United States
Environmental Protection
Agency
Office of Water
(4204)
EPA 832-S-89-002
June 1i99
ETI
Free Water Surface Wetlands
for Wastewater Treatment
A Technology Assessment
June 1999
w
-------
Free Water Surface Wetlands
for Wastewater Treatment
A Technology Assessment
Prepared for
U.S. Environmental Protection Agency
Office of Wastewater Management
U.S. Department of the Interior
Bureau of Reclamation
City of Phoenix, Arizona
l With funding from the
I Environmental Technology Initiative Program
Environmental Technology Initiative
Prepared by
Humboldt State University
Environmental Resources Engineering Department
Arcata, Californai
CH2M-HJII
Gainesville, Florida
PBS&J
Phoenix, Arizona
June 1999
-------
Table of Contents
Table of Contents i
List of Tables v
List of Figures vi
List of Equations x
List of Acronyms and Symbols . xi
Acknowledgments xiii
Section 1 Introduction to Free Water Surface Treatment Wetlands 1-1
Background , 1-1
Introduction to the Technology 1-2
Treatment Wetland Forms & Functions 1-3
Other Benefits of Treatment Wetlands 1-5
Historical Development of the Technology 1-6
Application of the Technology 1-10
Summary of Technology Issues 1-12
Organization of this Report 1-12
Section 2 Methods for Technology Assessment 2-1
Data Sources.. 2-1
Technology Workshop and Peer Review 2-7
Data Quality and Validation 2-8
Section 3 Wetland Processes 3-1
Wetland Hydrology 3-1
Water Balance 3-1
Input Wastewater Flowrate 3-4
Precipitation 3-4
Evapotranspiration 3-4
Output Wastewater Flow 3-5
Exfiltration to Groundwater (Infiltration) 3-5
Meteorological Effects on Wetland Water Budget 3-5
Wetland Hydraulics 3-7
Wetland Hydraulic Definitions 3-7
Water Depth , 3-7
Surface Area 3-8
Volume 3-8
Wetland Porosity or Void Fraction.... 3-8
Hydraulic Detention Time 3-9
Hydraulic Loading Rate 3-10
-------
TABLE OF CONTENTS
Section 3 Wetland Processes (continued)
Wetland Hydraulics (continued)
Water Conveyance 3-10
Aspect Ratio 3-11
Internal Flow Patterns Effects/Physical Facilities 3-11
Water Balance Effects on Wetland Hydraulics and Water Quality 3-12
Wetland Biogeochemistry 3-13
Total Suspended Solids 3-16
Processes 3-16
Settieable Solids Reduction-Anaerobic Decomposition 3-18
Biochemical Oxygen Demand 3-18
Chemical Oxygen Demand 3-19
Dissolved Oxygen 3-20
Nitrogen 3-22
Phosphorus 3-24
Hydrogen Ion 3-25
Metals 3-26
Thermal Effects in Wetlands 3-27
Constituent Characteristics 3-29
Aquatic Vegetation , 3-30
Types of Wetland Vegetation..... 3-30
Vegetation Patterns 3-31
Role of Aquatic Plants in Controlling Treatment Processes 3-34
Section 4 Performance Expectations 4-1
Approach to Performance Evaluation 4-1
Methodology of Performance Evaluation 4-2
BOD Performance 4-4
Database Assessment 4-4
Temporal BOD Performance 4-5
BOD Permit Compliance 4-9
TSS Performance 4-10
Database Assessment 4-10
Temporal TSS Performance 4-12
TSS Permit Compliance 4-13
Nitrogen Performance 4-14
Organic Nitrogen Performance 4-14
Ammonia Nitrogen Performance 4-15
Total Kjeldahl Nitrogen Performance 4-16
Nitrate and TIN Performance 4-18
Total Nitrogen Performance 4-19
Nitrogen Permit Compliance 4-21
Ammonia Nitrogen 4-21
Total Nitrogen 4-21
-------
TABLE OF CONTENTS
Section 4 Performance Expectations (continued)
Total Phosphorus Performance 4-22
Database Assessment 4-22
Temporal Phosphorus Performance 4-24
Total Phosphorus Permit Compliance 4-25
Fecal Coliform Performance 4-26
Database Assessment 4-26
Temporal Fecal Coliform Performance 4-28
Fecal Coliform Permit Compliance 4-28
Metals 4-28
Other Performance Considerations 4-30
Wetland Background Concentrations 4-30
Natural Variability 4-31
Section 5 System Planning and Design Considerations 5-1
Planning Considerations 5-1
Role of Wetlands in the Watershed 5-2
Additional Benefits/Habitat Considerations 5-5
Effluent Quality Considerations 5-5
Wetland Treatment System Objectives 5-6
Permitting 5-7
Public Access 5-9
Hydrological Considerations 5-10
Precipitation and Evapotranspiration 5-10
Groundwater .....5-11
Ice and Snow 5-11
Engineering Considerations 5-11
Pre-Treatment Requirements 5-11
Soils, Slope, and Subsurface Geology 5-11
Percolation and Use of Liners 5-12
Inlet/Outlet Types and Placement 5-12
Wildlife/HabitatConsideration 5-13
Environmental Impact 5-14
Land Use..... 5-14
Insect Vectors ., 5-14
Odors 5-15
Wildlife and Ecological Attractive Nuisances 5-16
FWS Wetlands & Bird Strike Issues 5-16
Wetland Sizing.. 5-16
Approaches to Sizing 5-16
Assessment of Predictive Equations 5-17
Areal Loading Rate Method 5-19
Design Approach to Sizing 5-20
in
-------
TABLE OF CONTENTS
Section 6 Lessons Learned and Recommendations 6-1
Information Management , 6-1
Database Maintenance and Analysis 6-1
Planning 6-3
Multiple Benefits and Public Access 6-3
Environmental Education and Interpretation Centers 6-4
Open Water/Emergent Vegetation Ratio 6-5
Site Topography and Soils....... 6-6
Wetland Hydraulics 6-7
Inlet/Outlet Structures 6-7
Flow Measuring Devices 6-8
Internal Drainage 6-8
Engineering 6-9
Berm Construction and Specifications 6-9
Wetland Configuration and Shape 6-10
Sediment Storage Zone at Inlet.. 6-10
Wetland Planting 6-11
Impermeable Barrier and liner Materials 6-14
Operation and Maintenance 6-14
Management of FWS Constructed Wetlands 6-15
Potential Nuisance Conditions 6-16
Vegetation Management Implications 6-16
Mosquito Control 6-17
Process Control 6-18
Monitoring Requirements 6-18
Considerations for Minimizing Variability in Effluent Quality , 6-21
Research Studies 6-21
Critical Research Issues 6-22
Appendix A - References
IV
-------
TABLE OF CONTENTS
List of Tables
TABLE 1-1 Timeline of selected events in wetland treatment technology (adapted
from Kadlec and Knight 1996) 1-7
TABLE 1-2 Percentage distribution of NADB FWS treatment systems by wetland
type and level of pretreatment 1-10
TABLE 2-1 Listing of major treatment wetland conferences,., „ 2-2
TABLE 2-2 EPA Publications on Free Water Surface Treatment Wetlands 2-3
TABLE 2-3 Books with focus on Free Water Surface Treatment Wetlands - in
chronological order 2-4
TABLE 2-4 Journals that regularly publish articles dealing with treatment
wetlands , 2-4
TABLE 2-5 Desired Minimum information/Criteria for FWS Wetland Systems 2-5
TABLE 2-6 FWS Wetlands used for performance evaluation (Technology
Assessment Sites) .2-6
TABLE 2-7 Panelists for the Mesa, Arizona, workshop held February 2 through
4,1996 2-7
TABLE 3-1 Mechanisms and factors that affect the potential for removal or
addition of water quality constituents in FWS wetlands (Adapted
from Stowell et al. 1980) 3-14
TABLE 3-2 Some common wetland plants and depths of occurrence used in FWS
and floating aquatic constructed wetland , 3-31
TABLE 3-3 Submerged surface area of wetland vegetation, normalized for a
depth of 0.5m 3-35
TABLE 4-1 Water quality constituent data availability for the FWS constructed
wetland systems included in this assessment, identified in Table 2-6 4-3
TABLE 4-2 Summary of performance data and loadings for systems analyzed in
this assessment (listed in Table 4-1) , .4-4
TABLE 4-3 Total phosphorous removal rates for non-forested treatment
wetlands 4-22
TABLE 4-4 Metal removal data from free water surface treatment wetlands 4-29
TABLE 4-5 Long-term average annual outflow concentrations for lightly loaded
FWS wetlands in the NADB , 4-31
TABLE 4-6 Expected range of background concentrations for constituents of
interest... 4-31
TABLE 5-1 Equations used to compute the performance of FWS constructed
wetlands... , .....5-17
TABLE 5-2 Range of areal loading rates for FWS constructed wetlands 5-20
TABLE 6-1 Percent of dominant plant species areal coverage of the Enhancement
Wetlands of the Arcata Marsh and Wildlife Sanctuary 6-13
TABLE 6-2 Suggested minimum monitoring requirements for a FWS constructed
wetland 6-20
-------
TABLE OF CONTENTS
List of Figures
FIGURE 1-1 Definition sketches for constructed wetlands: (a) free water surface
constructed wetland with emergent vegetation, (b) free water surface
wetland with an open water zone/ and (c) constructed floating aquatic
plant treatment system (adapted from Kadlec and Knight 1996) 1-4
FIGURE 1-2 Ecosystem and communities of a FWS (USEPA 1993b) 1-6
FIGURE 1-3 Percentage of all communities utilizing FWS constructed wetlands
based upon community size (NADB, n = 135), ....1-11
FIGURE 1-4 Distribution of FWS constructed wetlands utilized for treating
wastewater by State - not including pilot projects or demonstration
projects 1-11
FIGURE 2-1 Influent BOD loading rates for FWS Wetland Systems in the NADB 2-9
FIGURE 3-1 Components of overall wetland water mass balance (Kadlec 1993) 3-2
FIGURE 3-2 Total annual losses (+) and gains (-) from evapotranspiration and
precipitation in cm (ET-P) (Flach, 1973) .....3-5
FIGURE 3-3 Monthly water budget for Arcata's wastewater treatment plant
(Arcata, California) showing the effects of precipitation and
evapotranspiration on the water budget in a Coastal wetland system
and the monthly water budget for the Tres Rios Hayfield Site basin
HI (Phoenix, Arizona) showing the influence increased ET and
reduced precipitation Has in arid regions 3-6
FIGURE 3-4 Conceptual partitioning of treatment processes through a FWS
wetland 3-16
FIGURE 3-5 Wetland TSS removal, resuspension, and internal generation
processes 3-17
FIGURE 3-6 Simplified portrayal of wetland carbon processing. Incoming BODS
is reduced by deposition of partieulate forms and by microbial
processing in floating, epiphytic, and benthal litter layers.
