long term
removals. The constructed wetlands at the
Oregon State University dairy also appeared
to be oxygen limited (Niswander et aL,
1996). The wetland cells are currently being
loaded at a much lower loading rate, which
does not deplete all of the oxygen. The
retention times have also been varied to find
the optimum retention times for achieving
the lowest outlet concentrations. This
information along with our previous findings
will alow us to make recommendations on
the optimum loading rates and retention
times for achieving both the maximum mass
removal and the lowest outlet
concentrations.
References
Niswander, S. F., J. k Moore, M. J.
Gamroth, and S. M. Skarda (1996). Treating
dairy flush water in a constructed wetland.
Proceedings of the 1995 James A. Vomocil
Water Quality Conference: Wetlands, Best
Management Practices, and Riparian Zones,
SR 957, Oregon State University Extension
Service, pp. 2-23.
U. S. Environmental Protection Agency
(1988). Design manual on constructed
wetlands and aquatic plant systems for
municipal wastewater treatment
(EPA/625/1-88/022).
H-31
-------
Tabte 2. Estimate of percentage cover by each species October 4,1994.
Pond
Number
4
5
6
7
8
9
Typha
latifotta
30
30
30
10
35
0
Scirpus
acutus
5
10
20
5
20
20
Grass*
20
60
50
80
5
5
Hydrocotyl
ranunculoides
0
0
0
0
30
0
Lemna spp.
45
0
0
5
10
75
*Glyceria occidentalis and Alopecurus geniculatus. (Both species appeared floating.)
Table 3. Summary or dairy data.
Parameter
BOD$ (rag/L)
COD (mg/L)
Total solids (mg/L)
TSS (mg/L)
TP(mg/L)
Ortho-P (mg/L)
NH3+NH4-N (mg/L)
D.O. (mg/L)
Conductivity
(fimhos/cm2)
Temperature (°C)
pH (std. units)
Inlet
705
1,628
1,195
542
33
167
126
1.08
2,279
10.6
7.5
Outlet
242
655
852
142
17
86
65
0.23
1,644
9.3
7.1
% Reduction
66
60
61
74
48
49
48
94
—
—
—
n-32
-------
To Main Dairy
Storage Tank
Wetland Storage
Tank
Outlets
«*-
^_
•*-
•*-
•*-
^_
L»,
I |
Plant Propogation
1 2
Deep Water
Zone
Ponds
3
4
' . ' 5
Treatment Cells
7
,
8I
Deep Water
Zone
Storage Pond
1
0 25 50 75
I | I
9
•
10
_J
100 feet
0 12.5 25 meters
Inlets
Figure 1. Site map of Oregon State University dairy wetland research site.
11-33
-------
Auburn University Constructed Wetlands for the Treatment of
Poultry Lagoon Effluent - A Case Study
D.T. Hffl and J.W. Rogers*
Introduction
Three series of dual cell free-water-surface
constructed wetlands were installed at the
Poultry Unit of the Alabama Agricultural
Experiment Station in Auburn, Alabama,
during the summer of 1992. The first two
series of cells were vegetated while the third
was left void of plants to serve as a blank.
Two pair of cells were vegetated and one
pair served as a control (far right in photo).
Each cell was 5,5 m wide, 10.5 m long and 1
m in depth. During construction, clay was
brought in to line the bottom of the cells to
reduce seepage. The constructed wetlands
were supplied water from the three lagoon
wastewater treatment System at the Poultry
Unit (Figure 1). Valves were located
between the third lagoon and the wetlands to
allow a choice of wastewater from the first
or third lagoon.
On October 15,1992, the first shipment of
plants arrived. Sagittaria landfolia was
sprigged in Series 1 and Phragmites
australis was sprigged in Series 2.
Wastewater from the third lagoon was then
applied to maintain a water depth of 7.5 cm.
During the winter of 1992-93 all of the
plants perished due to the lateness of
planting and the extremely cold weather in
January. In early April, Series 1 and 2 were
completely replanted. Series 1 once again
received Sagittaria landfolia and Series 2
received Scirpus spp. in addition to the
Phragmites australis. Water was supplied
from the third lagoon to maintain a depth of
10 cm. The water level was gradually raised
to 30 cm near the end of May. In mM-July
the influent source was changed to the first
lagoon to provide a heavier nutrient loading.
In addition to the three large wetlands, two
series of scaled down "model" wetlands were
constructed in the first blank cell. Each cell
of the model wetlands measured 0.6 m wide,
3.4 m long and 0.3 m deep. Wooden dowel
rods, 1.2 cm in diameter and 40cm in length,
were driven into the clay soil to provide an
inert growth medium. One series of the
model ponds received 5% (by volume) dowel
fill and the second received a 10% fill. The
effluent from these model ponds was piped
to the influent of the second blank celL Table
1 shows the fill rates for plants and dowels in
the respective cells or series.
Soil water lysimeters were installed in the
first and second cells of the first three series
at depths of 0.6 m and 1.3 m to monitor the
concentration of chemical parameters present
in the soil percolate. The lysimeters were
constructed of a white PVC tube with an
outside diameter of 4.8 cm and lengths of 1.2
Professor and Graduate Research Assistant, respectively, Department of Agricultural Engineering,
Auburn University, AL 11-34
-------
Table 1. Cell series with type of plants or
dowels and percent fill in each.
Series No.
1
2
3
4
5
Plant or
dowels
Sagittaria
Sagittaria +
Phagmites
Blank
Dowels
Dowels
% fiH with
plants or
dowels
-10
<~S •'
0
-10
~s
m or 1.8 m. A porous ceramic cup was
attached to the bottom end of the PVC pipe.
Lysimeter samples were collected once a
month beginning August 6, 1993.
Beginning August 6,1993, wastewater
samples were collected twice a month until
March 9,1994. A 500 ml grab sample was
taken from the influent stream of the five
series. A 500 ml sample was also taken from
the effluent of each of the five series. COD
and BOD5 analyses were performed by
Standard Methods for the Examination of
Water and Wastewater (APHA, 1992). TKN
and NH4 analyses were performed according
to the methods of AOAC (1984).
Results and Discussion
This study was initially scheduled to run
from August 1993 until June 1994 to
investigate the potential for treating dilute
poultry waste as weH as the variation in
treatment levels during the winter months.
Due to an instability in the supply of
wastewater to the wetlands, it was
involuntarily halted on March 9,1994. The
data that were collected should provide
necessary information for a preliminary
investigation of changes in the treatment
levels due to seasonal variations. It should be
noted that the poultry research unit added
considerably more water than would be used
by the typical poultry producer. In addition,
some spring water was seeping into the
lagoons. For these reasons, the lagoon
wastewater was much more dilute than that
of most poultry lagoons.
All series of wetlands were operated at a
constant depth of 30 cm during the period of
the study. A constant hydraulic loading rate
of 3.1 cm/day was applied to each series of
wetlands. This resulted in a COD and TKN
loading rate of 145 and 30 kg/ha-day
respectively. In order to determine detention
time and also to validate the plant fill, a
water column displacement test, where the
volume of water present in a column of the
pond is determined and compared to the
theoretical volume had the plants not been
there, was performed on the vegetated Series
1 and 2. The expected values of 10% and 5%
were surprisingly close to the measured
values of 10.7% and 6.7% for Series 1 and 2,
respectively. The detention time was
calculated to range from 11.1 to 11.6 days in
the three series.
Examining the data in Table 2, it is evident
that constructed wetlands are capable of
treating dilute poultry waste. The vegetated
series (Series 1 and 2) provided better
treatment than the dowel series (Series 4 and
5). Series 3, the blank, performed equivalent
to the vegetated series in all aspects except
COD removal.
The initial expectation was for the vegetated
series to outperform the blank. This however
did not happen. It is believed that the age of
H-35
-------
Table 2. Removal efficiencies (%) during the entire study period.
Series No.
1
2
3
4
5
BOD5
49.8
45.7
48.5
30.2
35.2
COD
60.7
61.5
54.3
28.0
39.8
TKN
42.8
56.7
45.7
20.3
27.1
NH4
37.6
52.8
44.3
19.9
26.4
P04
36.8
34.0
36.7
8.0
20.4
K
28.4
12.4
8.8
14.5
22.3
the system is responsible for this. Like most
biological systems, constructed wetlands
require a period of time to become
established. Litter and stems within the
column form emersed solid surfaces which
enhance bacterial growth. According to
Kadlec and Knight (1996), the biological
activity on these surfaces represents the
principle mechanism for treatment within the
constructed wetland.
Thus, the immature system that was the
subject of this study had a relatively
undeveloped "biofilm" which is needed to
provide high levels of treatment. Treatment
levels should increase as the constructed
wetlands get older and the plants and litter
spread to fill in void areas. As the system
matures, microbiological activity and, thus,
treatment efficiency, will increase. The
system which served as the blank series is a
much less complex aerobic to facultative
system containing algae and suspended
microorganisms. Since algae are fast
growing, a system such as the blank should
yield better treatment sooner after
establishment.
To investigate the difference in treatment
levels, the total study period was broken
down into five seasonal periods. The first
period began August 1 and ended October
31, 1993. This period is representative of the
warm season. November 1 started the cool
period with the first hard frost that killed the
vegetation abo"v&the water line. Period 2
was defined as beginning November 1 and
ending December 1. During this period it
was too cold for continuous plant growth
but warm enough during most days (ie.,
water temperature > 10° C) to sustain
bacterial growth. The third period, referred
to as the cold period, started December 2
and ended February 3. During this period the
average water temperature was below 10° C.
The fourth period began February 4 and
continued until March 9 when the project
was halted due to the interruption in the
influent wastewater supply. This fourth
period represented the recovery stage when
the water temperature rose above 10° C, yet
plant growth had not restarted. A fifth period
was planned to continue until June when the
water was warm and the plants were
growing. However, due to wastewater
supply problems only the first four periods
were completed.
Examining Figures 2 and 3, graphical
representations of COD and BOD5
reductions for the four study periods, it is
clear that the ponds containing plants
provided better treatment than the dowel or
blank series. This strongly suggests that live
H-36
-------
wetland plants provide some manner of
growth enhancement in the bacteria
responsible for the removal of organic
matter. Period 2 produced mixed results with
regard to COD and BODS reductions. Series
1 had a slight increase in treatment efficiency
while Series 3 showed a dramatic increase
from 50% to 70%. Series 2,4 and 5 showed
decreases in treatment performance. As the
temperature decreased during period 2,
microbial activity should have slightly
decreased causing treatment to also
decrease. If the plants played an active part
in the treatment process, the decrease in the
treatment levels in the vegetated series
should have declined more after the first
frost than the dowel rod series. Series 2 (5%
plants) did drop in performance. Series 1
(10% plants) sHghtly increased in
performance. This series contained
Sagittaria which was killed back immediately
following the first frost. However, the plants
did start growing back until the next frost
killed them again. This cycle of growth and
die-back continued until the middle of
December when complete death occurred. It
is possible that these plants continued to play
a role in the treatment process white
re-growing.
The treatment in the dowel series also
decreased during the third period. The 10%
fill series suffered a larger decrease than the
5% fill series. This would be expected since
the 10% series contained a larger population
of microorganisms due to the amount of
media surface area present. As the water
temperature decreased, more bacteria were
present to slough off in Series 4 causing a
higher organic loading than in Series 5.
Period 4 provided some interesting results as
the COD removal increased slightly in the
vegetated series. In the dowel rods,
however, treatment levels increased greatly
from 8.2% to 28.9% and 37.9% to 52.6%
for Series 4 and 5 respectively. The BODj
removal was different. BOD removal
dropped sharply in Series 1 and 2 while
Series 4 and 5 showed dramatic increases in
treatment levels. It is possible that as the
water began to warm, dead plant material
present in Series 1 and 2 began decomposing
adding an additional organic load to the
system. Also, the dowel rods continued to
present a solid attachment site while the
decomposing plants resulted in a loss of
attachment sites.
The TKN and NH4 removal data shown in
Figures 4 and 5 revealed few surprises. Once
again the vegetated and blank series
provided better treatment during the first
periods. During the fourth period, the TKN
and NH4 removal dropped in Series 1, 2 and
3. The decrease in organic removal during
this period suggests a low oxygen content
which could also inhibit the nitrification
process and significantly decrease nitrogen
removal through the nitrification-
denitrification pathway. The performance of
the dowel series showed a large increase,
possibly due to a solid point of attachment
which the plants could not offer.
The concentration data for the chemical
parameters in the lysimeter samples are
contained in Table 3 and most were within
acceptable ranges with no cause for alarm,
except the NO3 levels. High nitrate
concentrations may cause the health problem
methemoglobinemia, better known as
bluebaby syndrome. The phosphorus
concentration in the lysimeter samples did
not differ statistically among the different
depths or cells. Series 1 was statistically
H-37
-------
different than Series 2 and 3. The difference,
however, was very small (<0.5 mg/L and
could have been due to the plant species or
variation in initial soil concentrations.
Examining the nitrogen parameters in the
lysimeter samples, several interesting trends
are noted. Series 3 had significantly higher
average concentrations of TKN than Series 1
or 2. This was expected since the wetland
plants provide oxygen to the root zone, thus
promoting nitrification. The NO3
concentration also varied with plant volume.
Series 1, containing 10% plants, had a mean
NO3 concentration of 21.1 mg/L-N while
Series 2, containing 5% plants, had a lower
concentration of 14.9 mg/L-N. Series 3, the
blank cells without any plants, had the lowest
concentration of NO3 at 12,2 mg/L-N. TKN
also showed a significant decrease going
from the 0.6 m depth to the 1.3 m depth in
Series 2 and 3. There were also no
significant differences for the nitrogen
parameters between the first and second cells
in each series.
Conclusions
It should be stated that as constructed
wetlands are biological systems, a period of
time is required for the ponds to become
fully established. In previous research by
DeJong (1976), starting effects were found
to be present until the third year of
operation. Due to funding restraints, this
study had to take place the first year the
wetlands became operational However,
these preliminary conclusions shall be the
basis of future research at this site.
These constructed wetlands have shown the
capability of providing treatment to dilute
poultry wastewater within the first year of
operation. BOD5 removal was almost 50%,
while COD removal varied from 40% to
50%. Overall, the treatment levels in the
vegetated series were not much greater than
in the blank series. However, the vegetated
systems should show increased levels of
treatment as they become better established.
The effect of plant presence on wastewater
treatment during the cold winter months was
difficult to determine due to the variation of
the influent quality. The most obvious effect
occurred in the early spring when the
temperature began to rise. During this time,
the treatment provided by the vegetated
series decreased while treatment in the dowel
series increased. This is believed to be
caused by the decomposition of dead plant
material which might drive the system into an
anaerobic state, therefore hurting
performance. This suggests that the
harvesting of the plants during the winter
months may improve the quality of the
effluent from this type of system.
The lysimeter data raises concerns about the
levels of nitrates present in the groundwater
below these systems. Further studies should
be conducted to determine if this is a site
specific problem or a problem inherent with
this design of wetlands systems.
References
AOAC (1984); Official Methods of Analysis:
Association of Official Analytical Chemists.
New York.
APHA (1992); Standard Methods for the
Examination of Water and Wastpwater:
American Public Health Association,
Washington, DC, 18th edition.
H-38
-------
DeJong, J. (1976); The purification of Kadlec, R, H. and R. L. Knight (1996);
wastewater with the aid of rush or reed Treatment Wetlands: Lewis Publishers, Boca
ponds; In ljo|og|cal Control of Water Raton, FL.
Pollution, ed. Joachim Tourbier and Robert
W. Pierson; University of Pennsylvania
Press, Inc.
n-39
-------
Valve Center
Lagoon 1
Lagoon 2
Cell! Cell 2
Series 1
Series 2
Series 3
^Series 4&5
Figure 1. Overview of lagoon and constructed wetland treatment systems.
! Pencdl E2 PotocJZ S3 PencaO SI Pund4
A
\\
^
!S
§
•
s
CM co v m
to co co co
03 CO CD 03
d i~ fc~ I™
(U
-------
Constructed Wetlands
for
Animal Waste Treatment
Appendix A
Questionnaires for
the Case Histories
-------
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or
or cooperating researchers:
a. Name: Dr. Frank J. Humenik
Address: N.C. State University
Bio. & Acr. Encr. . Box 7625
Raleiah. NC 27695-7625
Phone: 919/515-6767 Fax:919/515-6772
Your involvement/experience with this wetland system:
b. Name; Dr Patrick 'G. Hunt -
Address: USDA-AES
2611 West Lucas Street
Florence. SC 29501-1241
Phone:803/669-52Q3 Fax: 803/669-6970
Your involvement/experience with this wetland system;
Others included on separate sheet: Ic. 8 Id. H
2. Location of the system (description or address):
Duplin County, North Carolina
Location in terms of latitude and longitude: Lat:35° 02
Long: 77° 57'
Approximate percentage of funding for this project was
provided by:
EPA 74.2% State Water Quality Agency % TVA %
USDA 12.3% State Experiment Station 10.1%
Other: Private sources: Murphy Family Farms 3%
NC Pork Producers Assoc. Q.
A-l
-------
Ic. Name: Mark Rice
Address North Carolina State University
Bio. & Acr. Encr. , Box 7625
Raleiah. NC 27695-7625
Phone: 919/515-6794 Fax: 919/515-6772
Your involvement/experience with this wetland system:
Id. Name: Dr.Maurice Cook
Address North Carolina State University
gojl Science Dept.. Box 7619
Raleiah. NC 27695-7619
Phone: 919/515-7303 Fax:919/515-2167
Your involvement/experience with this wetland system:
le. Name: Dr. Stephen_.Broome
Address North Carolina State University
Soil ScienceDept.. Box 7619
Ba-1-eierh^ NC 27695-7619
Phone: 919/515-2643 Fax: 919/515-2167
Your involvement/experience with this wetland system:
If. Name: Dr.Ariel Szoai
Address USDA-ARS
Soil, Water .&. Plant Res. Ctr
2611 W. Lucas Street, Florence. SC 29501
Phone: 803/669-5203 Fax:803/669-6970
Your involvement/experience with this wetland system:
A-2
-------
Ig. Name: George Stem
Address:USDA-Natural Resource Conservation
Service, 4405Bland Road, Suite 205
Raleigh, NC 27609
Phone: 919/873-2102 Fax: 919/873-2156
Your involvement/experience with this wetland system:
Ih. Name: Mike Sugg
Address USDA-NaturalResource Conservation
Service, Kenansyille Field Office, PO Box 277
Kenansville, NC 28349
Phone: 910/296-2120 Fax:910/296-2122 •
Your involvement/experience with this wetland system;
li. Name: Gary Scalf
Address Murphy Family Farms
P. O. Box 759
Rose Hill. NC 28458
Phone: 910/289-2111 Fax:
Your involvement/experience with this wetland system:
A-3
-------
4. Animal information:
Type: s_win.e Number: 260,,0 Avg. Wgt. 25.
Type:, , Number: Avg. Wgt.
Type; Number: Avg. Wgt.
Type: Number: Avg. Wgt.
Type: Number: Avg. Wgt.
5. Date wetland system was installed: yr.1992 month August
6. Date system became operational (first waste discharge):
yr 1993 month June
7. Type of pretreatment (check one or describe):
One cell lagoon X Two lagoon cells
Waste Storage Pond (as defined by SCS or NRCS):
Settling Basin
Other:
8. Approximate age of pre-treatment units when wetland became
operational: 2 yrs.
9. Had sludge ever been removed from the pre-treatment unit
prior to installation of the wetland system? yes ; yi
prior to installing wetland. No x
10. Basis for design of wetland system (i.e., 90 Ibs.
BOD/ac/day; minimum 7 day detention time; etc.) If a
wetland/pond/wetland system was used, include design basis
for pond and for entire system:
Nitrogen loading of 3 kcr/ha/dav
11. Whose design criteria did you use? (TVA, SCS, private
industry, etc.) List any references:
SCS^ National Bulletin Mo. 210-1-17
12. Total surface area (water surface) of wetland cells:
720 m2 or _ acres
13 . Total surface area of pond cells : n_a _ m2 or na acres
14, Surface area of the entire system including embankments
3 .goo m2 or _ acres
15, Average water depth: 7 . 6 cm
16. Average slope of cells from upstream to downstream: 0 .2
A-4
-------
17. Provide a sketch in-plan view 'showing the cells,
embankments, and general location in relation to
, pretreatment units and other significant features. Provide
dimensions for the cells and embankments. If preferred,
attach engineering drawings from a publication and note
here.
SEE ATTACHED FIGURE
18. Was the waste flow diluted prior to being discharged into
the wetland system? No Yes x
Source of dilution water was: Well
Was dilution ratio changed during the course of the study?
Yes x dilution ratio became % fresh water and
wastewater
Data in table, Items 25 and 26 is for initial dilution
aygrage o£ 10,;.! and 6:1 dilution ratios
final dilution
(The initial dilution ratio was 10:1 in 1993. It was
changed to approximately 6:1 in 1994 and in February 1996
was changed to 2:1.
19. Avg. Daily warm season flows (non-storm flows):
System 1: Influent 2900 liters/day
System 1: Effluent 2230 liters/day
System 2: Influent 3400 liters/day
System 2: Effluent 2050 liters/day
20. Avg. daily cool season flows (non-storm):
System 1: Influent 2700 liters/day
System 1: Effluent 1720 liters/day
System 2: Influent 3250, liters/day
System 2: Effluent 2360 liters/day
21. Describe flow controls at influent end of system (i.e.,
valves, swivel pipe, pump on timer, etc.)
valves
22. Any problems with clogging of influent pipes/valves with
debris, struvite, animals, etc.?
There has been some clocrcfincr of the inlet valves.
Methods of control, if applicable:
Valves are flushed pn a regular basis to inhibit buildup
A-5 ' ••
-------
23. Describe water level controls in the cells (i.e., downstream
swivel pipe): Adjustable elbows between series cells.
acHustable weir at the effluent end.
24. Any clogging of pipes between cells or from final outlet
cells? No
Methods of control, if applicable:
A-6
-------
25. Provide average concentrations in mg/L (or other appropriate units) for
influent into the system and effluent from the system for warm (April-
October) and cold (November-March)
Average for system 1
Parameter
W = warm season
C = cold season
Average Concentration or Reading
Influent Effluent • Percent Change
BODS(W)
%
**? C
BOD5(C)
COD(W)
75
64
14.7
COD(C)
115
71
38.3
NH3+NH4-N(W)
NH3+NH4-N(C)
Org-N(W) TKN
41
92.7
Org-N(C) TKN
78.5
10.5
86.6
Total P (W)
Total P (C)
PO4-P (W)
7.5
2.6
65.3
PO4-P (C)
13.5
8.5
37
TSS (W)
TSS (C)
D.O.(W)
N/A
N/A
N/A
D.O.(C)
N/A
N/A
N/A
Fecal coliforms (W)
(No/100 ml)
Fecal coliforms (C)
No/100 ml)
Fecal strep. (W)
N/A
N/A
N/A
Fecal strep. (C)
N/A
N/A
N/A
PH (W)
(standard units)
PH (C)
Water temp. (W)
(deg. C)
N/A
N/A
N/A '
Water temp. (C)
N/A
N/A
N/A
A-7
-------
26. Provide same information, for only theupstream cell(s). If a bank of
cells is used, provide average data for that bank of upstream cells.
ttp»tream call (a)
System 1
Parameter
W «= warm season
C » cold season
BOD5 (W)
BOD5(C)
COD(W)
COD (C)
NH,+NH4-N (W)
NHj+NH<-N(C)
Org-N(W) TKN
Org-N(C) TKN
Total P (W)
Total P (C)
PO4-P (W)
PO4-P (C)
TSS (W)
TSS (C)
D.O. (W)
D.O. (C)
Fecal coliforms (W)
(No/ 100 ml}
Fecal coliforms (C)
No/ 100 ml)
Fecal strep. (W)
Fecal strep. (C)
pH (W)
(standard units)
PH (C)
Water temp. (W)
(dag. C)
Water temp. (C)
Average Concentration or Reading
Influent Effluent Percent Change
;""•" "
*. 'V.^L ..... '
t-* ' ,"ir r, " " ,*
75
114.5
rv '"33* ' "
L *\"> '
N/A
N/A
\ i
f
f" '* f A ' ' 1
f/; , "
7.5
13.5
N/A
N/A
W« ^. •, •-•,,- : ... *{
!! "- ,'- " _-':,
N/A
N/A
'" '•' ,8 '
:: 8',
N/A
N/A
'
^' ! . ^
38
52
8,5
•
N/A
N/A
-*
' _ - ,
5
10.5
' 4*
-
N/A
N/A
N/A
N/A
7.7
7,7
N/A
N/A
i ^ , .