Decomposition processes create a return flux 3-19
FIGURE 3-7 BOD and COD effluent concentration before and during tap water
loading to Arcata Pilot Project wetland 3-20
FIGURE 3-8 Vertical distribution of DO in a submergent plant zone of the Arcata
Enhancement Marsh..... 3-21
FIGURE 3-9 Vertical distribution of DO in an emergent plant zone of the Arcata
Enhancement Marsh .3-22
FIGURE 3-10 Nitrogen transformation processes in wetlands (Gearheart 1998,
unpublished data) 3-23
FIGURE 3-11 Influent and effluent phosphorus in the Arcata Pilot Project I FWS
wetlands, Second Pilot Project, 1982. Cell 5 was loaded at 0.75
kg/ha-d, and Cell 3 at 0.15 kg/ha-d (Gearheart 1993) 3-24
VI
-------
TABLE OF CONTENTS
FIGURE 3-12 Conceptual cycling of phosphorus forms in FWS constructed
wetlands. SRP: Soluble reactive phosphorus; POP: particulate organic
phosphorus; TSS-POP: form of POP in terms of a fraction of the total
suspended solids 3-25
FIGURE 3-13 Hydrogen ion (pH) buffering in system 3 at Listowel (Herskowitz
1986) 3-26
FIGURE 3-14 Metal sulfide burial processes in a wetland (Meyers 1998, personal
communication) ..3-27
FIGURE 3-15 Correlation between wetland water temperature and air
temperatures. Both northern (Listowel) and southern (Orlando
Easterly) systems show water temperatures that follow the mean
daily air temperature during warm months from nearby weather
stations (Kadlec and Knight 1996) 3-28
FIGURE 3-16 Distribution of BOD and COD concentration by form (settleable,
supracolloidal, or soluble) in oxidation pond effluent and treatment
marsh effluent from Arcata, California (Gearheart 1992) 3-30
FIGURE 3-17 Newly constructed wetlands require a startup period to attain full
vegetative cover. Ground level and aerial reconnaissance were used
to follow this process for the Tarrant County Project (Alan Plummer
Associates Inc. [APAI] 1995) ,. 3-32
FIGURE 3-18 Coverage of plants during the startup period of the Arcata Pilot
Project wetland 3-33
FIGURE 3-19 Stem, leaf and litter cumulative surface area for Typha spp. in
Houghton Lake discharge zone wetland (Kadlec, 1997) 3-36
FIGURE 3-20 Stem and leaf surface area for Scirpus acutis (hardstem bulrush)
and Typha latifolia (cattail) in Arcata Treatment Wetland (Gearheart et
al., 1999, publication in progress) 3-37
FIGURE 4-1 Average BOD loading rate versus effluent BOD concentration for
TADB sites 4-5
FIGURE 4-2 Monthly influent and effluent BOD values for Arcata's treatment
wetland 4-5
FIGURE 4-3 Monthly influent and effluent BOD values for Arcata's
enhancement wetland 4-6
FIGURE 4-4 Influent and effluent monthly BOD cumulative probability values
for West Jackson County, Mississippi 4-6
FIGURE 4-5 Influent and effluent monthly BOD for Lakeland, Florida. 4-7
FIGURE 4-6 Influent and effluent monthly BOD cumulative probability for Fort
Deposit, Alabama 4-7
FIGURE 4-7 Monthly BOD loading rate versus BOD effluent concentration for
Arcata Treatment Marsh 4-8
FIGURE 4-8 Cumulative monthly mass influent and effluent BOD for the Arcata
Treatment Wetland. The area between the two curves is
representative of the mass of BOD removed 4-9
Vll
-------
TABLE OF CONTENTS
FIGURE 4-9 Monthy TSS loading versus effluent TSS concentration for TADB
wetland systems 4-11
FIGURE 4-10 Cumulative probability distribution of monthly influent and
effluent TSS concentration for Fort Deposit wetland , 4-11
FIGURE 4-11 Weekly transect TSS concentration for Arcata's Cell 8 Pilot Project
with theoretical retention time of 6 days, receiving oxidation pond
effluent , 4-12
FIGURE 4-12 Weekly Influent and effluent TSS concentration for Arcata
Enhancement Wetland 4-12
FIGURE 4-13 Cumulative yearly mass influent and effluent TSS for Arcata
Treatment Wetland 4-13
FIGURE 4-14 Cumulative probability distribution of influent and effluent
organic nitrogen for West Jackson County, Mississippi 4-15
FIGURE 4-15 Ammonia nitrogen loading versus effluent ammonia
concentrations for TADB systems 4-15
FIGURE 4-16 Cumulative probability distribution of monthly influent and
effluent ammonia nitrogen from Beaumont, Texas. 4-16
FIGURE 4-17 Ammonia nitrogen removal for Beaumont, Texas, through 8 cells
with a total HRT of 17 days 4-16
FIGURE 4-18 Total Kjeldahl nitrogen loading versus effluent ammonia
concentrations for the TADB 4-17
FIGURE 4-19 Cumulative probability distribution of monthly influent and
effluent TKN from Central Slough, South Carolina 4-17
FIGURE 4-20 Nitrate nitrogen loading versus effluent nitrate concentrations for
the TADB 4-18
FIGURE 4-21 Cumulative probability distribution of monthly influent and
effluent nitrate concentrations for Orange County, Florida 4-19
FIGURE 4-22 Monthly influent and effluent of total inorganic nitrogen (TIN) for
the Arcata Enhancement Wetland 4-19
FIGURE 4-23 Total nitrogen loading versus effluent total nitrogen concentrations
for TADB wetland systems 4-20
FIGURE 4-24 Range of monthly inlet and outlet TN concentrations for cells 1
through 12 at the Iron Bridge FWS wetland near Orlando, Florida 4-20
FIGURE 4-25 Total phosphorus loading versus effluent phosphorus
concentrations for the TADB FWS systems 4-23
FIGURE 4-26 Cumulative probability distribution of monthly influent and
effluent total phosphorus concentrations for Central Slough, South
Carolina 4-24
FIGURE 4-27 Phosphorus pulsing, as illustrated in a pilot cell in Arcata,
California. Marsh 1 received tap water until June 1982 (no
phosphorus load), while Marsh 3 received oxidation pond effluent
(Gearheartl993).... 4-25
FIGURE 4-28 Influent FC versus effluent FC for the TADB systems 4-26
VJll
-------
TABLE OF CONTENTS
FIGURE 4-29 Cumulative probability distribution of influent and effluent fecal
coliform from Arcata Pilot Project Cell 8, CA (Gearheart et al. 1986). 4-27
FIGURE 4-30 Cumulative probability distribution fecal coliform from Arcata
Enhancement Wetland, California (Gearheart 1998, unpublished
data) 4-27
FIGURE 4-31 Variation in effluent BOD at the Arcata Enhancement Marsh 4-30
FIGURE 5-1 Diagram of a methodology for determining the appropriateness of
the use of a constructed wetland and the factors necessary for the
design of a multi-use constructed free surface wetland 5-4
FIGURE 5-2 Annual average area! BOD loading rate vs. annual average effluent
BOD concentration for NADB systems „ 5-20
IX
-------
TABLE OF CONTENTS
List of Equations
i
(3-1) ^-=Ql -Q0+Qc-Qb+Qsm+(P-ET-i)*A 3-2
at
Vs
(3.2) t=— 3-8
(3-3) Qave= ' ° 3-9
(3-4) q=-3- 3-9
(4-1) Ce = 3.42 + 0.262 Q „ 4_g
(5-1) — = -kap.C 5-16
dt
(5-2) C,=C0exp"kiS>t... 5-16
(5-3) kr = k2flecr'JO> 5-16
-------
TABLE OF CONTENTS
List of Acronyms and Symbols
ADEM
ADEQ
ASCE
AWRA
BOD
BOR
CBOD
CFU/100 mL
COE
C
cm
CT
d
DO
DP
EFF
ET
F
FAC
FAP
PC
FWS
ha
HEC2
HRT
IAW
IAWQ
kg
kg/ha-d
L
m
mm
mg/L
min
mL
MPN
NADB
NAWCC
NH3-N
NH4-N
NO3-N
NOD
NPDES
OrgN
pH
Alabama Department of Environmental Management
Arizona Department of Environmental Quality
American Society of Civil Engineers
American Water Resources Association
Biochemical oxygen demand
Bureau of Reclamation
Carbonaceous biochemical oxygen demand
Colony-forming units per one hundred miUiliters
U.S. Army Corps of Engineers
Centrigrade
centimeter
Crites, Tchobanoglous Model
Day
Dissolved oxygen
Dissolved phosphorus
Concentration reduction efficiency
Evapotranspiration
Farenheit
Florida Administrative Code
Floating aquatic plants
Fecal coliform
Free water surface
Hectare
U.S. Army Corps of Engineers computer program
Hydraulic residence time
pg.2-3
International Association on Water Quality
Kilogram
Kilogram per hectare per day
Liter
Meter
millimeter
Milligram per liter
minute
Microgram per liter
Milliliter
Maximum probable nitrogen
North American Treatment Wetland Database
North American Wetlands Conservation Council
Ammonia nitrogen
Ammonia nitrogen
Nitrate nitrogen
Nitrogenous oxygen demand
National Pollutant Discharge Elimination System
Organic nitrogen
hydrogen ion
-------
TABLE OF CONTENTS
PFR Plug flow reactor
POP Particulate organic phosphorus
ppb Part per billion
ppm Part per million
RCM Reed, Crites, Middlebrooks Model
RED Mass reduction efficiency
s second
SCDHEC South Carolina Department of Health and Environmental
Control
SFWMD South Florida Water Management District
SSF Subsurface flow wetlands
SRCSD Sacramento Regional County Sanitation District
SRP Soluble reactive phosphorus
TADB Technology Assessment Database
TIN Total inorganic nitrogen
TKN Total Kjeldahl nitrogen
TMDL Total maximum daily limit
TN Total nitrogen
TP Total phosphorus
TSS Total suspended solids
UV Ultraviolet
TVA Tennessee Valley Authority
USEPA U. S. Environmental Protection Agency
WEF Water Environment Federation
WPCF Water Pollution Control Federation
yr Year
XII
-------
TABLE OF CONTENTS
Acknowledgments
This report was prepared by the Environmental Resources Engineering Department of
Humboldt State University, Arcata, California, and by CH2M-HU1. Special
acknowledgment goes to George Tchobanoglous and Roland Wass for reviewing and
editing the final draft reports.
The following individuals reviewed the multiple draft documents and contributed
constructively to the final report.
Robert Bastian - USEPA Washington, D.C.
Robert Knight - Gainesville, FL
Jim Kriessl - USEPA Cincinnati, OH
Eric Stiles - Bureau of Reclamation, Denver, CO
Principal funding for this work was provided by the U.S. Environmental Protection
Agency, under an Environmental Technology Initiative (ETI) grant to the U.S. Bureau of
Reclamation for support of technology development related to the Tres Rios multi-
purpose constructed wetland project in Phoenix, Arizona. Project sponsors include
Robert Bastian and Robert E. Lee of the U.S. Environmental Protection Agency
Municipal Technology Branch, Marvin Murray of the U.S. Bureau of Reclamation, and
Paul Kinshella of the City of Phoenix. A number of treatment wetland practitioners
participated in a technology assessment workshop in Mesa, Arizona, in February 1996
and in March 1997, supplied information for this report, and provided a peer review of
the final draft report.
Data summarized in this assessment were obtained from several sources. The North
American Treatment Wetland Database (NADB) report prepared by Robert Knight,
Robert Kadlec, and Sherwood Reed under contract to the U.S. Environmental Protection
Agency provided an initial point of entry into selecting sites to be brought up to date
and for systems that met the data quality criteria. Project officers for the NADB project
were Mary E. Kentula and Richard Olson at the Environmental Research Laboratory in
Corvallis, Oregon, and Donald Brown at the Risk Assessment Engineering Laboratory in
Cincinnati, Ohio. In addition, considerable data were obtained from owners,
consultants, and researchers working on FWS constructed wetlands. Listed below are
the wetland systems added to the database and individuals that provided data for this
report.