>
49.3
54.6
74:2
^ ""
N/A
N/A
'
33.3
22.2
_ -
N/A
N/A
'-
N/A
N/A
3.75
3.7,5
N/A
N/A
27. For the data in the above tables, what was the period of observation?
mo. 5 yr. 93 to mo. 10 yr. 94
to mo. 10
A-8
-------
25. Provide average concentrations in mg/L (or other appropriate units) for
influent into the system and effluent from the system for warm (April-
October) and cold (November-March)
Average for system 2
Parameter
W = warm season
C = cold season
Average Concentration or Reading
Influent Effluent Percent Change
BOD5(W)
BOD5(C)
COD (W)
75
64
14.7
COD(C)
75
81
8% increase
NH3+NH4-N(W)
NH3+NH4-N(C)
Org-N(W) TKN
41
3.5
96.9
Org-N(C) TKN
78.5
9.5
87.9
Total P (W)
Total P (C)
tn
PO4-P (W)
7.5
1.1
85.3
P04-P (C)
13.5
48.1
TSS (W)
TSS (C)
D.O.(W)
N/A
N/A
N/A
D.O.(C)
N/A
N/A
N/A
Fecal coliforms (W)
(No/100 ml)
Fecal coliforms (C)
No/100 ml)
Fecal strep. (W)
N/A
N/A
N/A
Fecal strep. (C)
N/A
N/A
N/A
pH (W)
(standard units)
pH (C)
Water temp. (W)
(deg. C)
N/A
N/A
N/A
Water temp. (C)
N/A
N/A
N/A
A-9
-------
26. Provide same information for only the upstream cell(s). If a bank of
cells is used, provide average data for that bank of upstream cells.
Upstream cell(s)
System 2
Parameter
W = warm season
C = cold season
BODS (W)
BODS (C)
COD (W)
COD(C)
NHj+NHt-NCW}
NHj+NH4-N(C)
Org-N(W) TKN
Org-N(C) TKN
Total P (W)
Total P (C)
PO4-P {W}
PO4-P (C)
TSS (W)
TSS (C)
D.O. (W)
D.O. (C)
Fecal col i forms (W)
(No/100 ml)
Fecal coli forms (C)
No/100 ml)
Fecal strep. (W)
Fecal strep. (C)
pH (W)
(standard units)
pH (C)
Water temp. (W)
(deg. C)
Water temp. (C)
Average Concentration or Reading
Influent Effluent Percent Change
75
114.5
* *' '33
N/A
N/A
v, .. .,
•sJ. » • -
7.5
13.5
N/A
N/A
i"> "• '• ;• v
N/A
N/A
, - „ 8__ _ , ,
8
N/A
N/A
35
50.5
4
1
N/A
N/A
2
9
N/A
N/A
N/A
N/A
7.8
7.8
N/A
N/A
53.3
55.9
87.9
- v "
N/A
N/A
.*".'- •'
73.3
33.3
N/A
N/A
' ' *"\
N/A
N/A
;'2.5 •- '
2.5'' -'
N/A
N/A
27. For the data in the above tables, what was the period of observation?
mo. 5 yr. £2
to mo. 10 yr.
A-10
-------
28. How many sampling days were included in the study? days
May 27, 1993 to March 31, 1995 (245 observations)
February 4, 1994 to March 31, 1995 (51 observations)
29. What were maximum and minimum influent concentrations for:
a. Ammonia {NH3 + NH4-N) : Max._JJJL_mg/l Min.0.7 mg/1
b. Total suspended solids: Max. M/A_mq/l Min. N/A mg/1
c. ' *Total phosphorus (TP) : Max. 22...- mg/1 Min. j.. 7 mg/1
d. Phosphate (P04-P5 : Max. 31 ma/1 Min. 0.8 mg/1
*February 4,1994 to March 31, 1995 (51 observations)
30. Which of the following represented a problem at your site
-{Elaborate as needed or place N/A if not applicable) :
a. Insects destroyed particular plant species:
N/A
b. Muskrats or other animals created problems:
N/A
c. Plants were killed at upper end of cells because
concentrations of ammonia or suspended solids were too
high:
; N/A
d. Evaporation rates were so high in summer that plants at
lower end of system were killed or stressed:
The hydraulic flow; rate is .increased during the summer
to maintain flow through the system,.
e. Discharge limits could not be met for certain
constituents in the final effluent:
List: The dischargelimitsfor phosphoruscould not
bemet at anytime,and the nitrogenlimits
only at low loading., rates during warm seasons.
f. Phosphorus concentrations in the final effluent tended
to increase over .time: .
g. Water management is a major problem. A water balance
should be developed for any new systems which include:
t^(l) drawing down lagoon levels in late fall to
accommodate winter storage of flush water and waste,
rainwater on the lagoon surface, and runoff water;
«^(2) controlling water released to wetland system
based on seasonal changes;
S (3) collecting water discharged from the system and
recycling or land applying;
v^(4) accounting for rainwater on the surface of the
wetlands A-11
-------
(comments on need for water balance)
h. Mosquitoes were a problem: ft.two-year study was
conductedby Mike_ Strinaham in the Entomology
Department at NCSU and found no significant population
increase due to the wetlands
I. Uneven distribution of wastewater across all cells in a
multi-cell system:
i. Others :
More thoughts on problems included on separate sheet a
31. What place do constructed wetlands have in managing animal
wastes? 'What are the drawbacks? Should they be permitted
for discharge?
The wetland should be a component in an overall waste
management system. Based on our experience the wetlands are
not able to consistently meet discharge requirements even at
low nitrogen loading rates.
32, What additional research is needed on these systems?
The exact role and function of the wetlands and their most
effective sequence in a waste treatment system need to be
determined.
A-12
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or cooperating researchers:
a. Name: Thomas A. McCaskey
Address: Dept. of Animal and Dairy Sciences
Auburn University
Auburn University. AL 36849
Phone: 334-844-1518 FAX: 334-844-1519
Your involvement / experience with this wetland system:
Principle Investigator
b. Name:
Address:
Phone: . FAX:
Your involvement / experience with this wetland system:
Others included on separate sheet: Ic.n Id. n.
2. Location of the system (description or address): Sand Mountain Agricultural
Experiment Station. Crossville. AL
Location in terms of latitude and longitude: Lat. Long.
3. Approximate percentage of funding for this project was provided by:
EPA 3 % State Water Quality Agency % TVA2£L% USDAJ_%
State Experiment Station 74 % Other
A-13
-------
4. Animal information:
Type: Swine Number:_5QQ&t Avg. Wgt. 150 Ib
Type: Number: Avg. Wgt.
Type: Number: Avg. Wgt.
Type: Number: .Avg. Wgt.
Type: Number: Avg. Wgt
5. Date wetland system was installed: yr. 1988 month fall
6. Dale system became operational (first waste discharge): yr 1990 month Nov
7. Type of pre-ireatment (check one or describe):
One cell lagoon Two lagoon cellsj
Waste Storage Pond (as defined by SCS or NRCS):
Settling Basin
Other:
8. Approximate age of pre-treatment units when wetland became operational: 1.2 yrs
9. Had sludge ever been removed from the pre-treatment unit prior to installation of the wetland
system? yesjL_ ; yrs prior to installing wetland. no x
10. Basis for design of wetland system (le., 90 Ibs BOD/ac/day; minimum 7 day detention time;
etc.) If a wetland / pond / wetland system was used, include design basis for pond and for entire
system: 65 Ibs BOD/ac/day. See attache diagram
11. Whose design criteria did you use? (TVA, SCS, private industry, etc.) List any references:
TVA - Don Hammer
12. Total surface area (water surface) of wetland cells: m2 or L_ acres
13. Total surface area of pond cells: m2 or 0.5 acres
14, Surface area of the entire system including embankments: m2 or _UL acres
15, Average water depth: 15.24 cm
16. Average slope of cells from upstream to downstream: < 1 %
A-14
-------
17. Provide a sketch in plan view showing all cells, embankments, and general location in relation
to pretreatment units and other significant features. Provide dimensions for the cells and
embankments. If preferred, attach engineering drawings or drawings from a publication and note
here.
See attached sketch
18, Was the waste flow diluted prior to being discharged into the wetland system? No_
Yes x , dilution ratio wasJJQ. % fresh water and 7Q % wastewater
Source of dilution water was: gravity flow from a farm pond
Was dilution ratio changed during the course of the study? No
Yes , dilution ratio became % fresh water and _% wastewater
Data in table, Items 25 and 26 is for initial dilution final dilution.
19. Avg. daily warm season flows (non-storm flows): NOTE: for all weather events
Influent 21.248 liters/day Effluent 14.146 liters/day
20. Avg. daily cool season flows (non-storm): NOTE: for all weather events
Influent 30.990 liters/day Effluent 48.770 liters/day
21. Describe flow controls at influent end of system(i.e., valves, swivel pipe, pump on timer, etc.)
Swivel pipe with flows checked daily '
22. Any problems with clogging of influent pipes/ valves with debris, struvite, animals, etc.?
Only occassionally
Methods of control, if applicable:
Hand
23. Describe water level controls in the cells (i.e.,downstream swivel pipe): ;
Downstream swivel pipe
24. Any clogging of pipes between cells or from final outlet cells? No
However, a rock filter, containing a buried pipe to collect and distribute effluent to the next
cell or storage pond, and the surface of this filter would need occassional cleaning
Methods of control, if applicable: Rake and hand cleaning
25. Provide average concentrations in mg/L (or other appropriate units) for influent into the
system and effluent from the system for warm (April - October) and cold (November - March)
seasons.
A-15
-------
Average for the system*
Parameter
W=warm season
C=eold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS (W) All seasons
DOBS (C)
COD(W)
319.9
64,2
79,6
COD (C)
Org-N (W)
73.7
12.2
83.4
Org-N (G)
Total P (W)
Total P (C)
PO4-P (W)
N/A
TSS
D.O, (W)
N/A
D.O. (Q
FecaJ coliforms
(Noj'lOOml)
94.6
wO ill, C
(Noj'lOOml)
Fecal strep.
4.7 x 104
1.9 x 103
96.0
pH(W)
(standard units)
Water tenip. CftO deg C.
14.9
* Data was not provided for warm and cool seasons; table is all seasons average for the system.
A-16
-------
26. Provide same information for onlythe upstream cell(s). If a bank of cells is used, provide
average data for that bank of upstream cells.
Upstream cell(s)
***data provided only for average and not by season***
Parameter
W=warm season
C=cold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS (W) all seasons
Kmmmsmm&mmzxxxxA
DOD5(Q
COD (W)
319.9
107,7
66.3
NH,+NH4-N (W)
62.S
Org-N (W)
18.1
6.4
64,6
Ofg-N(C)
Total P (W)
Total P (C)
PO4-P (W)
N/A
TSS fW)
D.O. (W)
N/A
D.O. (C)
Fecal cbliforms fW)
(NO./100 ml)
Fecal strep.
Fecal stfcp.
pH (W)
(standard units)
Water temp. (W)
(deg.C)
Air temp only, see
previous sheet.
Water temp. (C)
27. For the data in the above tables, what was the period of observation?
mo.,
yr. 90 to mo.
-------
28. How many sampling days were included in the study? days
29. What were maximum and minimum influent concentrations for:
a. Ammonia (NH3 + NH4-N): Max. mg/L Min. __mg/L
b. Total suspended solids: Max mg/L Min. mg/L
c. Total phosphorus (TP): Max mg/L Min mg/L
d. Phosphate (PO4-P): Max mg/L Min mg/L
30. Which of the following represented a problem at your site (Elaborate as needed or place N/A
if not applicable):
a. Insects destroyed particular plant species:
Yes. Cattapillars invaded cattail cells at different times.
b. Muskrats or other animals created problems:
Yes. Muskrats burrowed into embankments. One cell was drained though an
outside embankment. Control of muskrats was essential.
c. Plants were killed at upper end of cells because concentrations of ammonia or
suspended solids were too high:
No
d. Evaporation rates were so high in summer that plants at lower end of system were
killed or stressed:
NO :
e. Discharge limits could not be met for certain constituents in the final effluent:
List: If a discharge were allowed, limits could not.be met, at some time for many
Constituents.
f. Phosphorus concentrations in the final effluent tended to increase over time:
g. Water management is a major problem. A water balance should be developed for any
new systems which include:
(1) drawing down lagoon levels in late fall to accommodate winter storage of flush
water and waste, rainwater on the lagoon surface, and runoff water;
(2) controlling water released to wetland system based on seasonal changes;
(3) collecting water discharged from the system and recycling or land applying;
(4) accounting for rainwater on the surface of the wetlands.
(comments on need for water balance)
A-18
-------
h. Mosquitoes were a problem:
i. Uneven distribution of wastewater across all cells in a multi-cell system: No
j. Others:
More thoughts on problems included on separate sheet n
31. What place do constructed wetlands have in managing animal wastes? What are the
drawbacks? Should they be permitted for discharge?
32. What additional research is needed on these systems?
A-19
-------
Primary lagoon
i
Primary lagoon
*
Secondary
lagoon
•w
/
Farm poad
T3
a
o
ft
6J)
**M4
.3
s
/
Figure 1. Generalized flow diagram of the Sand Mountain Constructed Wetland
System at the Agricultural Experiment Station, Crossville, AL
A-20
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or cooperating researchers:
a. Richard Reaves
Biologist
SO/lnternational, Environmental Group
781 Neeb Road
Cincinnati, OH 45233-4625
Phone:513-922-8199 fax:513-922-9150
email: rreaves@cinternet.net
Involvement/experience with the project;
Water quality monitoring, data analysis, reporting to regulatory agencies,
outreach activities
b. Brain Miller
Coordinator, Marine Advisory Services
Illinois-Indiana Sea Grant Program
1159 Forestry Building
Purdue University
West Lafayette, IN 47907-1159
Phone:317-494-3586 fax:317-496-2422 . .
email: brian_miller@acn.purdue.edu
Involvement/experience with the project:
Reporting to regulatory agencies, outreach activities
c. Paul DuBowy
Associate Professor
Dept. of Wildlife & Fisheries Sciences
Texas A&M University
210 Nagle Hall
College Station, TX 77843-2258
phone: 409-845-5765 fax: 409-845-3786 '
email: p-dubowy@tamu.edu
Involvement/experience with the project:
Reporting to regulatory agencies, outreach activities, landowner-agency
coordination
2. Location of the system:
Tom Brothers' Dairy
5377 East 800 North ,
Syracuse, IN 46567
Location in terms of latitude and longitude:
Lat.:41°20'30" Long.: 85° 46'30"
A-21
-------
3. Approximate percentage of funding for this project was provided by:
EPA 319 funds administered by the Indiana Department of Environmental
Management funded approximately 90% of this project; Purdue University
provided the remaining 10% of the funding.
4. Animal Information:
Type: dairy Number: 80 Avg.Wgt: 630 kg
5. Date Wetland system installed:
yr. 1994 month: November
6, Date system became operational (first waste discharge):
yr.: 1995 month: May
7. Type of pretreatment:
A stack pad for drying manure coupled with a septic pit for solids removal from liquid waste.
8. Approximate age of pretreatment units when system became operational:
10yr.
9, Had sludge ever been removed from the pre-treatment unit prior to installation of the
wetland system?
Yes; 1 year prior to wetland installation.
10, Basis for design of wetland system:
The presumptive method for sizing wetland basin based on 65 pounds of BOD
per acre per day, modified by designing wetland basin to accommodate 25-year,
24-hour storm event.
11. Whose design criteria did you use?
SCS guidelines were used, the design was done by Indiana SCS State Engineer.
12. Total surface area (water surface) of wetland celts:
Cell 1 approximately 900 m2
Cell 2 approximately 950 m2
A-22
-------
13. Total surface area of pond cells:
approximately 1200 m2
14. Surface area of entire system including embankments:
approximately 4000 m2
15. Average water depth:
inflow to cell: 4 cm cell outflow: 20 cm
16. Average slope of cells from upstream to downstream:
0.25%
17. Provide a sketch in plan view showing all cells, embankments, and general location
in relation to pretreatment units and other significant features. Provide dimensions for
the cells and embankments. If preferred, attach engineering drawings from a publication
and note here.
Drawings are attached: a general schematic and various engineering drawings.
18. Was the waste flow diluted prior to being discharged into the wetland system?
No, but surface runoff from surrounding uplands (pasture and no-till row crop)
was diverted and directed into the first cell approximately 60% of the way from
the inlet to the outlet. Most times this water was cleaner than wastewater in the
cell and provided a dilution at the point of entry. Occasionally, the runoff
contained higher concentrations of various contaminants and the entry point
provided a form of step loading for the first cell.
19. Average daily warm season inflows (non-storm flows):
Influent: 750 L/d, milkhouse washwater and waste runoff.
Effluent: not measured, most of the growing season, there was no outflow resulting from
high evapotranspiration rates.
20. Average daily cool season inflows (non-storm flows):
Influent: 750 L/d, milkhouse washwater and waste runoff.
Effluent: not measured.
A-23
-------
21. Describe flow controls at Influent end of system:
Gravity flow by overflow from the septic pit into system piping. Switching valves
allow direction of waste flow into either of the wetland ceils or sequentially
through both cells. This allows continued operation if one cell is out of service.
Horizontal slotted PVC pipe provided inflow distribution across the width of the
cells.
22. Any problems with clogging of influent pipes/valves with debris, struvite, animals,
etc.?
Occasional blockage of slotted PVC pipe with debris.
Methods of control, if applicable:
manual removal of debris
23. Describe water level controls in the cells:
Bottom fed risers with adjustable 4-inch plastic depth regulators in each cell.
Surface overflow from holding pond into infiltration area
24. Any clogging of pipes between cells or from final outlet cells?
Occasional blockage of slotted PVC pipe that redistributed flow across the head
of cell 2 with debris (primarily tree leaves in fall). No problems with blockage of
outflow riser structures of pipes between cells.
Methods of control, if applicable:
manual removal of debris
25. Provide average concentrations in mg/L or other appropriate units for entire system
warm season (May-October) and cold season (November-April) influent and effluent:
I redefined warm and cold season for this system. Warm season does not start at this system
until May.
There are two things worth noting when examining these numbers.
1. There was a substantial difference in the quality of wastewater entering the system between
the two years. The total averages for the two years are probably lower than would be
expected for long-term operation because wetland influent quality during the first year was
exceptionally good. Influent quality during the second year was probably typical for a dairy
of this size,
2, Effluent values are given for the outflow of the second cell. The system has a holding pond
downstream of the second cell that can provide additional treatment after effluent leaves the
second cell. However, because the open pond attracts wildlife and provides an exceptionally
good environment for algal growth, it also can result in decreases in water quality after
leaving the second cell.
A-24
-------
Parameter
W = warm season
C = cold season
BOD5(W)
BODS (C)
COD (W)
COD(C)
(NH3+NH4)-N (W)
(NH3+NH4)-N (C)
Org-N (W)
Org-N (C)
Total-P {W)
Total-P (C)
PO4-P (W)
PO4-P (C)
TSS(W)
TSS (C)
DO(W)
DO (C)
Fecal eoliforms (W)
colonies per 1 00 mL
Fecal conforms (C)
colonies per 100 mL
Feca! Strep. (W)
Fecal Strep. (C)
pH(W)
, PH(G)
Water temp. (W)
Water temp. (C)
Average Concentration or Reading
Influent
372.08
696.83
not measured
not measured
74.36
192.18
13.25
25.32
13,31
19.16
8.94
8.32
81
74
0.99
2.70
96
222
not measured
not measured
7.58
7.26
20.71
6.67
Effluent
22.84
63.89
not measured
not measured
11.58
9.08
6.81
6.46
3,15
1.83
2.60
1.35
83
L_ 17
5.05
6.86
4
8
not measured
not measured
7.74
7.44
21.73
10.75
Percent change
-93.9%
-90.8%
not determined
not determined
-81.4%
-95.3%
-48.6%
-74.5%
-76.33%
-90.4%
-70.9%
-83.8%
+2.5%
-77.0%
+410.1%
+154.1%
-95.8%
-96.4%
not determined
not determined
+2.1%
+2.5%
+4.9%
+61.2%
A-25
-------
25. Provide average concentrations in mg/L or other appropriate units for upstream cell
only warm season (May-October) and cold season (November-April) influent and effluent:
(Warm season does not start at this system until May.)
Parameter
W = warm season
C = cold season
BODS(W)
BODS (C)
COD (W)
COD (C)
(NH3+NH4)-N (W)
(NH3+NH4)-N (C)
Org-N(W)
Org-N (C)
Totai-P (W)
Total-P (C)
PO4-P (W)
PO4-P (C)
TSS(W)
TSS (C)
DO(W)
DO(C)
Fecal coliforms (W)
colonies perl 00 mL
Fecal coliforms (C)
colonies per 100 mL
Fecal Strep. (W)
Fecal Strep. (C)
pH(W)
PH(C)
Water temp. (W)
Water temp. (C)
Average Concentration or Reading
Influent
372.08
696.23
not measured
not measured
74.36
192.18
13.25
25.32
13,31
19.16
8.94
8.32
81
74
0.99
2.70
96
222
not measured
not measured
7.58
7.26
20.71
6.67
Effluent
68.28
174.24
not measured
not measured
50.47
87,93
8.79
9.65
8.58
5.56
7.66
4.52
71
43
1.84
3.56
14
43
not measured
not measured
7.66
7.37
20.04
6.94
Percent change
-81.6%
-75.0%
not determined
not determined
-32.1%
-54.2%
-33.7%
-61.9%
-35.5%
-71.0%
-14.3%
-45.7%
-12.3%
-41.9%
+85.9%
+31 .9%
-85.4%
-80.6%
not determined
not determined
+1.1%
+1.5%
-3.2%
+4.0%
27. For the data in the above tables, what is the period of observation?
for most parameters: May, 1994 to October, 1995
for pH, DO, and temperature July, 1994
to
October, 1995
28. How many sampling days were included in the study?
33 sampling dates for most parameters
29 sampling dates for pH, DO, and water temperature
29. What were maximum and minimum influent concentrations for:
Parameter
Ammonia ({NH3+NH4)-N)
TSS
TP
PO4-P
Maximum (mg/L)
435.32
757
36.44
29.01
Minimum (mg/L)
0.00
9
0.59
0.00
A-26
-------
30. Which of the following represented a problem at your site (elaborate as needed or
place N/A if not applicable).
a) Insects destroyed a particular plant species: N/A
b) Musk rats or other animals created problems:
Deer trampled largest cattails, there was no observable adverse impact to the system.
c) Plants were killed at the upper end of cells because concentrations of ammonia or
suspended solids were too high: N/A
d) Evapotranspiration rates were so high in summer that plants at lower end of system
were killed or stressed:
Due to bottom slope of cells, plants at upper end of system were stressed from low water
in summer,
e) Discharge limits could not be met for certain constituents in the final effluent:
This was a no discharge system so this aspect was not applicable.
f) Phosphorus concentrations in final effluent tended to increase over time:
Total-P concentrations were higher at the end of the 2-year period, but inflow
levels were much higher at this time. Higher outflow rates probably resulted
from increased loading rather than phosphorus saturation within the wetland
system.
g) Water management is a major problem. A water balance should be developed for any
new systems which include:
1) drawing down lagoon levels in late fall to accommodate winter storage of flush water
and waste, rainwater on the lagoon surface, and runoff water;
2) controlling water released to wetland system based on seasonal changes;
3) collecting water discharged from the system and recycling or land applying;
4) accounting for rainwater on the surface of wetlands.
(comments on the need for water balance)
A) With gravity flow into a terminal infiltration area, item number three on the above list is less
critical. However, design must include major storm events in infiltration area design for
proper functioning.
B) Wetland design should include average expected rainfall as part of the flow volume to be
treated. Water volume from major storm events should be handled separate from the
wetland basins, either with storage ponds or lagoons.