Gustine, CA Mac Walker - Larry Walker & Assoc., Davis, CA
Ouray, CO Tom Andrews - Southwest Wetlands, Santa Fe, NM
Hemet, CA Stella Denison - Eastern Munic. Wtr. Dist, Hemet, CA
Sacramento Regional, CA Glen Dombeck - Nolte & Assoc., Sacramento, CA
Columbia, MO Robert Kadlec - Wetland Mgmt Services, Chelsea, MI
Houghton Lake, MI Robert Kadlec - Wetland Mgmt. Services, Chelsea, MI
Minot, ND Don Hammer - Norris, TN
Kill
-------
TABLE OF CONTENTS
Lakeland, FL Robert Knight - CH2M-HU1, Gainesville, FL
Mount Angel, OR John Yarnall - WesTech, Salem, OR
Arcata, CA Robert Gearheart - Humboldt State Univ., Arcata, CA
Phoenix, AZ Roland Wass - City of Phoenix, AZ
West Jackson Co., MI Bill Raddey - MGCRWA, MI
Manila, CA Wiley Buck - Manila Community Serv. Dist., Manila, CA
Beaumont, XX Bill Benner - Beaumont, TX
The principal authors of the final report were Robert Gearheart, Brad Finney, Margaret
Lang, Jeffrey Anderson, and Sophie Lagace of Humboldt State University, while final
editing was provided by Roland Wass, PBS&J.
Not all reviewer comments were able to be resolved in this final
document. Participation by any specific individual in the
drafting and review process does not constitute approval or
endorsement of the technical content of this report. The U.S.
Environmental Protection Agency and the editors of the final
report accept full responsibility for any errors or omissions in
this document.
XIV
-------
SECTION 1
Introduction to Free Water Surface
Treatment Wetlands
The purpose of Free Water Surface Wetlands for Wastewater Treatment: A Technology
Assessment is to assess the application, performance, and scientific knowledge of
free water surface (FWS) wetlands to treat municipal wastewater and to meet
other societal and ecological needs. This report is not intended to cover subject
areas to the extent needed for actual design and operation. Rather, the objective
of this assessment is to produce a document that public works engineers,
consulting engineers, regulatory agency representatives, researchers and citizens
can use to evaluate the feasibility of FWS treatment wetland technology. The
scope of this document includes a summary of the treatment processes operating
in FWS treatment wetlands, a summary and evaluation of FWS treatment
wetland performance, and discussion of important issues in the planning,
design, and operation of FWS treatment wetlands.
Background
Free water surface wetlands have been engineered for water quality treatment in
the United States since the early 1970s. Design information and operational
performance data for these systems have been accumulating since that time and
has led to the rapid development of a growing collection of literature. A number
of efforts have been undertaken to summarize information from diverse data
sources into a collection of performance descriptions. The most complete effort
to date was the development of the North American Constructed Wetland
Database (NADB) funded by the U.S. Environmental Protection Agency (EPA)
(Knight et al. September 1993, NADB 1993, Brown and Waterman 1994).
The next step in assessing the performance of FWS treatment wetlands was to
compile the assembled data into a summary of the state of knowledge. This
technology assessment report describes the current understanding of processes
and the performance of FWS treatment wetlands. In addition, areas of
inadequate understanding are identified. The findings of this technology
assessment will be incorporated into an update (in progress) of the U.S.
Environmental Protection Agency's (EPA) FWS constructed wetland design
manual (EPA 1988a) and the Water Environment Federation (WEF) Manual of
Practice on Natural Systems (WEF, 1999), currently in preparation. Further, in
the time period since the data analysis was performed for this assessment, many
additional treatment wetland systems have become operational. Some of these
systems have operation and performance data that are currently being used by
researchers at Humboldt State University to update and provide a web-based
version of the NADB by the end of 2000.
1-1
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
In all, three draft technical assessment documents have been prepared. A
technical review team comprised of researchers, USEPA representatives,
consultants, Bureau of Reclamation representatives, Corps of Engineers
representatives, and municipal representatives extensively reviewed each
document. This final document, Free Water Surface Wetlands for Wastewater
Treatment: A Technology Assessment, is a culmination of an extensive effort to
create an accessible summary of the operating principles and performance
expectations of FWS treatment wetlands for wastewater treatment.
Introduction to the Technology
Wastewater polishing systems utilizing wetland plants have proven to be very
reliable. Wetland plants create an environment that supports a wide range of
physical, chemical, and microbial processes. These processes separately and in
combination remove total suspended solids (TSS), reduce the influent
biochemical oxygen demand (BOD), transform nitrogen species, provide storage
for metals, cycle phosphorus, and attenuate organisms of public health
significance. The biogeochemical cycling of macro and micronutrients within
the wetland is the framework for the treatment capability of a wetland system.
Valiela et al. (1976) describe the wastewater treatment capacity of natural
wetlands as follows:
"Wetlands seem to be better processors of wastes than estuaries
and coastal waters. It might be feasible to safely dispose of
effluents under carefully controlled conditions on marshlands
rather than deeper coastal areas where the elimination of
contaminants is not as effective and dispersal of contaminants is
more likely. We would like to emphasize, however, that the
wetland properties outlined above, and the consequent effects on
nutrients, heavy metals, hydrocarbons, and pathogens are features
of wetlands as they function naturally. They are in fact providing
free waste treatment for contaminated waters already."
Natural wetlands are ecosystems that occur in areas that are intermediate
between uplands and deep-water aquatic systems. Technical and regulatory
definitions of wetlands focus on the dependence of wetland ecosystems on
shallow water conditions which result in saturated soils, low dissolved oxygen
(DO) levels or anaerobiosis in soils, and colonization by adapted plant and
animal communities (Cowardin et al. 1979, Mitsch and Gosselmk 1993). The
ability of wetland ecosystems to improve water quality naturally has been
recognized for more than 30 years (Seidel 1964), During this period, the use of
constructed wetlands has evolved from a research concept to a relatively
successful, and increasingly popular, pollution control technology
(Tchobanoglous 1993; Kadlec and Knight 1996; Kadlec et al 2000; EPA 1999a).
1-2
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Treatment Wetland Forms and Functions
Three general types of shallow vegetated ecosystems are used for water quality
treatment: (1) free water surface (FWS) wetlands, (2) subsurface flow (SSF)
wetlands, and (3) floating aquatic plant (FAP) treatment systems. All three of
these vegetated treatment systems are operating in the U.S. for water quality
improvement. Early performance information for system types has been
published in a previous design manual (EPA 1988a), and a subsurface flow
technology assessment has already been completed (EPA 1993a). This
technology assessment report focuses only on the FWS treatment wetland
technology (Figure 1-1). In FWS treatment wetlands, water flows over the soil
surface from an inlet point to an outlet point or, in rare cases, water is completely
lost to evapotranspiration and infiltration within the wetland.
The technology began with the ecological engineering of natural wetlands for
wastewater treatment (Ewel and Odum 1984, Kadlec and Tilton 1979).
Constructed FWS wetlands are designed to mimic the hydrologic regime of
natural wetlands. Currently, application of the FWS treatment wetland
technology is almost exclusively through the construction of new FWS wetlands
designed to meet specific influent levels and effluent water quality goals and to
potentially enhance ancillary benefits associated with treatment wetland
systems.
This technology assessment includes performance data from both natural and
constructed free water surface wetlands. Such systems are similar in overall
function with some important exceptions. The principal differences between
natural and constructed treatment wetlands are structural—natural wetlands are
more likely to have a forested plant community and to include a well-developed
organic soil component than constructed wetlands. Natural wetlands are more
likely to be subject to variable inflows and water depths and have more stagnant
water zones outside the primary flow path that can reduce treatment efficiency.
Also, hydraulic efficiency, the ability to utilize the entire wetland area in the
process of water treatment, can be more nearly optimized in constructed
wetlands than in most natural wetlands.
Free water surface treatment wetlands function as land-intensive wastewater
treatment systems. Inflow water containing particulate and dissolved pollutants
slows and spreads through a large area of shallow water and emergent
vegetation. Particulates (typically measured as total suspended solids [TSS]) are
trapped and tend to settle due to lowered flow velocities and sheltering from
wind. The solids contain biodegradable organic matter, typically measured as
biochemical oxygen demand (BOD) components, fixed forms of total nitrogen
(TN) and total phosphorus (TP), and trace levels of metals and other recalcitrant
synthetic orgardes. These insoluble pollutants enter the biogeochemical cycles
within the water column and surface soils of the wetland. Colloidal materials
are subject to flocculation and are removed partially with the particulate fraction
described above. At the same time, soils and active microbial and plant
populations throughout the wetland environment sorb a fraction of the
dissolved BOD, TN, TP, and trace elements. These dissolved constituents also
enter the overall mineral cycles of the wetland ecosystem.
1-3
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
FIGURE 1-1
Definition sketches for constructed wetlands: (a) free water surface constructed wetland with
emergent vegetation, (b) free water surface wetland with an open water zone, and (c) constructed
floating aquatic plant treatment system (adapted from Kadlec and Knight 1996).
Distribution Pipe
Outlet Weir
/ / / /
Low Permeability Soil
Free Water Surface (Surface Flow)
Distribution Pipe
Outlet Weir
Lined Basin
Free Water Surface with Open Water Zone
Distribution Pipe
Outlet Weir
/ / s / / S// / / / / / / /
Lined Basin
Floating Aquatic Plant System
During the process of elemental cycling within the wetland, chemical free energy
is extracted by the heterotrophic biota, and fixed carbon and nitrogen are lost to
the atmosphere. A smaller portion of the phosphorous and other non-volatile
elements can be lost from the mineral cycle and buried in accreting sediments
within the wetland. Wetlands are autotrophic ecosystems, and the additional
carbon and nitrogen fixed from the atmosphere is processed simultaneously
with the pollutants introduced from the wastewater source. The net effect of
these complex processes is a general reduction in pollutant concentrations
between the inlet and outlet of the treatment wetland.
1-4
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Free water surface treatment wetlands have some properties in common with
facultative lagoons, but also have many important structural and functional
differences. Water column processes in the open-water zones within FWS
wetlands are nearly identical to similar zones within ponds. At the surface, an
autotrophic zone dominated by planktorac or filamentous algae or by floating or
submerged aquatic macrophytes limits light to the deep zones. The absence of
light in the deeper zones in both systems causes them to be dominated by
anaerobic microbial processes. The submerged macrophytes in deep zones also
afford sites for colonization of periphytic bacteria and provide substrate for
algae-biofflm development.
The shallow, emergent macrophyte zones present in FWS wetlands operate quite
differently than any zone within a facultative lagoon. Emergent wetland plants
tend to shade the water surface reducing algae growth and limiting water
reaeration processes that add dissolved oxygen to the water column. Secondary
populations of duckweed, covering the water surface and held in place by
emergent plants, may also hinder reaeration. Net carbon production in
emergent wetlands tends to be high compared to facultative ponds because of
much greater primary production of plant carbon. High production of plant
carbon and the resistance of plant carbon to degradation combines with a low
organic carbon decomposition rate in the oxygen deficient water column to
create significant differences in biogeochemical cycling rates in wetlands
compared to ponds and lagoons.
Other Benefits of Treatment Wetlands
In addition to water quality benefits, wetland systems have also been designed
and operated to provide wetland habitat for waterfowl and other wildlife (see
Figure 1-2). Many FWS treatment wetland systems are operated as wildlife
refuges or parks, as well as part of wastewater treatment, reuse, or disposal
systems (Wilhem et al. 1989, Gearheart et al. 1989, Knight 1997, EPA 1999b). In
some cases, FWS constructed wetland systems provide an area for public
education (interpretive center or informative displays) and outdoor recreation
(walking, jogging, bird watching). The design of multiple purpose use FWS
constructed wetlands has been significant. More than 40 percent of the NADB
secondary and 36 percent of the NADB tertiary treatment applications identified
one or more additional benefits beyond that of water quality improvements.
Some ecological benefits can be claimed for nearly all FWS constructed wetland
systems regardless of their stated objectives. Benefits are often claimed for FWS
constructed wetlands in areas where wetlands have been lost or degraded such
as the facultative ponds in the north central United States (South Dakota and
North Dakota) where existing degraded wetlands are used for seasonal storage.
Sometimes, the ancillary benefits of treatment wetlands work counter to those
processes that improve the water quality. For example, some treatment
wetlands are home, at least on a seasonal basis, for 1000s of birds and sometimes
100s of mammals, depending upon the location and scale of the system. While
residing within the treatment wetlands, their activities can add bacteria and
1-5
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
nutrients to the system. Wildlife activity may also re-suspend bottom sediments
increasing turbidity and potentially causing the export of nutrients, inorganic,
and organic constituents from the wetland. Wildlife induced water quality
degradations are often mitigated, however, due to design factors necessary to
achieve other water quality goals; e.g., providing flow time through a vegetated
emergent zone prior to discharge for algal control or denitrification.
FIGURE 1-2
Ecosystem and communities of a FWS (USEPA1993b).
Historical Development of the Technology
Treatment wetland technology using FWS wetlands has been under
development, with varying success, for nearly 30 years in the United States
(Table 1-1). Early laboratory studies in Germany examined the effects of
emergent plants on removal of organic compounds in industrial wastewater
(Seidel 1976). Constructed estuarine ponds with wetland vegetation were
loaded with municipal wastewater during the 1960s and early 1970s in North
Carolina (Odum 1985). Large-scale engineered natural wetland systems
receiving pretreated municipal wastewater were studied in Michigan (Kadlec et
al. 1993) and Florida (Ewel and Odum 1984) beginning in the early to mid-1970s.
Constructed marsh-pond-meadow systems were under study at the same time in
New York (Small and Wurm 1977). These research programs led to an
1-6
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
TABLE 1-1
Timeline of selected events in wetland treatment technology (adapted from Kadlec and Knight 1996).
Date
Location
Description
Selected Research Efforts
1952-Iate Plon, Germany
1970s
1967-1972 Morehead City, NC
1 971 -1 975 Woods Hole, MA
1972-1977
1973-1974
1973-1975
1973-1976
Houghton Lake, Ml
Dulac, LA
Seymour, WI
Brookhaven, NY
1973-1977 Gainesville, FL
1974-1975 Brillion, WI
1975-1977 Trenton, NJ
1976-1979 Eagle Lake, IA
1976-1982 Southeast Florida
1979-1982 Humboldt, SK
1980-1984 LIstowel, Ontario
1979-1982 Arcata, CA
Removal of phenols and dairy wastewater treatment
with bulrush plants by K, Seidel and R. Kickuth
Constructed estuarine ponds and natural salt marsh
studies of municipal effluent recycling by H.T. Odum
and associates
Potential of natural salt marshes to remove nutrients,
heavy metals, and organics was studied by I. Valiela,
J.M. Teal and associates
Natural wetland treatment of municipal wastewater by
R.H. Kadlec and associates
Discharge of fish processing waste to a freshwater
marsh by J.W, Day and coworkers
Pollutant removal in constructed marshes planted with
bulrush by Spangler and coworkers
Meadow/marsh/pond systems by M.M. Small and
associates
Cypress wetlands for recycling of municipal
wastewater by H.T. Odum, K. Ewel, and associates
Phosphorus removal in constructed and natural marsh
wetlands by F.L. Spangler and associates
Small enclosures in the Hamilton Marshes (freshwater
tidal) were irrigated with treated sewage by Whigham
and coworkers
Assimilation of agricultural drainage and municipal
wastewater nutrients in a natural marsh wetland by
C.B. Davis, A.G. van der Valk, and coworkers
Nutrient removal in natural marsh wetlands receiving
agricultural drainage waters by F.E. Davis, A.C.
Federico, A.L. Goldstein, S.M. Davis, and coworkers
Batch treatment of raw municipal sewage in lagoons
and wetland trenches by Lakshman and coworkers
Constructed marsh wetlands were tested for treatment
of municipal wastewater under a variety of design and
operating conditions by Wile and associates
Pilot wetland treatment system for municipal
wastewater treatment by Gearheart and coworkers
1-7
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Date
Location
Description
1974-1988 NSTL Station, MS
1980 - 1989 Walt Disney World, FL
1986 Orlando, FL
1979-1998
1981-1984
1993
1994
1994
San Diego, CA
Santee, CA
Hemet, CA
Tres Rios, AZ
Sacramento, CA
Selected Full-Scale Projects
1972 Bellaire, M!
1973
1974
1975
1977
1978
1979
1979
1984
1986
1987
1987
1987-1988
1988
Mt. View, CA
Othfresen, West
Germany
Mandan, ND
Lake Buena Vista, FL
Houghton Lake, Ml
Dmmmond, Wl
Show Low, AZ
incline Village, NV
Arcata, CA
Orlando and Lakeland,
FL
Myrtle Beach, SC
Benton, Hardin, and
Pembroke, KY
Hayward, CA
Gravel-based, subsurface flow wetlands tested for
recycling municipal wastewaters and priority pollutants
by B.C. Wolverton and coworkers
Pilot-scale wetland work on a variety of wetland plants
Tom Debusk
Aquatic Plants for Water Treatment and Resource
Recovery by Ramesh Reddy, and Smith (1987)
1 mgd demonstration of treatment effectiveness of
water hyacinths as a front end to the raw wastewater
to potable water project
Subsurface flow wetlands were tested for treatment of
municipal wastewater by R.M. Gersberg and
coworkers
Effluent polishing, groundwater recharge
Metals removal, effluent polishing, groundwater
recharge, ecosystem restoration
Metals removal, ammonia reduction, temperature
reduction
Natural forested wetland receiving municipal
wastewater
Constructed wetlands for municipal wastewater
treatment
Full-scale reed marsh facility treating municipal
wastewater in an old quarry
Constructed ponds and marshes to treat runoff and
pretreated process wastewater from an oil refinery
Natural forested wetland was used for year-round
advanced treatment and disposal of up to
27,700 cubic meters per day (rrf/d) of municipal
wastewater
Natural peatland receiving summer flows of municipal
wastewater
Sphagnum bog receiving summer flows from a
facultative lagoon
Constructed wetland ponds for municipal wastewater
treatment and wildlife enhancement
Constructed wetlands for total assimilation (zero
discharge) of municipal effluent
Constructed marsh wetlands for municipal wastewater
treatment, wetland creation, and wildlife enhancement
Two large (> 480 ha) constructed wetlands for
municipal treatment
Natural Carolina bay wetlands for municipal
wastewater treatment
Constructed wetlands for municipal wastewater
treatment designed by the Tennessee Valley Authority
Rve basin 70 ha wetland for wildlife enhancement
1-8
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Date
1988
1989
1990
1991
1991
1991
1993
1993
1993
1995
1997
Location
Orange County, FI_
Sisseton, SD
W. Jackson County, MS
Columbus, MS
Huron, SD
Minot, ND
Everglades, FL
Beaumont, TX
Ouray, CO
Hidden Valley
(Riverside), CA
Cheney, WA
Description
Hybrid treatment system combining constructed and
natural wetland units
102 ha total assimilation wetland treating municipal
wastewater
Wildlife refuge linkage
First full-scale constructed wetland for advanced
treatment of pulp and paper mill wastewater
132 ha total assimilation wetland treating municipal
wastewater
Northern surface flow wetland (51 .2 ha) system for
municipal treatment during 180-day discharge season
Treatment of phosphorus in agricultural runoff in a
1 ,380-ha constructed filtering marsh
Large (263 ha) constructed marsh for municipal
wastewater polishing and public use
Effluent polishing
Nitrogen removal, wetland restoration, wildlife habitat,
groundwater recharge
Wildlife enhancement, groundwater recharge
increasing number of research and full-scale treatment wetland projects treating
a variety of wastewater from municipal, industrial, and agricultural sources.
Many of the earliest treatment wetlands were subsurface flow systems
constructed in Europe to treat partially pretreated municipal wastewater. Soil
and gravel-based subsurface flow wetlands are still the most prevalent
application of this technology in Europe and the United Kingdom (Cooper 1990,
Brix 1994a, EPA 1993a). Subsurface flow wetlands (SSF) using gravel substrates
have also been used extensively in the United States (Reed 1992). The goal of
such systems is to allow flow of polluted water through a gravel and root matrix
where over time contaminants are degraded by physical, chemical, and
biological processes.
Free water surface constructed and natural wetlands providing treatment
beyond the secondary level were built throughout the U.S. and Canada during
the 1980s and 1990s. In addition to providing advanced treatment, an increasing
number of these systems have been designed and operated to enhance wildlife
habitat and provide public recreation. Free water surface treatment/habitat
wetlands are typically much larger than subsurface flow wetlands, including
several systems greater than 400 hectares (ha) in size. In the United States, the
largest application of FWS treatment wetland technology to date is the over
16,000 ha of FWS wetlands for the treatment of agricultural drainage in south
Florida. Other large applications include the 89 ha wetland of Orange County,
Florida, for agricultural drainage and the 1200 ha Orlando, Florida, wetland used
to polish municipal effluent.
1-9
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Application of the Technology
Free water surface treatment wetlands can be characterized by either their origin
(natural, constructed, hybrid) or by the level of pretreatment wastewater receives
prior to entering the wetland. As can be seen in Table 1-2, about 28 percent of
the North American Treatment Wetland Database (NADB) treatment systems
utilize natural wetlands, 69 percent of the wetlands are constructed, and 3
percent are hybrid systems. About 65 percent of the natural wetland systems are
receiving conventional secondary treated wastewater. More than 45 percent of
the constructed wetland systems are treating pond effluent and 22 percent are
treating conventional secondary effluent. Viewed from the perspective of
pretreatment levels, one-third of the wetland systems receive pond effluent, one-
third receive conventional secondary effluent, and the remaining third are
distributed among primary, advanced secondary, tertiary, and other.
These treatment systems were designed to meet a wide range of discharge
requirements including:
• MPDES secondary standards
• Total nitrogen
* Ammonia nitrogen
• Total phosphorus
• Total maximum daily limits (TMDL) requirements
* Advanced secondary (BOD and TSS = 10 mg/L)
• Water reuse - groundwater discharge
TABLE 1-2
Percentage distribution of NADB FWS treatment systems by wetland type and level of pretreatment.
Level of Pretreatment
Primary
Secondary
Advance Secondary
Tertiarv
Ponds
Other
None
Unknown
Number
6
45
11
4
45
4
7
13
Natural
(%}
33
53
18
50
2
25
43
15
Constructed
(%)
67
44
82
25
96
75
57
69
Hybrid
(%)
0
0
0
25
2
0
0
0
Other
(%)
0
2
0
0
0
0
0
0
Unknown
(%}
0
0
0
0
0
0
0
15
FWS constructed wetlands have been applied to a wide variety of community
sizes, however, nearly half of the existing systems are in communities with less
than 1,000 people. The fraction of systems serving (or, in the case of pilot
systems, located in) communities of different populations is summarized in
Figure 1-3. About 30 percent of the FWS systems have been built in communities
1-10
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETUNDB
with 1,000 to 10,000 people. There are four full-scale wetland treatment systems
serving communities with populations ranging from 100,000 to 1,000,000
(Beaumont, Texas, Orlando, Florida, Hayward, California, and Riverside,
California). Demonstration projects operated by Phoenix, Arizona,
Albuquerque, New Mexico, and the Sacramento, California, Regional
Wastewater Facility are examples of locations for potential future large
community applications.
FIGURE 1-3
Percentage of all communities utilizing FWS constructed wetlands based upon community size
g 60%
I 50%
| 40%
U 30%
f 20%
•g 10%
* 0%
1,000 10,000 100,000 1,000,000
Population Class Upper Limit
(NADB, n = 135).
The largest number of FWS treatment wetlands are located in the states of South
Dakota and Florida (Figure 1-4). These states utilize both constructed and
natural wetland systems. California has the next largest number of projects, the
majority of which are designed to meet effluent polishing and water reuse
objectives.