C) Water level management is the most important task that producers must do once wetland
systems are in place. Adjusting flows and volumes to allow vegetation establishment and to
handle high flow volumes is critical. This is especially important in seasonal climates such
as the upper midwest. The period of high flow from rainfall and snowmelt coincides with
early spring when temperatures are low and microbial activity is low. High volumes can
move through wetland systems at a time when treatment efficiency is low. Proper design
and water level management allows storage of excess liquids for release to the wetland
during times when microbial activity and waste treatment are high.
A-27
-------
h) Mosquitoes were a problem:
The landowners stated that they noticed no problems with mosquitoes during the two
years of the study.
1) Uneven distribution of wastewater across all ceils in a multi-cell system: N/A
31. What place do constructed wetlands have in managing animal wastes? What are the
drawbacks? Should they be permitted for discharge?
1. Constructed wetlands can be successfully integrated into an onfarm total waste management
system. Site conditions (topography, soils) are prime determinants of whether a particular
operation will find wetland treatment systems an economically viable option for inclusion in
waste management systems. The degree of treatment needed from a wetland system will
vary depending upon the level of pretreatment and the needs of particular producers.
2. A particular drawback to widespread use of these systems is the extreme variation in
regulations governing animal waste treatment systems. Regulations impose particular
design and/or operation constraints on wetland systems. A system that works well in one
state may be excluded from another state because of the regulatory climate.
3. The issue of discharge is rather complex. Discharge may not be desirable in all situations.
A producer wishing to obtain nutrient benefits through land application of wastewater will not
want to treat wastewater to the point of meeting discharge standards, on the other hand,
some producers may need to eliminate excess water. In such a situation, discharge
becomes a desirable goal. If one looks at individual onsite wetland systems, animal waste
treatment wetlands that obtain NPDES discharge standards should be permitted for
discharge into waterways, provided that NPDES monitoring is done to determine
compliance. It seems reasonable to expect the same degree of monitoring as industry and
municipalities face for discharge. This may not be an economically viable option for many
animal production operations. Water quality monitoring and reporting are additional time
and money expenses that a producer may not be able to afford. However, if a total
watershed approach is adopted where onsite wetland systems discharge through other
collective treatment wetlands, discharge without onsite monitoring may be practical. As with
any unmonitored situation, the potential for bad actors exists. Without onsite monitoring,
there is no way to guarantee wetland systems will be properly maintained.
32. What additional research is needed on these systems?
1, Further work is needed on integrating site-specific waste treatment systems into a full
watershed-level NFS control program.
2, Additional work is needed to support life expectancy projections for animal waste treatment
wetlands, examining effects of management practices, climate, and different solid loading
rates.
3. Given the public perception of the risks associated with water-borne pathogens,
investigations into the mechanisms of pathogen removal and design criteria that can
enhance pathogen removal are probably warranted.
A-28
-------
Cropland
Yard runoff collection
Drying
pad -
Manure pit
Runoff diversion
Cell 2
Holding
pond
Flow control Switching valve
Infiltration
Area
Figure 1. Generalized layout of waste treatment system at Tom Brothers' Dairy
A-29
-------
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or cooperating researchers:
a. Name: Dr. Charles M. Cooper
Address: USDA-ARS
National Sedimentation Laboratory
P. O. Box 1157. Oxford. MS 38655
Phone: 601-232-2933 FAX: 601-232-2915
Your involvement / experience with this wetland system:
Supervised management and monitoring of wetland treatment system
b. Name: Samuel Testa III
Address: USDA-ARS
National Sedimentation Laboratory
P. O. Box 1157. Oxford. MS 38655
Phone: 601-232-2933 FAX: 601-232-2915
Your involvement / experience with this wetland system:
Water quality sampling and monitoring, and maintenance of wetland treatment system
Others included on separate sheet: Ic.n Id. n
2. Location of the system (description or address): Scott Dairy Farm. 5 miles ESE Hernando. MS
Location in terms of latitude and longitude: Lat. M34°45' Long. W89°54'
3. Approximate percentage of funding for this project was provided by:
EPA % State Water Quality Agency % TVA % USDAJflQ_%
State Experiment Station % Other
A-31
-------
4. Animal information:
Type: Hostein dairy Number: 60-100 range Avg. Wgt. 1000-1 200 ib
Type: _ Number: _ Avg, Wgt. _
Type: _ Number: _ Avg. Wgt. _
Type: _ Number: _ _ Avg. Wgt. _
Type: _ Number: _ Avg. Wgt. ............................
5. Date wetland system was installed: yr. 1990 month April
6. Date system became operational (first waste discharge): yr 1991 month April
7. Type of pre- treatment (check one or describe):
One ceE lagoon x Two lagoon cells _
Waste Storage Pond (as defined by SCS or NRCS): _
Settling Basin _
Other: _ _
8. Approximate age of pre-treatment units when wetland became operational: 1 yr
9. Had sludge ever been removed from the pre-treatment unit prior to installation of the wetland
system? yes _ ; _ yrs prior to installing wetland. no
10. Basis for design of wetland system (ie., 90 Ibs BOD/ac/day; minimum 7 day detention time;
etc.) If a wetland / pond / wetland system was used, include design basis for pond and for entire
system: _ See attached computation sheets _ ;
11. Whose design criteria did you use? (TVA, SCS, private industry, etc.) List any references:
USDA-SCS: Jimmy Wilson. District Engineer. New Albany. MS
12. Total surface area (water surface) of wetland cells: 432 m2 or acres
13. Total surface area of pond cells: _Q_ m2 or acres
14. Surface area of the entire system including embankments: s7QQ m2 or acres
15. Average water depth: « 25 cm
16. Average slope of cells from upstream to downstream: >• 1 %
A-32
-------
17. Provide a sketch in plan view showing all cells, embankments, and general location in relation
to pretreatment units and other significant features. Provide dimensions for the cells and
embankments. If preferred, attach engineering drawings or drawings from a publication and note
here.
See attached plan
18. Was the waste flow diluted prior to being discharged into the wetland system? No_i_
Yes , dilution ratio was % fresh water and % wastewater
Source of dilution water was:
Was dilution ratio changed during the course of the study?
Yes_, dilution ratio became % fresh water and % wastewater
Data in table, Items 25 and 26 is for initial dilution final dilution.,
19. Avg. daily warm season flows (non-storm flows):
Influent _I25A_ liters/day Effluent_57JL liters/day
20. Avg. daily cool season flows (non-storm):
Influent 1368 liters/day Effluent 749 liters/day
21. Describe flow controls at influent end of system(le., valves, swivel pipe, pump on tinier, etc.)
4000 li|gr constant head tank led to inflow at cells where drilled hole in PVC pipe end
cap gave desired volume/time. Different sized hpleg in threadedJendcapsjgQBldJje^se4. to
change flow rates
22. Any problems with clogging of influent pipes/ valves with debris, struvite, animals, etc.?
Original equipment gate valves for control (and ball valves) needed frequent remedial
action. Drilled endcaps fed from head tank rarely required maintenance,
Methods of control, if applicable:
Lagoon discharge to wetland protected by ¥2. inch hardware cloth
23. Describe water level controls in the cells (Le.,downstream swivel pipe): .
Downstream swivel nice
24. Any clogging of pipes between cells or from final outlet cells? None
Methods of control, if applicable: Outflow from cells first passed though a submerged
perforated (*A inch holes) 3-in. PVC pipe (with endcaps) that spanned the lower end of
the cell, plumbed at center to discharge pipe.
A-33
-------
25, Provide average concentrations in mg/L (or other appropriate units) for influent into the
system and effluent from the system for warm (April - October) and cold (November - March)
seasons.
Average for the system
Parameter
Wswarm season
C=cold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS(W)
mg/L
BOD5 (C)
COD(W)
244
50
COD(Q
343
80
77
(W) "
Org-N (W)
Org-N (C)
Toial P (W)
mg/L
T«al P (C)
PtVP* (W)
8.59
4,96
42.2
P04-P* (Q
11.20
6.40
42.9
TSS (W)
D,0. (W)
2.87
1.43
50.2
D,O.(Q
4.77
2.58
45.8
Fecal coUforms (W)
(No^lOOml)
Fecal coliforms, (C)
(NoJlOOml)
Fecal strep. (W)
Fecal strep. (C)
pH(W)
(standard units)
pH(C)
Water temp. (W) degC
22.5
20.4
9.1
Water temp. (Q
11.2
9.4
16.0
* Filterable ortho-P
A-34
-------
26. Provide same information for only the upstream cellfs). If a bank of cells is used, provide
average data for that bank of upstream cells.
Upstream cell(s)
NOT TAKEN—
Parameter
W=warm season
Ocold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS (W)
BOD5(C)
COD(W)
COD(C)
NHj-fNH4-N (W)
(C)
Org-N (W)
Org-N (C)
Total P (W)
Total P(C)
PO4-P (W)
P04-P (C)
TSS (W)
TSS (C)
D.O. (W)
D.O. (C)
Fecal coliforms (W)
(No./100ml)
Fecal coliforms. (C)
(NO./100 ml)
Fecal strep. (W)
Fecal strep. (C)
PH(W)
(standard units)
pH(C)
Water temp. (W)
(deg.C)
Water temp. (Q
27, For the data in the above tables, what was the period of observation?
mo. 5 yr. 91 to mo. _
yr. 94
A-35
-------
28. How many sampling days were included in the study? 181 days
29. What were maximum and minimum influent concentrations for:
a. Ammonia fNH3 + NHrN): Max. 30.8 mg/L Mm. _JH_mg/L
b. Total suspended solids: Max 466 mg/L Min. 0 mg/L
c. Total phosphorus (TP): Max 69 mg/L Min 1.3 mg/L
d. Phosphate (PO4-P)*: Max 24 mg/L Min 0.9 mg/L
*Filterable ortho-P
30. Which of the following represented a problem at your site (Elaborate as needed or place N/A
if not applicable):
a. Insects destroyed particular plant species:
Grasshoppers caused moderate damage during late summer of one year to Sfrjpuy
validusin monoculture.
b. Muskrats or other animals created problems:
No
c. Plants were killed at upper end of cells because concentrations of ammonia or
suspended solids were too high:
, NO plant deathjwas seen, but noticeable yellowing and stunting ocurred immeidiately
after first wastewater introduced.
d. Evaporation rates were so high in summer that plants at lower end of system were
killed or stressed:
Mo. but water column became very shallow at times.
e. Discharge limits could not be met for certain constituents in the final effluent:
List: N/A
f. Phosphorus concentrations in the final effluent tended to increase over time:
Yes. See attached graph. ^^___
g. Water management is a major problem. A water balance should be developed for any
new systems which include:
(1) drawing down lagoon levels in late fall to accommodate winter storage of flush
water and waste, rainwater on the kgoon surface, and runoff water;
(2) controlling water released to wetland system based on seasonal changes;
(3) collecting water discharged from the system and recycling or land applying;
(4) accounting for rainwater on the surface of the wetlands.
A-36
-------
(comments on need for water balance)
Design should allow for accomodating strorn events.
h. Mosquitoes were a problem: Never
i. Uneven distribution of wastewater across all cells in a multi-cell system: No
None noticed ',
j. Others: ,
More thoughts on problems included on separate sheet n
31. What place do constructed wetlands have in managing animal wastes? What are the
drawbacks? Should they be permitted for discharge?
See attached sheet
32. What additional research is needed on these systems?
See attached sheet
A-37
-------
Question #31.
Constructed wetlands are useful as secondary treatment of animal wastes, but at present should probably
only be used at the tertiary level for final polishing of wastewater. At present, generality of design is a
drawback, as systems attempt to deal with a variety of pollutants and physical/environmental conditions.
More site-specific design criteria are necessary for particular contaminants and specific conditions,
which may also lead us to a better knowledge of the processes occurring in constructed wetlands.
Current permitting should not be allowed for release into surface waters of the nation unless the systems
are greatly overdesigned, and capable of handling the large variability inherent to these living systems.
Question #32.
Constructed wetland systems for waste treatment so far have mostly been general, catch-all systems,
Design criteria need to be developed for particular pollutants and site characteristics, allowing better
control and predictability. Recent research at our wetland system indicates that some nitrate-reducing
bacteria also reduce iron. This may be important when iron is present at the site, there is potential for
nitrate-reducing bacteria to be already present. Additional research is also needed concerning longevity
of these systems. It has already been demonstrated that phosphorus trapping potential decreases sharply
after relatively short times of operation, and we have information on the processes causing this effect.