FIGURE 1-4
Distribution of FWS constructed wetlands utilized for treating wastewater by State - not including
pilot projects or demonstration projects.
1-11
-------
SECTION 1 INTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Summary of Technology Issues
The scope of this technology assessment is to present information that may be
used to determine whether FWS wetlands are appropriate for achieving specific
water quality and treatment goals. The technical tasks of primary importance
for this technology include:
• Estimating accurately the influent flows and pollutant loads to the FWS
treatment wetland
• Estimating wetland performance and the area and volume required to
satisfy limiting water quality treatment goal(s)
• Developing wetland hydrology and hydraulic design and operating
criteria to attain levels of performance comparable to the performance of
the operating systems used to derive empirical rate constants
• Creating and maintaining the physical, chemical, and biological wetland
system components necessary to achieve expected pollutant-processing
rates
The first of these tasks, the need to predict design loading, is a standard
procedure for conventional wastewater treatment technologies and is not
addressed. The remaining three tasks specific to the design and operation of
FWS treatment wetland technology are covered in this report
Numerous ancillary issues are also important in the design and operation of
FWS treatment wetlands, but are not covered in detail in this technology
assessment. These include conventional civil engineering design criteria for
dikes and levees, water inlet and outlet control structures, and soil compaction
and grading; mechanical design details for flow measurement devices; and
architectural/landscape design details for operator and public access.
Construction and operation issues are also important including clearing and
grubbing requirements, plant selection and plant maintenance techniques, water
level control, avoidance of nuisance conditions from mosquitoes or odors,
operator and public safety, and wildlife management. These and other related
issues for FWS treatment wetland technology are treated in greater detail in a
number of sources related to FWS wetland design and operation (Mitsch &
Gosselink 1993, Hammer 1996, Arizona Department of Environmental Quality
[ADEQ] 1995, Kadlec and Knight 1996, Reed et al. 1995, EPA 1999a and 1988b,
and Water Pollution Control Federation [WPCF, now WEF] 1989; Kadlec et al
2000).
Organization of this Report
This technology assessment report is not intended to provide detailed design
guidance, but rather to present a summary of existing knowledge about FWS
treatment wetland processes and performance. The goal of this report is to
summarize nearly 30 years of FWS treatment wetland information. Many of the
1-12
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
volumes documenting the development of FWS treatment wetland technology
are briefly described herein and are cited in the Reference Section.
Section 2 discusses methods used to prepare this technology assessment report.
Data sources are described and information concerning data quality and
validation are presented. Information is also presented regarding a FWS
treatment wetland technology assessment workshop convened in Mesa,
Arizona, from February 2-4,1996, to guide development of this report.
Section 3 summarizes key components of the physical, chemical, and biological
processes that occur in FWS treatment wetlands. These fundamentals are
essential for presenting and interpreting FWS wetland performance data.
Subject areas covered in this section include wetland hydrology, wetland
hydraulics, wetland treatment processes, wetland vegetation and vegetation
patterns, and wetland thermal effects.
Section 4 presents and discusses several fundamental principles to evaluate and
summarize FWS treatment wetland performance. Wetland background
constituent concentrations, normal ranges of stochastic variability, and the
general pattern of pollutant removal efficiencies are identified. Finally, wetland
system performance is compared to regulatory permit limitations.
Section 5 identifies some system planning and design considerations. The
overall goals of a FWS constructed wetland and the role they play within a
watershed in terms of wildlife habitat value and water quality is examined.
Environmental impact and permit issues associated with constructed wetlands
are also summarized in this section. Discussed are important issues concerning
wetland system planning from a community level perspective. The current FWS
constructed wetland design models and methods are introduced along with a
discussion of their assumptions. Finally, Section 5 includes discussion of
construction, operation, and maintenance considerations, as well as monitoring
and management suggestions.
Section 6 provides specific recommendations regarding the use and further
development of a database for FWS constructed wetlands. Potential nuisance
conditions, open water/emergent vegetation areas, major components of
wetland civil design and construction, issues surrounding wildlife enhancement
wetlands, multiple benefits and public access, and general operation and
maintenance considerations are discussed. Last, a list of critical operational
research issues is presented that if answered, would enhance the current
understanding and application of FWS constructed wetlands to treat municipal
and domestic wastewater flows.
1-13
-------
SECTION 2
Methods for Technology Assessment
Technology development is an incremental process, in which initial research,
guided by information from related fields, provides a preliminary assortment of
observations, speculation, and conclusions. Promising observations and
conclusions become the basis for the design and scope of research efforts. As a
technology develops, subsequent applications can typically be categorized as
those that extend the experience with the technology or those that advance the
state of knowledge about the technology. In the advancement of FWS treatment
wetland technology, many efforts have been made towards data compilations
and feasibility assessments rather than explicit experimental studies with clear
questions, replicated design, and adequate controls. The two categories are not
mutually exclusive; both applications contribute to the development and
acceptance of FWS treatment wetland technology, but they differ in their
contribution to the advancement of the technology.
In this technology assessment, those treatment wetland applications that have
been documented most thoroughly are identified and emphasized. Further
preference is given to research applications designed to advance the state of
knowledge about FWS treatment wetland processes and performance.
Data Sources
Information concerning FWS treatment wetlands has been published in
numerous locations including agency-funded reports, wetland system design
feasibility reports, system operational summaries, project case histories, technical
research papers in refereed and non-refereed journals and books, conference
proceedings, annotated bibliographies, design handbooks, electronic databases,
and general wetland reference books. Primary sources are too numerous to
include here, but citations for many of these references can be found in the
documents listed in the Reference Section.
A sequence of treatment wetland conferences has been held in the U.S. and
abroad beginning in the mid-1970s. A list of the major conferences and, when
available, the literature citation for conference proceedings is provided in
Table 2-1,
EPA has published a number of studies and summaries concerning FWS
treatment wetlands. Titles and citations for these documents are summarized in
Table 2-2. At least four states have published research syntheses and guidelines
for consideration of treatment wetlands (Alabama Department of Environmental
Management [ADEM] 1988, Arizona Department of Environmental Quality
[ADEQ] 1995, Florida Administrative Code [FAC] 1989, South Carolina
Department of Health and Environmental Control [SCDFffiC] 1992). The
2-1
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-1
Listing of major treatment wetland conferences.
Date
May 1976
February 1978
November 1 978
July 1979
September 1979
June 1981
June 1982
July 1986
June 1988
August 1988
September 1989
September 1 990
September 1990
June 1991
October 1991
July 1992
September 1992
December 1992
November 1994
April 1994
July 1995
September 1995
May 1996
September 1996
Location
Ann Arbor, Ml
Tallahassee, FL
Lake Buena Vista, FL
Higgins Lake, Ml
Davis, CA
St. Paul, MN
Amherst, MA
Orlando, FL
Chattanooga, TN
Arcata, CA
Tampa, FL
Cambridge, UK
Show Low, AZ
Arlington, VA
Pensacola, FL
Pinetop-Lakeside, AZ
Columbus, OH
Sydney, Australia
Guangzhou, China
Lafayette, IN
Fayetteville, AR
Tampa, FL
Fort Worth, TX
Vienna, Austria
Description
Freshwater Wetland and Sewage Effluent Disposal
(Tiltonetal. 1976)
Environmental Quality Through Wetlands Utilization
(Drew 1978)
Wetland Functions and Values (Greeson et al. 1 978)
Freshwater Wetland and Sanitary Wastewater Disposal
(Sutherland and Kadlec 1979)
Aquaculture Systems for Wastewater Treatment
(Bastian and Reed 1979)
Wetland Values and Management (Richardson 1981)
Ecological Considerations in Wetlands Treatment of
Municipal Wastewaters (Godfrey et al. 1 985)
Aquatic Plants for Water Treatment and Resource
Recovery (Reddy and Smith 1987)
Constructed Wetlands for Wastewater Treatment
(Hammer 1989)
Wetlands for Wastewater Treatment and Resource
Enhancement (Allen and Gearheart 1988)
Wetlands: Concerns and Successes (Fisk 1 989)
Constructed Wetlands in Water Pollution Control
International Association on Water Quality (IAWQ) 2nd
(Cooper and Findlater 1990)
Municipal Wetlands (City of Show Low Public Works
Department)
Created and Natural Wetlands in Controlling Non-Point
Source Pollution (Olson 1992)
Constructed Wetlands for Water Quality Improvement
(Moshiri1993)
Effluent Reuse and Constructed Wetlands (Arizona
Hydrological Society Summer Seminar)
INTECOL Wetlands Conference (Mitsch 1994)
Wetland Systems in Water Pollution Control IAWQ 3rd
(Pilgram 1992)
4th International Conference on Wetland Systems for
Water Pollution Control (IAWQ 1994)
Constructed Wetlands for Animal Waste Management
(DuBowy and Reaves 1 994)
Animal Waste and the Land-Water Interface (Steele
1995).
Versatility of Wetlands in the Agricultural Landscape
(Campbell 1995)
Constructed Wetlands for Animal Waste Management
(DuBowy, in preparation)
5th International Conference on Wetland Systems for
Water Pollution Control (Perfler and Huberl, in
preparation)
2-2
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-2
EPA Publications on Free Water Surface Treatment Wetlands.
Aquaculture Systems for Wastewater Treatment. R.K. Bastian and S.C. Reed, eds.
EPA 430/9-80-006. MCD 67.
University of California, Davis - Wetland Conference Proceedings. EPA, 1979.
The Effects of Wastewater Treatment Facilities on Wetlands in the Midwest. EPA 905/3-83-002.
1983.
Freshwater Wetlands for Wastewater Management. Region IV Environmental Impact Statement.
Phase 1 Report. EPA 904/9-83-107. 1983.
The Ecological Impacts of Wastewater on Wetlands: An Annotated Bibliography.
EPA 905/3-84-002. 1984.
Freshwater Wetlands for Wastewater Management Handbook. EPA 904/9-85-135. 1985.
Report on the Use of Wetlands for Municipal Wastewater Treatment and Disposal.
EPA 430/09-88-005. 1988.
Design Manual. Constructed Wetlands and Aquatic Plant Systems for Municipal Wastewater
Treatment. EPA 625/1-88/022.1988.
Constructed Wetlands for Wastewater Treatment and Wildlife Habitat. 17 Case Studies.
EPA 832-R-93-005. 1993.
Constructed Wetlands Treatment of Municipal Wastewater. Process Design Manual.
EPA625-R-99-010. Cincinnati, Ohio: Technology Transfer Branch. 1999.
Treatment Wetland Habitat and Wildlife Use Assessment Executive Summary. EPA 832-S-99-001.
1999.
Draft Guiding Principles for Constructed Treatment Wetlands: Providing for Water Quality and
Wildlife Habitat. Prepared by the Interagency Workgroup on Constructed Wetlands. Available on-
line at
reference section of this document also contains many detailed studies on FWS
treatment wetlands.
Books dealing specifically with treatment wetlands are listed in Table 2-3.
Journals that commonly publish articles about treatment wetlands are listed in
Table 2-4.
Free water surface treatment wetland data summaries exist in a number of
locations and include various synthesis papers (North American Wetlands
Conservation Council [NAWCC] 1995, Watson et al. 1989, Water Pollution
Control Federation [WPCF] 1989). The most widely used source of treatment
wetland design and operational performance data is the North American
Treatment Wetland Database (NADB) (Knight et al. 1993, Knight et al,
September 1993, NADB 1993). This electronic database includes information
from 203 treatment wetland systems at 176 sites in North America. Of these
systems, 140 are FWS treatment wetlands of which 125 treat municipal
wastewater, 9 treat industrial wastewater, and 6 treat stormwater.