Processes controlling other pollutant dynamics are much less well understood. Long-term capabilities of
these systems for treating other pollutants are also less well known, and procedures for re-attaining lost
treatment capability need to be developed. Routine maintenance procedures may be developed which
help sustain the treatment capability of constructed wetlands, but much more information is necessary
concerning the physical and biological processes before these procedures can be ascertained.
A-38
-------
Pitch
Rainwater in(et-pipe-^*\
Inflow Outflow
LAGOON
Direction of Flow
i
e
f
5 C
k,
f
t
i
»
>
»
5
k
Outflow/Inflow Outfl
Inflow
^ CELL #1
1
in IB
X
;|
i
b 1 c.
inflow
^ CELL #3
6m
e
vaiv
^
^ CELL #4 II
1 Outflow
*^— CELL #2
^^ Walkway
fi Outflow
x24m
ach
es 1,2, and 3
40m x 52m
Valve
Figure 1. Layout of lagoon\wetland cell system at Hernando
A-39
-------
-------
Site Information Summary
Type project: Constructed wetland for treating dairy wastewater
Contact details:
Paul Hermans. P. Ag.
Essex Region Conservation Authority
360 Fairview Avenue, West
Essex. Qnt. CN N8M 1Y6
Ph. (519) 776-5209 '
Fax #:(519)776-8688
Site Name: Maiden Valley Farms Constructed Wetland
Sytem Name; Constructed wetland for dairy operation Mastes
Country: Canada
Province/State: Ontario
City/Community: Woodske
Description of Waste Production Facility:
200 head dairy: 65 milking cows: 200 gallons of milkhouse waste
per day; manure lot runoff
Wastewater Pretreatment: Solids separation in barnyard
Stormwater: Watershed Area: 0.25 ha
% Impervious (Roofs, parking lots, etc.) 100%
Wetland Hydrologic Type (surface flow, subsurface flow, hybrid, etc)
surface flow wetland
Number of Cells: three: pond/marsh/pond
Cell configuration and dimensions: see attached drawing
Cell length: 120 m Cell width: 5 m ;
Cell bottom slope: 0.25% Cell avg depth: 0.30 m (avg. 0-0.6 m)
Substrate material: clay
Predominant plant species: cattail
Captial cost: $18.000
Climatic Data:
Avg # frost free days: 154 (May 4 - Qct 6)
Avg annual temp (°C) 8t9 Avg winter temp -3.4
Annual snowfall (cm) 100 Annual rainfall (cm): 712
Elevation (msl): 182m
A-41
-------
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or cooperating researchers:
a. Name: James A. Moore
Address: Department of BioresourceEngineering
Oregon State University
Corvallis. OR 97330
Phone: 5417737-2041 FAX: 541/737-2082
Your involvement / experience with this wetland system:
PrincipalInvestigator
Email; moorei@ccmail.orst.edu
b. Name: Steven F. Miswander
Address: Dept. of Bioresource Engineering
Oregon State University
Corvallis. OR 97330
Phone: 541/737-6296 FAX: 541/737-2082
Your involvement / experience with this wetland system:
Graduate Research Assistant:
Email: niswandsSpandora.bre. or st.edu
Others included on separate sheet: lc.B9 Id. D
2. Location of the system (description or address): o-ro.grm Rtqt
Campus Dairy
Location in terms of latitude and longitude: Lat.M 37' Long. 123 12'
3. Approximate percentage of funding for this project was provided by:
EPA 20 % State Water Quality Agency % TVA ._% USDA6J1_%
State Experiment Station 15 % Other Oregon Dairy Farmers Association - 5%
A-43
-------
4. Animal information:
Type: Milking Number: 175 Avg. Wgt. 1.400 Ibs
Type: Number: Avg. Wgt.
Type: Number: Avg. Wgt.
Type: Number: Avg. Wgt.
Type: Number: Avg. Wgt.
5. Date wetland system was installed: yr. 1992 month May
6. Date system became operational (first waste discharge): yr 1993 month October
7. Type of pre-treatment (check one or describe):
One cell lagoon Two lagoon cells
Waste Storage Pond (as defined by SCS or NRCS):
Settling Basin
Other: Solids separator
8. Approximate age of pre-treatment units when wetland became operational: N/A
9. Had sludge ever been removed from the pre-treatment unit prior to installation of the wetland
system? yes ; yrs prior to installing wetland. no
10. Basis for design of wetland system (Le., 90 Ibs BOD/ac/day; minimum 7 day detention time;
etc.) If a wetland / pond / wetland system was used, include design basis for pond and for entire
System: 1QQ mg NH,/!; Total solids 1,500 mg/1
BOD 74 kg/ha-day
11. Whose design criteria did you use? (TVA, SCS, private industry, etc.) List any references:
USEAP 1988. Design manual on constructed wetlands and aquatic plant systems
tor municipal wastewater treatment.EPA/bz5/1-88/uzz.
12. Total surface area (water surface) of wetland cells: 784 m2 or acres
13. Total surface area of pond cells: 98 m2 or acres Deep water sections
Pond 4 and 9
14. Surface area of the entire system including embankments: 2,500 m2 or acres
15. Average water depth: an cm
16, Average slope of cells from upstream to downstream: o.s %
A-44
-------
17. Provide a sketch in plan view showing all cells, embankments, and general location in relation
to pretreatment units and other significant features. Provide dimensions for the cells and
embankments. If preferred, attach engineering drawings or drawings from a publication and note
here.
See figure 1-5, Note these are design drawings and acutal size of ponds is
" slightly different.
18. Was the waste flow diluted prior to being discharged into the wetland system? No
Yes_jL, dilution ratio waP'% fresh water md^$k wastewater
Source of dilution water was: recycled water from pond 10
Was dilution ratio changed during the course of the study? NO
Yes , dilution ratio became % fresh water and % wastewater
Data in table, Items 25 and 26 is for initial dilution final dilution
19. Avg. daily warm season flows (nOn-storm flo*ws):
Influent liters/day Effluent z_ liters/day * currently developing H20 budget
34,750
20. Avg. daily cool season flows (non-storm):
Influent liters/day Effluent _? liters/day
34,750
21. Describe flow controls at influent end of system(Le., valves, swivel pipe, pump on timer, etc.)
recycle water -> pump on timer—•
wastewater - electric ball valave on timer
flows to pond -1" manual ball valves
22. Any problems with clogging of influent pipes/ valves with debris, struvite, animals, etc.?
No problems when operating all ponds at same detention time.
We are nowhaving problems with operating 2 ponds at 2 day DT and 4 ponds' at 7 days.
Methods of control, if applicable:
23. Describe water level controls in the cells (i.e.,downstream swivel pipe):
PVC standpipe cut to, proper height
24. Any clogging of pipes between cells or from final outlet cells?
Methods of control, if applicable:___
A-45
-------
25. Provide, average concentrations in mg/L (or other appropriate units) for influent into the
system and effluent from the system for warm (April - October) and cold (November - March)
seasons.
Average for the system
Parameter
W=warm season
O=cold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS (W)
BODS (C)
CQDCW)
2.812
1.245
56
COD(Q
1,686
896
47
NH,+NH4-N (W)
NH}+NH4-N (C)
Org-N (W)
225
109
52
Ofg-N (C)
117
68
42
Total P (W)
44,9
Total P (C)
PO«-P (W)
PO4-P (C)
4.9
*only
AT ft gi3ITIT\1 Offl
TSS (W)
,444.
TSS (C)
D.O, (W)
2.72
0.15
94
D.O, (Q
5.14
0.28
95
Fecal conforms (W)
(No./iOO ml)
Rxal coliforms. (C)
(No^lOO ml)
Ffccai strep. (W)
Fecal strep. (C)
pH(W)
(standard units)
pH(O
Water temp. (W)
(deg.C)
12.9
12.1
Water temp. (C)
Total Solids (w) 3,329
Total Solids (c) 1,586
A-46
1,736
958
48
35
-------
26. Provide same information for only the ppstream cell(s:). If a bank of cells is used, provide
average data for that bank of upstream cells.
Upstream cell(s)
Parameter
W=warm season
C=co!d season
Average Concentration or Reading
Influent
Effluent
Percent Change
BODS (W)
BODS (C)
COD (W)
COD (C)
(W)
NH3+NH4-N (C)
Org-N (W)
Org-N (C)
Total P (W)
Total P (C)
PO4-P (W)
PO4-P (C)
TSS (W)
TSS (C)
D.O. (W)
D.O. (C)
Fecal conforms (W)
(NO./100 ml)
Fecal coliforms. (C)
(No.7100 ml)
Fecal strep. (W)
Fecal strep. (C)
PH (W)
(standard units)
pH (C)
Water temp. (W)
(deg.C)
Water temp. (C)
27. For the data in the above tables, what was the period of observation?
mo.
to mo.
yr.
A-47
-------
28. How many sampling days were included in the study? 27 days
29. What were maximum and minimum influent concentrations for:
a. Ammonia (NH3 + NHLj-N): Max. 301 mg/L Min. 12 mg/L
b. Total suspended solids: Max 1,705 mg/L Min. 75 mg/L
c. Total phosphorus (TP): Max 115 mg/L Min 3.5 mg/L
d. Phosphate (PO4-P): Max 12.0 mg/L Min 1.2 mg/L Note only 8 samples
30. Which of the following represented a problem at your site (Elaborate as needed or place N/A
if not applicable):
a. Insects destroyed particular plant species:
N/A
b. Muskrats or other animals created problems:
Nutria destroyed vegetation and burrowed through berms during
Electric fence was installed around entire site.
c. Plants were killed at upper end of cells because concentrations of ammonia or
suspended solids were too high:
N/A. We are interested in whatcauses death of plants.We conducted a
study and found that high ammonia (71,000 mg/1) did not inhibit
wetland plant seed germination. We think the death of plants may be
volatile acids.
d. Evaporation rates were so high in summer that plants at lower end of system were
killed or stressed:
N/A ,
e. Discharge limits could not be met for certain constituents in the final effluent:
List: N/A no discharge
f. Phosphorus concentrations in the final effluent tended to increase over time:
W/A —,
g. Water management is a major problem. A water balance should be developed for any
new systems which include:
(1) drawing down lagoon levels in late fall to accommodate winter storage of flush
water and waste, rainwater on the lagoon surface, and runoff water;
(2) controlling water released to wetland system based on seasonal changes;
A-48
-------
(3) collecting water discharged from the system and recycling or land applying;
(4) accounting for rainwater on the surface of the wetlands.
(comments on need for water balance)
All four are extremely Important fgr .Interpretation nf concentrations +
In July 1995 we installed a H20 level recorder in pond 10 to help us
construct a H20 balance for our site.
h. Mosquitoes were a problem: N/A
i Uneven distribution of wastewater across all cells in a multi-cell system: only
after switching to 2 and 7 day retention times.
j. Others:
More thoughts on problems included on separate sheet a
31. What place do construced wetlands have in managing animal wastes? What are the
drawbacks? Should they be permitted for discharge?
I. Two main purposes:
1) for mass removal of nutrients, solids, and organics. which would
reduce amount of wastewater that would need to be disposed of.
2) for final polishing of pretreated, wastewater before discharge.
II. Drawbacks: Require large land area.
III.Depends on circumstances but generally yes.
32. What additional research is needed on these systems?
Design criteria and a better understanding of internal processes.
A-49
-------
Ic. Information on cooperating researchers:
MikeJ. Gamroth
Animal Science Department
Oregon State University
Corvallis, OR 97330
Ph: (541) 737-3316 FAX: (541) 737-4174
Co-investigator
Id. Steven M. Skarda
Dept. of Bioresource Engineering
Oregon State University
Corvallis, OR 97330
Ph: (541) 737-6296 FAX: (541) 737-2082
Lab Technician
A-50
-------
15.5
Observation
wells
Parlor waste
treatment
cells
Diluted waste
treatment cells
Effluent storage cell
Figure 1: Plan View of OSU Constructed Wetland Cells
A-51
-------
23
15.5
48,5
3.75*
95
•120.5
1
4'
..25
Figure 2: Dimensions of Constructed Cells
A-52
-------
-6.23' &—•
Berm
Pond
2.5'
-..I fl
•15.5
I
slope of pond sides
Figure 3: Cross section "A" of Wetland Cells
12'-
10'
B
Deep water section
Figure 4: Cross section "B" of Wetland Cells
A-53
-------
Hi iking
Parlor waste
supply |me 4"
Dairy waste
storage lank
finlnal waste
supply I ine 4'
Fence
Figure 5: OSU Constructed Wetlands Piping Schematic
A-54
-------
CONSTRUCTED WETLANDS
FOR TREATING ANIMAL WASTES
QUESTIONNAIRE
1. Information on person(s) completing this questionnaire or cooperating researchers:
a. Name: David T. Hill
Address: Agricultural Engineering Pent
Auburn University
Auburn University. AL 36849-5417
Phone: 334-844-3531 FAX: 334-844-3530
Your involvement / experience with this wetland system:
Researcher. Ooeralor
b. Name:
Address:
Phone: FAX:
Your involvement / experience with this wetland system:
Others included on separate sheet: Ic.n Id. o
2. Location of the system (description or address): Ala. Agricultural Experiment Station.
Poultr Research Unitr Aubum UflfaEfTSftVi Alt
Location in terms of latitude and longitude: Lat. 3236.87 Long. 85 25.86
3. Approximate percentage of funding for this project was provided by:
EPA % State Water Quality Agency _ % TVA _ % USDA40
State Experiment Station 60 % Other _
A-55
-------
4. Animal information:
Type! Poultry Number: 10.000 layers Avg. Wgt. 1.8kg
Type: Number: , Avg. Wgt _
Type:i ,,.. , Number: Avg. Wgt.
Type: ' Number: Avg. Wgt.
Type: Number: Avg. Wgt.
5. Date wetland system was installed: yr. 199% month June
6. Date system became operational (first waste discharge): yr!222_ month .
7. Type of pre-treatment (check one or describe):
One ceH lagoon Two lagoon cells
Waste Storage Pond (as deined by SCS or NRCS):
Settling Basin
Other: Three lagoon cells: post wastewater delivered to wgtlin.4 from cell $1
8. Approximate age of pre-treatment units when wetland became operational: :: ll.yfs
9. Had sludge ever been removed from the pre-treatment unit prior to installation of the wetland
system? yesjs_ ; _1_ yrs prior to installing wetland. no
10. Basis for design of wetland system (ie., 90 Ibs BOD/ac/day; minimum 7 day detention time;
etc.) If a wetland / pond / wetland system was used, include design basis for pond and for entire
system: $5 Ibs BQD/ac/day. 11 day HRT
11. Whose design criteria did you use? (TVA, SCS, private industry, etc.) List any references:
USDA-SCS ,
12. Total surface area (water surface) of wetland cells: 781 m2 or acres
13. Total surface area of pond cells: 3250 m2 or acres
14. Surface area of the entire system including embankments: 4031 m2 or acres
15. Average water depth: 25 cm
16. Average slope of cells from upstream to downstream: 1 %
A-56
-------
17. Provide a sketch in plan view showing all cells, embankments, and general location in relation
to pretreatment units and other significant features. Provide dimensions for the cells and
embankments. If preferred, attach engineering drawings or drawings from a publication and note
here.
18. Was the waste flow diluted prior to being discharged into the wetland system? Nox*
Yes_, dilution ratio was % fresh water and % wastewater
Source of dilution water was:
(* Wastewater was dilute for a poultry facility because of high water use for flushing.)
Was dilution ratio changed during the course of the study?
Yes , dilution ratio became % fresh water and % wastewater
Data in table, Items 25 and 26 is for initial dilution final dilution
19. Avg. daily warm season flows (non-storm flows):
Influent _25QQ_ liters/day Effluent liters/day
20. Avg. daily cool season flows (non-storm):
Influent _25flflL liters/day Effluent _liters/day
21. Describe flow controls at influent end of system(Le., valves, swivel pipe, pump on timer, etc.)
Valves
22. Any problems with clogging of influent pipes/ valves with debris, struvite, animals, etc.?
Methods of control, if applicable:
Manual cleanin
23. Describe water level controls in the cells (Le., downstream swivel pipe):
Downstream swivel PIPS
24. Any clogging of pipes between cells or from final outlet cells? Yes
Methods of control, if applicable: Manual cleaning
A-57
-------
25. Provide_average,concentrations in mg/L (or other appropriate units) for influent into the
system and effluent from the system for warm (April - October) and cold (November - March)
seasons.
Average for the system
Parameter
W=warm season
C=cold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BOD5CW)
BODS (C)
COD(W)
400.0
205.0
48.8
COD(Q
300.0
155.0
48.3
(W)
Org-N (W)
25.0
5.5
78
Org-N(C)
20.0
64
Total P (W)
Total P(Q
PO<-P (W)
P04-P(C)
TSS (W)
ISSCQ
D.O. (W)
D.O.(Q
F«al colifcfins (W)
Fecal colifonns. (C)
(Noj'lOOEal)
Fecal strep. (W)
Fecal strep. (Q
pH(W)
(standard units)
pH{Q
Water tenq). (W)
(dcg.C)
86.6 F
85.5 F
Water temp. (C)
55.6 F
54.2 F
A-58
-------
26. Provide same information for only the upstream cell(sX If a bank of cells is used, provide
average data for that bank of upstream cells.
Upstream cell(s)
NOT TAKEN—
Parameter
W=warm season
C=cold season
Average Concentration or Reading
Influent
Effluent
Percent Change
BOD5(W)
BOD5(C)
COD(W)
COD(Q
(W)
-N (Q
Org-N (W)
Org-N(Q
Total P (W)
Total P(Q
(W)
PO,-P{C)
TSS (W)
TSS(C)
D.O. (W)
D.O.(Q
Fecal colifonns (W)
(NoJlOOml)
Fecal colifonns. (C)
(NoJlOOml)
Becal strep. (W)
Fecal strep. (Q
pH(W)
(standard units)
pH(Q
Water tenqp. (W)
(deg. Q
Water ten?). (C)
27. For the data in the above tables, what was the period of observation?
mo. ij yr. 93 to mo.
- 94
A-59
-------
28. How many sampling days were included in the study? 14 days
29. What were maximum and minimum influent concentrations for:
a. Ammonia (NH-, + NHrN): Max. 126 mg/L Min. 72.2 mg/L
b. Total suspended solids: Max mg/L Min. mg/L
c. Total phosphorus (TP): Max 36.5 mg/L Min 15 mg/L
d. Phosphate (PO4-P): Max mg/L Min mg/L
30. Which of the following represented a problem at your site (Elaborate as needed or place N/A
if not applicable):
a. Insects destroyed particular plant species:
NO
b. Muskrats or other animals created problems:
NO
c. Plants were killed at upper end of cells because concentrations of ammonia or
suspended solids were too high:
NO
d. Evaporation rates were so high in summer that plants at lower end of system were
killed or stressed:
NO :
e. Discharge limits could not be met for certain constituents in the final effluent:
List: We weren't concerned with discharge limits
f. Phosphorus concentrations in the final effluent tended to increase over time:
Only in the cool season
g. Water management is a major problem. A water balance should be developed for any
new systems which include:
(1) drawing down lagoon levels in kte fall to accommodate winter storage of flush
water and waste, rainwater on the lagoon surface, and runoff water;
(2) controlling water released to wetland system based on seasonal changes;
(3) collecting water discharged from the system and recycling or land applying;
A-60
-------
(4) accounting for rainwater on the surface of the wetlands.
(comments on need for water balance)
This was a research unit. We were not concerned with lagoon operation. We maintained
constant flow to the wetland cells. . ____^______
h. Mosquitoes were a problem: Yes-- but they were not bjtiqg mosquitoes
i Uneven distribution of wastewater across all cells in a multi-cell system: No
j. Others: Influent pipe plugging
More thoughts on problems included on separate sheet a
31. What place do constructed wetlands have in managing animal wastes? What are the
drawbacks? Should they be permitted for discharge?
Wetlands are questionable for managing animal wastes, they do not provide for final disposal
and are only another intermediate treatment— and a headache at that. I do not see them enjoying a
"place in the sun" in animal waste management operations. Phosphorus build-up over years will
be the main problem. In short, they are more trouble than they are worth. Should not be permitted
for discharge.
32. What additional research is needed on these systems?
Water balance, temperature variation caused by plants
C A-61
-------
Lagoon 1
Valve Center
1
1
1 -
Lagoon 2
V
1
Lagoon 3 >
=1=1
* Cell
Cel2
\
i
\
Series 1
Series 2
Series 3
Series 4 & 5
Figure 1. Overview of the lagoon and constructed wetland treatment system at
Auburn University's Poultry Research Unit
A-62
-------
Constructed Wetlands
for
Animal Waste Treatment
Appendix B
Typical Aquatic Plants Used in
Animal Waste Treatment
Constructed Wetlands
-------
-------
Typical Aquatic Plants Used for Animal Waste Treatment Wetland
A wide variety of species of wetland plants
have been used in municipal waste treatment.
The same cannot be said of wetlands used to
treat animal wastes. Some species have been
planted in animal waste wetlands but have
either not survived because of wastewater
strength or because they could not compete
successfully with more tolerant or more
aggressive species.
Presented here are a few species that have
performed successfully in the animal waste
environment. Most have been purposefully
introduced but others are natural invaders.
Certain species which were purposefully
introduced in one place might be natural
invaders in another.
Hydrocotyle umbellata
(Water pennywort, marsh pennywort)
The growth habit of the marsh pennywort is
a cross between emergent and floating. It is a
common natural invader in animal waste
constructed wetlands. Since it tolerates
partial shade, it readily occupies open areas
between taller emergent plants.
The most characteristic feature of marsh
pennywort is its circular leaves. Each leaf is
centrally attached on a long, slender stem; it
is scalloped on the edges with the scallops
appearing as shallow rounded teeth (crenate
margins). Small white flowers are produced
in simple, many-flowered clusters; each
cluster is formed on a narrow stem which
often rises above the height of the leaves.
Vegetative growth method: stolons or
rhizomes
Growth and spread rate: rapid
Persistence: perennial, non-persistent
Spacing when planted: N/A
Water regime: regular to permanent
inundation, <12 in (30 cm)
Other comments: Has proliferated in the first
cell of an animal waste wetland with NH4
concentrations greater than 80 ppm.
between taller plants in this wetland.
Pennywort's rounded, crenate leaf
Iris versicolor, virginicus
(Blue flag)
Blue flag is an elegant plant which produces
a large, pale blue to purple flower. The
B-l
-------
flower resembles the domestic, garden-
variety Dutch iris. The leaves are narrow, flat
and pointed, and they arise like a fan from
the base of the plant.
Blue flag has been used successfully in on-
site, subsurface constructed wetlands and in
municipal systems. Although it has been
planted in wetlands for treating animal
wastes, it has not been as successful. It
would probably fair best in a second or third
cell where wastewater strength has been
diminished. Without assistance, it may not
compete well with more vigorous species.
Vegetative growth method: bulb
Growth and spread rate: slow, <2 ft (60
cm)/yr
Persistence: perennial, persistent
Spacing when planted: 0.5-1.5 ft (15-45 cm)
Water regime: regular to permanent
inundation, 6 in (15 cm)
Blue flag (Iris versicolor, virginicus)
Juncus effusus
(Soft rush)
Rushes are grass-like, emergent plants which
grow in clumps. The pale green stems are
cylindrical and grow to 2 to 5 ft (0.6 -1.5 m)
tall. Flowers are produced on floral axis
(inflorescence) which is open and branched.
This flowering axis appears from late spring
through fall and emerges from the side of the
plant.
Juncus has been planted and survived in
animal waste wetlands. In some cases they
have been crowded out by other species.
Vegetative growth method: rhizomes
Growth and spread rate: slow, <15 in (6
cm)/yr
Persistence: perennial, persistent
Spacing when planted: 0.5-1.5 ft (0.15-0.45
m) O.C.
Water regime: regular to permanent
Soft rush (Juncus effusus)
Phragmites australis
(Common reed)
Phragmites is a perennial grass which estab-
lishes itself quickly and spreads rapidly
through its rhizomes. The stems of the plant
are slender and may reach 9 to 10 ft (2.7 -
3.0 m) tall. The leaf blades are about 1V4 in
(3.2 cm) wide and about a foot (0.3 m) long.
Silky white hairs about V4 in (0.6 cm) long
are located at the junction of the leaf sheath
and blade.