2-3
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-3
Books with focus on Free Water Surface Treatment Wetlands - in chronological order.
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and Agricultural, edited by
Hammer, D. A., Lewis Publishers, Michigan, 1989.
Natural Systems for Wastewater Treatment - Manual of Practice, Water Environment Federation
(formally Water Pollution Control Federation), 1989.
Constructed Wetlands for Water Quality Improvement, edited by Moshiri, G. A., Lewis Publishers,
Boca Raton, 1993.
Wetlands, Second Edition, by Mitsch, W, J, and J. G. Gosselink, Van Nostrand Reinhold, New
York, 1993.
Natural Systems for Waste Management and Treatment, Second Edition, by Reed, S. C.,
R. W. Crites, and E. J. Middlebrooks, McGraw-Hill Inc., New York, 1995.
Treatment Wetlands, by Kadlec, R. H., and R. L. Knight, Lewis Publishers, Boca Raton, 1996.
Creating Freshwater Wetlands, Second Edition, by Hammer, D. A., Lewis Publishers, Boca Raton,
1996.
Small and Decentralized Wastewater Management Systems, by R.W. Crites, and George
Tchobanoglous, McGraw-Hill Inc., New York, 1998
Constructed Wetlands for Pollution Control: Process, Performance, Design, and Operation, by
International Water Association on Water (IWA) Specialist Group on Use of Macrophytes in
Pollution Control. Scientific Technical Report No. 8, IWA Publishing, 2000,156 pg.
TABLE 2-4
Journals that regularly publish articles dealing with treatment wetlands.
Aquatic Botany American Water Resources Association
(AWRA) Journal
Canadian Journal Fisheries and Aquatic Science Ecological Applications
Ecological Engineering Ecological Modeling
Hydrobiologia Journal of Environmental Quality
Soil Science Water Environment Research (formerly
Environmental Science & Technology Journal of the Water Pollution Control
Federation)
Water Environment Technology Water Research
Water Resources Journal Wetlands
Wetlands Journal Water Science & Technology (IAWQ)
To fully evaluate the performance of full-scale FWS wetlands treating municipal
Wastewater, the following data and operational information (Table 2-5) should
be available.
2-4
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-5
Desired Minimum Information/Criteria for FWS Wetland Systems.
Informational/Data Category or Criteria
1. Municipal wastewater treatment objective with NPDES or equivalent discharge permit for
target contaminants,
2. Wetland type — constructed, natural, or hybrid,
3. Systems have been in operation longer than 3 years and at least 2 years of operating
data are available for the wetland,
4. Spatial dimensions of the system are well characterized,
5. Influent and effluent flow rates are available for independent wetland cells for a minimum
time period of monthly averages,
6. Influent and effluent constituent concentrations are available for independent wetland
cells,
7. Wetlands continuously discharge,
8. Minimize use of data from leaky or Infiltrating (extraneous flows in or out) wetlands,
9. Minimize use of multiple cell wetlands without intermediate flow rate and constituent
concentration data,
10. Particulate and soluble fractionated constituent data, and
11. Surface mapping (vegetated vs. open area) characterization is available on a regular
basis.
No full-scale FWS treatment wetland system has been identified for which all of
the data and operational information listed above is available. Forty FWS
treatment wetland systems were judged to meet enough of conditions 1 through
6 listed above to allow adequate evaluation of the system performance. These
40 systems also include FWS treatment wetlands operating across the range of
feasible pollutant loading rates. The FWS systems meeting the minimum
requirements for system evaluation are listed in Table 2-6 and are the principle
sources of data used for this technology assessment. For the purposes of this
document, these sites will be referred to as the Technology Assessment Sites, and
the data associated with these sites will be referred to as the Technology
Assessment Database (TADB).
While most of these sites were represented in the NADB, several additional sites
were added, and addih'onal data from NADB sites were incorporated where
available. Source information is given whenever necessary for data or
information used in this report.
2-5
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-6
FWS Wetlands used for performance evaluation (Technology Assessment Sites; Source: TADB).
System
Arcata Pilot 1 Cell 8
Arcata Pilot II
Arcata Treatment
Arcata Enhancement Allen
Arcata Enhancement
Beaumont
Benton Cattail
Benton Woolgrass
Brookhaven Meadow Marsh
Cannon Beach
Central Slough
Clermont Plot H
Columbia
Fort Deposit
Gustine (89-90)1 A
Gustine (89-90)16
Gustine (89-90) 1C
Gustine (89-90) 1 D
Gustine (89-90) 2A
Gustine (89-90) 2B
Gustine (89-90) 6D
Gustine (94-97)
Houghton Lake
Iron Bridge
Lakeland
Listowel 4
Manila
Minot
Ml. Angel
Orange County
Ouray
Pembroke FWS 2
Poinclana Boot
Reedy Creek WTS1
Reedy Creek OFWTS
Sacramento
Sea Pines Boggy Gut
Tres Rios Hayfield
Vereen Bear Bay
West Jackson County
State
CA
CA
CA
CA
CA
TX
KY
KY
NY
OR
SC
FL
MO
AL
CA
CA
CA
CA
CA
CA
CA
CA
Ml
FL
FL
ONT
CA
ND
OR
FL
CO
KY
FL
FL
FL
CA
SC
AZ
SC
MS
Pretreatment
Pond
Pond
Pond
Pond
Pond
Pond
Secondary
Secondary
Primary
Adv Sec
Pond
Secondary
Adv Primary
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Tertiary
Secondary
Pond
Pond
Adv Sec
Pond
Tertiary
Pond
Secondary
Adv Sec
Tertiary
Tertiary
Secondary
Adv Sec
Adv Sec
Pond
Pond
Seasonal Origin
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
x Natural
Natural
Natural
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
x Natural
Natural
Constructed
Constructed
Constructed
x Constructed
x Constructed
Hybrid
Constructed
Constructed
Natural
Natural
Natural
Constructed
Natural
Constructed
Natural
Constructed
Area
(ha)
0.04
0,37
1.87
4.40
11.20
222,00
1.50
1.50
0.32
7.00
31.60
0.20
38.30
6.00
0.39
0.39
0.39
0.39
0.39
0.39
0.39
9.38
75.00
494.00
498.00
0.13
0.55
50.18
3.57
89.00
0.89
0.93
46.60
35.00
5.90
6.07
20.00
2.61
69.00
22.70
Flow
(m3/day)
46
327
6700
5186
5186
79494
815
819
48
1814
1788
25
54287
584
164
82
41
164
174
164
144
2563
4378
45521
26550
27
244
16886
2320
6682
718
287
746
12677
3719
3975
6017
3477
879
6257
2-6
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
Technology Workshop and Peer Review
A preliminary draft technical assessment document was prepared by Sherwood
C. Reed in cooperation with Parsons Engineering Science, Inc., under contract
with EPA. An invited workshop was convened in Mesa, Arizona, from February
2 to 4,1996, to provide additional input to the technology assessment process.
This workshop consisted of presentations and discussions of 17 FWS treatment
wetland technology issues by a group of panelists who are published
practitioners in this field of expertise (Table 2-7). Not all of the FWS treatment
wetland professionals with valuable information could be invited to participate
on this panel. However, the panel consisted of a cross-section of the types of
specialists who are active in the design and operation of this technology.
These specialists brought a broad mix of experience related to different
wastewater types, wetland configurations, wetland design, wetland data
analysis and research, and science or engineering educational backgrounds. A
revised document was prepared by Robert L. Knight and Robert H. Kadlec in
cooperation with CH2M HILL, under contract with the City of Phoenix to
complete tasks supported by a grant from EPA. Another revision was prepared
by Robert A. Gearheart and George Tchobanoglous with extensive input from
numerous reviewers. This revision reflects the data presented and discussed
and insights offered by panelists at the workshop.
TABLE 2-7
Panelists for the Mesa, Arizona, workshop held February 2 through 4,1996.
Andrews, Tom L. Southwest Wetlands Group
Crites, Ron Brown and Caldwell (formerly with Nolle and Associates)
DeBusk, Thomas A. Azurea, Inc.
Dortch, Mark U.S. Army Corps of Engineers
Gearheart, Robert A. Humboldt State University
Hammer, Donald A. Hammer Resources, inc.
Kadlec, Robert H. Wetlands Management Services
Knight, Robert L Private Consultant (formerly with CH2M HILL)
Mitsch, William J. Ohio State University
Moore, James Oregon State University
Payne, Victor W.E. Payne Engineering
Reed, Sherwood C. Environmental Engineering Consultants
Reddy, Ramesh University of Florida
Schueler, Thomas R. Center for Watershed Protection
Schwartz, Larry Camp Dresser & McKee
Stiles, Eric Bureau of Reclamation
Tchobanoglous, George University of California, Davis
2-7
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
Due to the large number of contributors and the monumental efforts made by so many
wetland scientists and practitioners in the development of this document, participation
in the drafting and review process for this technology assessment cannot, and does not
constitute approval or agreement by any participant with the content of this final report.
Data Quality and Validation
Data related to wetland design, operation, and performance exist, but are
variable with respect to quality. Some design information is estimated from
plans and specifications and has not been confirmed by as-built field
measurements. Thus, wetland area estimates may be inaccurate because of
difficult construction conditions, berm erosion during and following
construction, or imprecise aerial photo interpretation. Similarly, water depths
are rarely measured at more than a few points, and topography due to final
grade variation or due to natural wetland conditions results in depth estimates
that may be questionable.
Water flow rates can be measured with considerable accuracy, given
state-of-the-art equipment and adequate calibration techniques. However, few
facilities have a high level of instrument sophistication, and many only routinely
estimate inflows or outflows, and not both. Internal flows are rarely measured.
Considerable error has been observed in flow measurements from many
treatment wetland facilities.
Numerous methods are available for analysis of water quality constituents.
These methods tend to range from those requiring minimal sophistication to
those methods employed in scientific research facilities. Significant variability
exists in the accuracy of water quality data from different FWS treatment
wetlands and analyzed at different facilities.
The inherent stochastic behavior of natural systems and the inescapable error
introduced during the collection and analysis of wetland characterization and
performance information is reflected in the NADB by data exhibiting a wide
range of quality, accuracy, and precision. Some of the included data sets have
large quantities of data collected from well-funded projects (i.e. large scale pilot
projects), and the reported data is good quality and accurate. However, many
data sets in the NADB have questionable flow rates and constituent
concentration values, as is cautioned by the USEPA in the user instructions for
the NADB. For this reason, greater or lesser reliance is warranted for conclusions
formulated from different sites. When multiple data sets are included in an
analysis/ some of the uncertainty reflected in the results is likely due to
measurement imprecision, while the rest is due to variables not included in the
analysis.
The NADB was developed to identify sites and was not an attempt to analyze
data or assess data quality. The NADB provides cautionary information on data
quality with no attempt to review and reject questionable data. Although
recognized errors have been corrected for this technology assessment, it is
inevitable, due to the large amount of data presented that some of the results in
2-8
-------
SECTION 2 FOR TECHNOLOGY ASSESSMENT
this assessment will contain inaccuracies. The NADB operational data summary
also disproportionately represents the southeastern US region and FWS
treatment wetland systems with secondary or better quality influents. This
situation needs to be considered carefully when attempting to draw conclusions
regarding regional differences.
The majority of the systems in the database are lightly loaded systems with
relatively low influent BOD and TSS concentrations. In several of these systems,
the effluent BOD is greater than the influent BOD. Figure 2-1 shows that as of
1993,50 percent of the systems (and over 70 percent of the observations)
documented in the NADB had average organic loads of less than 5 kg
BOD/ha-d. Approximately 28 percent of the systems measured had organic
loads less than 1 kg BOD/ha-d. Only 21 percent of the systems documented
received loading within the suggested range for secondary effluents from 12 to
50 kg/ha-d (calculated from hydraulic loading rate ranges suggested by Watson
et al. in Hammer, 1989 for secondary and polishing treatment).