B-2
-------
The stems of this plant are leafy up to the
flowering head. The flower head or panicle is
a loose, feathery cluster of flowers which
may be over a foot (0.3 m) long. It will be
tawny, brown, purplish or silvery in color.
As the panicles age, the color becomes more
silvery due to the long hairs associated with
the flowers or spikelets.
Vegetative growth method: rhizomes
Growth and spread rate: rapid, >1 ft
(30 cm)/yr '
Persistence: perennial, persistent
Spacing when planted: 2,0-6.0 ft (0.6-
1.8 m) r
Water regime: seasonal to permanent
inundation, up to 2 ft (60 cm)
Other comments: Because of its aggressive
growth habits, Phragmites may be consi-
dered to be a pest plant in some parts of the
country, and its use in constructed wetlands
may be discouraged. However, it has f
performed well in animal waste wetlands
where relatively high pollutant loads
occurred.
Sagittaria spp.
(Duck potato, arrowhead)
Duck potato has large, thick leaves and
conspicuous white flowers. It gets its name
from the potato-like corms often produced
underground. Sagittaria consists of several
species whose leaves can vary considerably
in size and shape. Two species which have
been used successfully in animal waste
constructed wetlands include S. lancifolia
and S. latifolia. They are generally good
competitors with other more aggressive
species.
The leaves are typically 4 in (10 cm) wide
and up to 2 ft (1.6 m) long and grow as a
fan-like rosette from underground rhizomes.
The base of the plants are full and can
mound above the water level in shallow
water.
Sagittaria latifolia in bloom
The white flowers are showy with three
petals. They are on thick stalks that, often
extend a foot (0.3 m) or more above the
leaves.'-" ••. . ,.".:""." ' • ; ••
B-3
-------
Vegetative growth method: runners, tubers
Growth and spread rate: rapid, >1 ft (30
cm)/yr
Persistence: perennial, non-persistent
Spacing when planted: 2-6 ft (0.6-1.8 m)
Water regime: regular to permanent
inundation, up to 2 ft (60 cm)
Other considerations: The mounds which
form at the base of the plants may, as the
wetland matures, affect water flow and
hydraulic retention time. More information
will be needed on this topic.
Scirpm sp,
(Bulrush)
Several species ofScirpus have been used in
municipal wastewater treatment. A common
variety used for animal waste treatment is S.
validus (soft stem bulrush).
A colony of bulrush
S. validus typically grows in colonies and can
reach 10 ft (3 m) in height. The stems are
about 3/4 in (2 cm) thick at the base and
taper to a point at the top. The bulrush does
not have obvious leaves, but only sheaths at
the base of the stem. A floral axis appears at
the top of the stem and has several drooping
stalks that have irregularly clustered
spiketets.
Vegetative growth method: rhizomes
Growth and spread rate: rapid, >1 ft (30
cm)/yr
Persistence: perennial, persistent
Spacing when planted: 2 - 6 ft (0.6 - 1.8m)
Water regime: regular to permanent
The floral axis at the top of a bulrush stem
Typha spp.
(Cattail)
Cattails are perhaps the most easily recog-
nized of all wetland plants. They grow
prolifically in animal waste constructed.
wetlands and appear to be as adaptable to
high strength wastes as any plant.
The flower spike of cattail
B-4
-------
The plant gets its name from the cylindrical
flower spike which is packed with tiny
flowers. The spikes are cinnamon brown and
the leaves can reach 7 ft (2.1 m) in height.
The spikes can grow more than a foot long.
The cattail leaves are flat with rounded
backs. They are typically 1 in (2.5 cm) wide
and can grow to 5 to 8 ft (1.5 - 2.4 m) tall.
The leaves are sheathed together at the base.
Vegetative growth method: rhizomes
Growth and spread rate: rapid, >1 ft (30
cm)/yr
Persistence: perennial, persistent
Spacing when planted: 1 - 6 ft (0.6 -1.8 m)
Water regime: irregular to permanent
inundation, up to 1 ft (30 cm)
Other comments: Cattails are sometimes
attacked by caterpillars which strip the
leaves. This is a temporary occurrence and
the plants revive. The effect on treatment is
thought to be inconsequential
A caterpillar attacks the leaf of this cattail
Zizaniopsis miliacea
Giant cutgrass, Southern wild rice
Cutgrass is a perennial grass that grows to
10 ft (3 m) tal. It forms very dense stands
and has proliferated in animal waste
constructed wetlands. It has not been a
natural invader of these wetlands.
The stems are single leaf bkdes with widths
of 1 Vi to 2 Yz in (3.8 - 6.4 cm). The margins
of the blades are upwardly scabrous (cutting
or rough to the touch), hence the name
"cutgrass."
A stand of cutgrass in the second cell of a
wetland used to treat swine lagoon effluent.
The flowers of the plant form on a floral
axis, which is a loose and irregular branching
cluster or panicle. These panicles will reach
1 to 2 ft (30 to 60 cm) in height. Flowering
occurs from April to July in the South.
Vegetative growth method: rhizomes
Growth and spread rate: rapid
Persistence: perennial, persistent
Spacing when planted: 2- 4 ft (0.6 - 1.2 m)
Water regime: regular to permanent
inundation, up to 3 ft (0.9 m)
References:
CH2M Hill and Payne Engineering (1997),
Constructed Wetlands for Livestock
Wastewater Management: Literature
Review, Database, and Research Synthesis;
B-5
-------
prepared for the EPA's Gulf of Mexico Aulbaeh-Smith, C, A. and S. J. de Kozlowski
Program through the National Council of the (1990); Aquatic and Wetland Plants of South
Paper Industry for Air and Stream Carolina; South Carolina Aquatic Plant
Improvement (NCASI) and the Alabama Soil Management Council in cooperation with
and Water Conservation Committee; South Carolina Water Resources
Montgomery, AL. Commission; Columbia, SC
B-6
-------
Constructed Wetlands
for
Animal Waste Treatment
Appendix C
As-Excreted Waste and
Wastewater Production Values
for Livestock and Poultry
-------
-------
Wastewater Volumes and As-Excreted Waste Values
Table C-l. Swine: As-Excreted Values
Constituent
Mass
Volume
Nitrogen
Phosphorus
BOD5
Units*
kg/d
(Ib/d)
nf/d
(ft3/d)
kg/d
(Ib/d)
kg/d
(Ib/d)
kg/d
(Ib/d)
Growers
18.1 - 99.8 kg
(40-2201bs)
28.8
(63.4
0.028
(1.0)
0.19
(0.42)
0.07
(0.16)
0.94
(2.08)
Replacement
Gilts
14.9
(32.8)
0,015
(0.53)
0.11
(0.24)
0.04
(0.08)
0.49
(1.08)
Sows
Gestation
12.3
(27.2)
0.012
(0.44)
0.09
(0.19)
0.03
(0.063)
0.38
(0.83)
Lactation
27,2
(60.0)
0.027
(0.96)
0.21
(0.47)
0.07
(0.15)
0.91
(2,00)
Boars
9.3
(20.5)
0.009
(0.33)
0.07
(0.15)
023
(0.05)
0.30
(0.65)
Nursing/
Nursery Pigs
2.7-18.1 kg
(6 - 40 Ib)
48.1
(106)
0.048
(1.70)
0.27
(0.60)
0.11
(0.29)
1.54
(3.40)
Table C-2. Dairy; As-Excreted Values
Constituent
Mass
Volume
Nitrogen
Phosphorus
BOD5
Units*
kg/d
,(lb/d)
rtf/d
(fl?/d)
kg/d
(Ib/d)
kg/d
(Ib/d)
kg/d
(Ib/d)
Cow
Lactating
36.3
(80.0)
0.037
(1.30)
0.20
(0.45)
0.032
(0.07)
0.73
(1.60)
Dry
37.2
(82.0)
0.037
(1.30)
0.16
(0.36)
0.023
(0.05)
0.54
(1.20)
Heifer
38.6
(85.0)
0.037
(1.30)
0.14
(0.31)
0.018
(0.04)
0.59
(1.30)
*Units per 454 kg (1,000 Ib) of animal weight; Source: USDA-NRCS, 1992.
C-l
-------
Table C-3. Beef: As-Excreted Values
Constituent
Mass
Volume
Nitrogen
Phosphorus
BOD5
*
Units*
kg/d
Ob/d)
m3/d
(ftVd)
kg/d
Ob/d)
kg/d
Ob/d)
kg/d
Ob/d)
Feeder
340 - 499 kg
(750- 1,100 Ib)
High
Forage Diet
26.8
(59.1)
0.027
(0.95)
0.14
(0.31)
0.05
(0.11)
0.62
(1.36)
High Energy
Diet
23.2
(51.2)
0.023
(0.82)
0.13
(0.30)
0.043
(0.094)
0.62
(1.36)
Yearling
205 - 340 kg
(450- 750 Ib)
26.4
(58.2)
0.026
(0.93)
0.14
(0.30)
0.045
(0.10)
0.59
(1.30)
Cow
28.6
(63.0)
0.028
(1.00)
0.15
(0.33)
0.054
(0.12)
0.54
(1.20)
Table C-4. Poultry Layers: As-Excreted Values
Constituent
Mass
Volume
Nitrogen
Phosphorus
BODS
Units*
kg/d
Ob/d)
mVd
(ft3/d)
kg/d
Ob/d)
kg/d
Ob/d)
kg/d
Ob/d)
Layer Hen
27.4
(60.5)
0.026
(0.93)
0.38
(0.83)
0.14
(0.31)
1.68
(3.70)
*Units per 454 kg (1,000 Ib) of animal weight; Source: USDA-NRCS, 1992.
C-2
-------
Table C-5. Volume of Milkhouse and Parlor Wastes
Washing
Operation
Bulk tank
Automatic
Manual
Pipeline
In parlor
Miscellaneous equipment
Cow preparation
Automatic (estimated avg.)
Manual
Milkhouse floor
Parlor floor w/o flushing
Parlor and holding area
with flushing
Parlor only
Parlor and holding area
Holding area only
Water Volume
Liters
140 - 230
115-150
240-475
115
7.6
0.95 - 1.9
40-75
150 - 285
75-115
95 - 150
40-75
Gallons
50-60
30-40
75-125
30
2
0.25 - 0.50
10-20
40-74
20-30
25-40
10-20
Volume Per
wash
wash
day
wash/cow
day
day
cow/day
Table C-6. Minimum Total Daily Flush Volumes tor Swine
Swine type
Sow and litter
Pre-nursery
Nursery pig
Growing pig
Finishing pig
Gestating sow
Flush Volume
L/head
130
8
15
40
60
95
Gal/head
35
2
4
10
15
25
C-3
-------
-------
Constructed Wetlands
for
Animal Waste Treatment
Appendix D
Conversion Tables
-------
-------
Conversion Table
9.2903 x 10'2
0.3048
m
10.764
3.2808
0.4047
mile
1.6093
ac
2.471
m
0.5907
2.8317 x 10
r2
2.8317x10
,-2
35.314
35.31
gal/min »*L/s
6.309 x lO'2
3J854
L/s»* gal/min
15.85
0.264
0.454
Ib/ac *»• kg/ha
1.12
2.204
kg/ha»* Ib/ac
0.89
I Temperature:
°F=1.8(C°)
°C = 0.5556(°F - 32)
Milligrams per liter (mg/L): 1 mg/L is 1 milligram (mass) in 1 million parts (volume..i.e., liter). If
the liquid has a specific gravity similar to water, 1 mg/L =1 ppm. In concentrations below about
7,000 mg/L this relationship is generally true. A one percent solution has a concentration of
10,000 ppm, which equals 1 gm in 100 grams of water.
One ppm is approximately equal to 1 gallon of water by weight (8.34 Ibs) in one million gallons of
water (gallons). So, 1 ppm = 8.34 x 10"6 Ibs/gal or 0.00834 lbs/1000 gal., assuming a specific
gravity similar to water.
In wastewater applications, the fraction of solids can be high, and the relationship between ppm
and Ibs/gal is not totally accurate. However, for approximation purposes, concentrations of
nutrients in wastewater, for example, can be converted from mgTL or ppm to lb/1000 gal by
multiplying by 8.34 x 10"3. Thus, 200 mg/L N would convert to approximately 1.7 Ibs N/1000 gaL
-------
------- |