FIGURE 2-1
Influent BOD loading rates for FWS Wetland Systems in the NADB.
Frequency %
—»— Cumulative %
2.5 5 10 15 20 25 50 100
BOD Loading (kg/ha-day)
150 200
Most of the lightly loaded systems have effluent concentrations of BOD close to
the influent concentration, and in some cases, the effluent BOD levels are higher
than the influent. More than 44 percent of the influent BOD measurements for
FWS wetlands in the NADB were less than 10 milligrams per liter (mg/L) (32
percent less than 5 mg/L). Nearly 60 percent of these systems had effluent BOD
values less than 5 mg/L. Some of these systems with low effluent BOD were
moderately loaded systems, but most were very lightly loaded.
Because of legitimate concerns about data variability, quality, or relevance, it is
important to examine information from multiple systems; to look for consistent
trends among systems and over time; and to question and understand
conflicting results. It is also prudent to look to multiple, independent data sets to
validate apparent trends and conclusions. The level of confidence in the
2-9
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
conclusions stated in this report is proportional to the availability of
corroborating evidence and is indicated, when appropriate, throughout the text.
2-10
-------
SECTION 3
Wetland Processes
Free water surface (FWS) treatment wetlands are typically shallow vegetated
basins. They are designed and constructed to exploit physical, chemical and
biological processes naturally occurring in wetlands that provide for the
reduction of organic material, total suspended solids, nutrients, and pathogenic
organisms. FWS treatment wetlands take advantage of these natural treatment
processes by providing time for settling and for the wastewater to contact the
many different reactive surfaces found in wetlands. Wastewater normally has
higher nutrient concentrations than natural wetland influents, thus, many of the
wetland processes and constituent reductions proceed at increased rates. The
increased nutrient loads delivered to treatment wetlands generally result in
higher levels of biological production than that which occurs in natural wetlands
receiving non-wastewater inputs.
Important wetland processes, as they relate to FWS constructed wetlands, are
summarized in this section. Topics discussed include wetland hydrology,
hydraulics, biogeoehemistry, temperature effects, constituent characteristics, and
aquatic vegetation. The intent of this section is to provide the reader with a brief
introduction to wetland processes. For more detailed discussion of such pro-
cesses, the reader may refer to the books available on wetlands (see Table 2-3).
Wetland Hydrology
The hydrology of FWS wetlands, both natural and constructed, is often
considered the most important factor in maintaining wetland structure and
function, determining species composition, and developing a successful
wetlands project (Mitsch and Gosselink, 1993; Hammer, 1992; Kusler and
Brooks, 1988). Wetland hydrology directly influences and controls abiotic
factors, such as water and nutrient availability, aerobic and anaerobic conditions
in both the soil and water columns, water chemistry, soil salinity, soil conditions
(e.g. peat building), and water depth and velocity. In rum, biotic components of
a wetland (primarily vegetation) directly influence wetland hydrology through
processes, such as transpiration, interception of precipitation, peat building,
shading, wind blocks, and development of microclimates within the wetland.
The development of a water balance or budget, the standard method for
characterizing wetland hydrology, is described below.
Water Balance
The wetland water balance is used to quantify inflows and outflows of water to
and from the wetland, and the wetland volume or storage capabilities. The
flows and storage volume control the length of time water spends in the
wetland, and thus the opportunity for interactions between waterborne
3-1
-------
SECTIONS WETLAND PROCESSES
substances and the wetland ecosystem. A thorough understanding of the
dynamic nature of the wetland water balance, and how this balance affects
pollutants, is necessary for the planning and design of FWS constructed
wetlands.
Water enters natural wetlands via stream inflow, runoff, groundwater discharge
and precipitation, and water is lost through stream outflow, groundwater
recharge (infiltration), and evapotranspiration (Figure 3-1). These flows are
extremely variable and stochastic in nature, which can cause large water level
fluctuations to occur in natural wetlands.
FIGURE 3-1
Components of overall wetland water mass balance (Kadlec 1993).
Evapotranspiration = ET
Volumetric
Inflow = Q j
Precipitation = P
Berm Runoff= Qc
Berm Runoff = Qc
Surface Area = A
Depth = h
Bank loss =
Groundwater Exfiltration = I
Volumetric
Outflow = Q0
In contrast, FWS constructed wetlands are typically isolated from stream
inflows. Instead, their primary source of water is continuous wastewater inflow,
precipitation and runoff, while water losses are via surface discharge through
the outlet, evapotranspiration, and possibly percolation (if the wetland bottom
and sides are unlined and/or permeable). The dominant steady wastewater
inflow associated with FWS constructed treatment wetlands represents an
important feature that distinguishes them from many natural wetlands. A
dominant steady inflow, with little variation in water levels drives the ecosystem
toward an ecological condition that is somewhat different from a stochastically
3-2
-------
SECTIONS WETLAND PROCESSES
driven system. Dry-out does not normally occur in FWS constructed wetlands,
and only plants that can withstand continuous flooding will survive.
Although FWS constructed wetlands experience more constant inflows,
seasonally variable wastewater flows can combine with seasonally variable
precipitation and evapotranspiration to cause large differences in seasonal
hydrologic functions. An overall water balance is required to perform the
contaminant mass balance analyses necessary to predict or evaluate wetland
functioning.
The averaging time period over which the water balance components are
determined must be short enough (weekly to monthly) to capture seasonal
effects. In addition, the averaging time period must be compatible with the
frequency of water quality sampling. For instance, weekly water quality results
would normally be combined with weekly average flows to determine mass
removal rates. At a minimum, a detailed monthly or seasonal water balance, in
which all potential water losses and gains are considered, should be conducted
for any proposed FWS treatment wetland. An annual water budget will miss
important seasonal wetland water gains or losses, such as heavy periods of
winter precipitation or high summer evapotranspiration rates.
The overall dynamic water budget for a FWS constructed wetland can be stated
= Qi-Q0+Qc-Qh+Qsm+(P-ET-l)A (3-1)
as:
dV
dt
where:
dV/dt = rate of change in water volume (V) in the wetland with
time (t), [L3/t]
Qj = input wastewater flow rate, [L3/t]
Q0 = output wastewater flow rate, [L3/t]
Qc = catchment runoff rate, [L3/t]
Qjj = bank loss rate, [L3/t]
Qsm = snowmelt rate, [L3/t]
P = precipitation rate, [L/t]
ET = evapotranspiration rate, [L/t]
I = infiltration (or exfiltration) to groundwater, [L/t]
A = wetland top surface area, [L2]
3-3
-------
SECTIONS WETLAND PROCESSES
Each term in this water budget may be important for a given constructed
wetland, but rarely do all terms contribute significantly. The importance of the
primary components of Equation 3-1 will need to be determined prior to the
preparation of the wetland water budget. Some of the terms may be deemed
insignificant and can be neglected from the water budget equation (e.g., Qb, Q^
Qsm, are generally ignored). In addition, groundwater infiltration (I) can be
neglected if the wetland is lined with some type of impermeable barrier.
Input Wastewater Flowrate
The daily influent wastewater flow (Qi) will typically be the controlling inflow
into a FWS treatment wetland. If wastewater flowrates are not known, they can
be estimated using conventional engineering methods for predicting wastewater
flows, such as water usage records, or user numbers and typical wastewater per
capita flowrates found in the literature. The variability of wastewater flowrates
may need to be considered when conducting wetland water balances, especially
for small to medium size FWS treatment wetlands. Variable wastewater flows
include seasonal peaks from vacation communities and high infiltration and
inflow rates into coEection systems, the latter being a condition that should be
studied and minimized prior to treatment system design.
Precipitation
Depending on the time scale of the water budget, precipitation (P) data may be
required in daily, weekly, monthly, seasonal, or annual quantities. Precipitation
inflows into a wetland come from direct precipitation onto the wetland surface
area, and runoff from the wetland catchment (i.e., berms and roads). The effects
of precipitation on the wetland water balance can be significant, especially in
areas of high rainfall or snowfall rates. High seasonal precipitation can dilute
wetland pollutant concentrations, and the resulting effects may need to be
considered in a wetland pollutant mass balance.
Evapotranspiration
Evapotranspiration in a FWS constructed wetland is the combined water loss
due to evaporation from the water surface and transpiration from wetland
vegetation. Many FWS wetlands operate with small hydraulic loading rates. For
the 100 surface flow wetlands in North America, a hydraulic loading of 10.0
millimeters per day (mm/d) is found to be the 40th percentile (Knight et al.
September 1993). Evapotranspiration (ET) losses approach a daily average of 5.0
mm/d in summer in the southern U.S.; consequently, more than half the water
added daily may be lost to ET under these circumstances. Because ET follows a
diurnal cycle, with a maximum during early afternoon and a minimum in the
late nighttime hours, outflow from a FWS constructed wetland can cease during
the day in areas of high ET rates. As shown in Figure 3-2, with the exception of
the non-coastal, western U.S., annual water loss due to ET is largely replaced by
precipitation.
-------
SECTION 3 WETLAND PROCESSES
FIGURE 3-2
Total annual losses (+) and gains (-) from evapotranspiration and precipitation in cm (ET-P) (Flach,
1973).
Output Wastewater Flow
The output wastewater flow (Qo) corresponds to the amount of treated
wastewater (effluent) leaving the FWS constructed wetland. The outlet in a FWS
constructed wetland generally consists of some type of control structure that can
be used to regulate water depth. Increasing or decreasing the water level also
changes the wetland volume, which can influence the wetland water budget by
providing more or less water storage potential.
Exffiltration to Groundwater (Infiltration)
In a FWS constructed wetland, infiltration (I) is the loss of water that occurs into
the bottom soils or berms. The effect is to reduce the amount of water remaining
in a wetland and change the potential for each constituent transformation.
Effluent constituent load(s) calculated at the surface discharge point from the
wetland can be further reduced by the loss of certain soluble constituents as the
water percolates from the system and infiltrates into the soil. If the FWS
constructed wetland is lined with some type of impermeable barrier, percolation
can be neglected in the water balance.
Meteorological Effects on Wetland Water Budget
FWS constructed wetlands generally have a more consistent hydrology than
natural wetlands. However the variability of wastewater inflows, and the
seasonal and stochastic nature of precipitation and ET can produce a variable
seasonal hydrology in these wetlands.
3-5
-------
SECTIONS WETLAND PROCESSES
The effects of precipitation and ET on monthly outflows from the Arcata,
California, FWS constructed wetland system and the Phoenix, Arizona, Tres Rios
Hayfield FWS treatment wetland HI are shown in Figure 3-3. In reference to
Arcata, California, the solid line is the wastewater outflow neglecting the effects
FIGURE 3-3
Monthly water budget for Arcata's wastewater treatment plant (Arcata, California) showing the effects
of precipitation and evapotranspiration on the water budget in a Coastal wetland system and the
monthly water budget for the Tres Rios Hayfield Site basin H1 (Phoenix, Ariiona) showing the
influence increased ET and reduced precipitation has in arid regions.
Arcata, California, Wastewater Treatment Plant Water Budget
300,000
450,000
400,000
Ǥ 350,000
o
1
o
£ 150,000 ir
en *"
Sj 100,000
50,000; •
Qout = Qin
-Qout = Qin •«• (P-ET)A
300,000 • f
250,000 • •
200,000
Date
Tres Rios HayfieldSHe Wetland Basin HI
Water Budget
O
&
o
5
"S
i-.
Ol
€1
&
60000 -
4n (vin .
95 AAA
n .
- p n Q n _ — Q _ _ _ _
y- U -w— w — »— w — 1( — ^^^Q^'^V-.xJB
f,' "A ^-A-* &-A--&-
?
^ff £ ' • O • Qo = (^
S"*&'**^"&. . ,^'"& •"A>"Qo=C? + (P-Er)A
Dale
3-6
-------
SECTIONS WETUND PROCESSES
of precipitation and ET (Qo = Q.), Note the increase and variable nature of
wastewater inflows for the months of December through April, which is caused
by seasonal infiltration and inflow into the collection system. The dashed line
indicates the wastewater outflow when monthly precipitation and ET are
included in the water budget (Q0 = Q{ + (P-ET)A). Precipitation increases
wastewater outflows during the wet weather season (November through April)
whereas ET reduces wastewater outflows during the warmer months of the year
(May through October). In contrast to coastal and the more temperate regions of
the country, Phoenix, Arizona, has an annual precipitation of about < 25 cm/yr
while ET can be as high as 1.2 cm/day in midsummer (AZMET, 1998). Water
budgets both neglecting and considering these meteorological effects for a
wetland in this area are shown below. The lack of precipitation allows the
effects of ET to be seen particularly in the May through July time periods.
Consideration of meteorological effects on the water budget is warranted in this
case, as the reduction in surface outflow from these systems as a result of ET can
result in the degradation of quality due to the evaporative concentration of salts.
Wetland Hydraulics
Wetland hydraulics is the term applied to the movement of water through the
wetland. Improper hydraulic design can cause problems with water
conveyance, water quality, odors, and vectors. For example, in a few instances
FWS constructed wetland design has failed to account properly for head loss,
with inlet over-flooding as the result. Important wetland hydraulic definitions
and basic wetland hydraulic principles are presented and discussed below.
Wetland Hydraulic Definitions
Water Depth
Compared to other aquatic treatment systems (lagoons for example), a wetland
can be designed to incorporate features that allow the system to be operated over
a wide range of depths from less than 10.0 cm to 1.5 m (4 in. to 5 ft). Depending
on bottom topography and slope of the water surface, the water depth will not
be equal at all locations in a constructed wetland. For natural wetlands and
some large FWS constructed wetlands, accurate determinations of water depth
may be difficult due to lack of survey data. However, many FWS constructed
wetlands are designed and constructed with strict engineering grade control and
detailed surveys. Consequently, the elevations of the bottom, berms, islands,
and inlet and outlet structures are known with some degree of accuracy. With
detailed elevations, accurate estimates of average water depth can be obtained.
Water depth in FWS constructed wetlands should be considered an operational
characteristic as well as a design characteristic. The effective depth of a wetland
will change with time as litter fall below the water surface and detritus buildup
on the bottom begin to reduce the depth, therefore, reducing the effective
hydraulic volume.
3-7
-------
SECTIONS WETLAND PROCESSES
Surface Area
The term surface area (A) can embody at least two different concepts when
considering FWS constructed wetlands. First, surface area can refer to the
wetted projection of the constructed wetland in plan view. This is relatively easy
to define using construction or "As-Built" drawings. If this information is not
available, aerial photography or a survey of the wetland water surface perimeter
can be conducted and produce accurate estimates. For most situations, the
surface area or the wetland footprint at the water surface is a good estimate of
the wetland bottom area. In the standard use of the term surface area, either a
contaminant mass or a depth is used to define effectiveness and utilization of the
system resulting in terms such as kg/ha/day (Ibs/acre/day) or cm/ha/day
(ft/acre/day).
Second, surface area can be referred to as the effective surface area, or the
amount of area that comes in routine contact with the water. The effective area
is available for the sorption of pollutants, or the attachment of microbial
communities. Although it would be appropriate for use in modeling the
performance of such systems, the effective surface area of a FWS wetland is
difficult to quantify. Not only does it include the hydraulically active portion of
the wetland bottom, but also the submerged surfaces of vegetation, litter, and
detritus.
Volume
The volume (V) of a FWS treatment wetland is the potential quantity of water
(neglecting vegetation, litter and peat) found in the wetland basin. As indicated
under the depth discussion, the volume changes with time for a given outlet
weir setting.
Wetland Porosity or Void Fraction
In a natural or constructed wetland, the vegetation, litter, and detritus occupy a
portion of the water column, thereby reducing the space available for water. The
porosity of the wetland (e), or void fraction, is the ratio of the theoretical or
empty basin volume to the actual volume available for water to occupy in a
wetland. Wetland porosity can be difficult to determine as it varies in the x-y
(horizontal) dimension due to plant species composition and distribution, and in
the vertical direction with lesser values near the bottom in the litter layer. As a
result, wetland porosity values listed in the literature can be highly variable and,
sometimes, not in good agreement. In a recent study, emergent vegetation was
found to occupy between 3 percent and 8 percent of the available volume
depending upon species and stem density (Lagrace et al., 2000). Literature
values as reported in Reed et al. (1995), shows wetland porosity values ranging
from 0.65 to 0.75 for vegetated wetlands, with lower numbers for dense mature
wetlands. Finally, Kadlec and Knight (1996) report that average wetland
porosity values are usually greater than 0.95, and as such, e = 1.0 can be used as a
good approximation.
3-8
-------
SECTIONS WETLAND PROCESSES
The overall effect of porosity is to reduce the wetland volume available for water
flow and storage. In turn, this reduction in volume reduces the amount of time
water remains in the wetland, and the potential for constituent removal to occur.
Lower wetland porosity values correspond to a lower fraction of the wetland
volume available for water, shorter hydraulic detention times, lower removal
efficiencies, and result in larger required wetland areas to achieve desired
treatment goals. To be conservative, a porosity (e) value of 0.7 to 0.9 could be
used in FWS constructed wetland design calculations, with lower e values for
densely vegetated wetlands, and higher e values for wetlands with more open
water areas.
Volume is not the only factor affected by vegetation density and porosity, head-
loss is equally important. The friction coefficient that controls head-loss through
the wetland depends on the vegetation density. Highly vegetated areas will
have a greater head-loss than open areas, and this increase may cause a
significant backwater effect and can lead to the development of preferential flow
paths. If this potential backwater is not accounted for in the FWS wetland
design, inlet flooding may occur as the wetland vegetation matures, density
increases, and the porosity decreases.
Hydraulic Detention Time
The theoretical (or nominal) hydraulic detention time (t) is the ratio between
flowrate and the wetland volume available for water flow, and includes the
volume reducing effects of vegetation (porosity). The theoretical hydraulic
detention time can be calculated as:
(3-2)
where: t = hydraulic detention time, [t]
V = volume of wetland basin, [L3]
e = wetland porosity, and
Q = flowrate, [L3/t]
The flowrate (Q) value used in the hydraulic detention time calculation is
generally one of two values: input wastewater flowrate (Q.) or average flowrate
(Qavg). The use of input wastewater flowrate (Q;) in Equation 3-2, results in the
inlet hydraulic detention time. The inlet hydraulic detention time neglects the
effects of precipitation, evapotranspiration, and infiltration, and assumes Qj = Qo.
The input wastewater flowrate (Qj) should only be used for preliminary
calculations, or when no measurement or estimate (i.e. water balance) of the
output wastewater flowrate (Qo) exists.
A more realistic measure of detention time can be computed using the average
flowrate (Qav) in Equation 3-2 to account for the effects of water gains and losses
3-9
-------
SECTIONS WETLAND PROCESSES
(precipitation, evapotranspiration and infiltration) that occur in a wetland. The
average flowrate can be estimated by:
Q avg =-^i±^ (3-3)
Accuracy of the theoretical hydraulic detention time calculation is dependent on
the measurements of depth, surface area, and the estimate of porosity. As
mentioned earlier, the theoretical detention time may also be a very poor
estimate of the actual hydraulic detention time due to hydraulic short-circuiting
(e.g., preferential flow paths). The modal detention time, which can be
determined by a tracer study, will always be shorter than the theoretical value
(sometimes less than half).
Hydraulic Loading Rate
The hydraulic loading rate (q) is the rainfall equivalent of whatever flowrate is
under consideration; however, it does not imply the physical distribution of
water uniformly over the wetland surface. The hydraulic loading rate is defined
as:
q=^ (3-4)
where: q = inlet hydraulic loading rate, [L/t]
Q = flowrate, [L3/t]
A = wetland surface area, [L2]
When the input wastewater flowrate (Qi) is used in Equation 3-4, the resulting
calculation is for the inlet hydraulic loading rate, which neglects the effects of
other hydrologic inputs and outputs such as precipitation, infiltration, and
evapotranspiration. Like hydraulic detention time, the average flowrate (Qaug)
can also be used in Equation 3-4, resulting in the average hydraulic loading rate,
accounting for water losses and gains in the wetland.
Water Conveyance
Water conveyance in FWS wetlands is complex hydraulically, varying in both
space and time due to changing inflow conditions and the stochastic nature of
hydrologic events. Water moves through FWS wetlands in response to a surface
water gradient from inlet to outlet, impeded by the friction created from
submerged plants, litter, peat, and the bottom and sides of the wetland. Some
type of outflow structure, such as an adjustable weir, typically is used to control
the water depth. The hydraulic profile of the water surface is dictated by these
factors, combined with the bottom slope and length-to-width ratio of the
wetland.
It is important to consider wetland hydraulics when designing a FWS
constructed wetland. The primary concern is to ensure that the wetland can
3-10
-------
SECTIONS WETLAND PROCESSES
handle all potential flows without creating significant backwater problems, such
as flooding the inlet structures or overtopping of berms.
Assessment of the head-loss from inlet to outlet can usually be done using
Manning's equation. When a more detailed head-loss calculation is required, or
the effects of precipitation, evapotranspiration, and infiltration need to be
considered in water conveyance calculations, then the simplified one-
dimensional flow procedure presented by Kadlec and Knight (1996) can be used.
In the case of complex geometry or irregular boundaries, more detailed
hydraulic modeling approaches may be required, such as the one-dimensional
HEC2 or HECRAS (U.S. Army Corps of Engineers Hydraulic Engineering
Center), or the two-dimensional Surface Water Modeling System (Engineering
Computer Graphics Laboratory, Brigham Young University).
Aspect Ratio
The aspect ratio is defined as the quotient of the average length of the major axis
and the average width of a wetland. Because the footprint of a wetland can have
a variety of shapes, it is the effective aspect ratio between inlets and outlets that
is important. In general, FWS treatment wetlands with high length to width
ratios are of greatest concern with respect to head-loss. However, some early
researchers reported that the treatment performance of FWS constructed
wetlands is better at higher aspect ratios (Wile et al. 1985).
For wetland systems with high length to width ratios, careful consideration
needs to be given to increasing costs associated with more lineal feet of berm
construction, head-loss, and the internal flow through the wetland. In some
instances weir overflow rate, location of inlets and outlets, and elevation of
berms may be as important or more important than the influence of the aspect
ratio on wetland performance.
Internal flow Patterns Effects/Physical Facilities
The low gradients found in FWS treatment wetlands result in very low water
velocities, approaching laminar flow in highly vegetated areas. This type of flow
regime produces quiescent conditions, an ideal situation for many of the
physical, chemical, and biological processes that occur in FWS wetlands.
Water does not flow through a FWS wetland in one flow direction or path.
Instead, water flows through a complex maze of submerged vegetation, litter,
detritus, and other obstructions (e.g., islands); forcing the water velocity to
increase and decrease and to continually change direction. Water in open areas
located away from submerged vegetation or accumulated bottom material is less
subject to friction and generally moves at faster velocities than water located in
densely vegetated areas. Open water zones are subject to wind-driven surface
flows, which can move at higher velocities than water below the surface, and
cause mixing to occur at different depths. Some areas of a FWS constructed
wetland, such as corners and behind islands, may become isolated from the
main flow path, creating pockets of dead space for which no or little water
3-11
-------
SECTION 3 WETLAND PROCESSES
exchange occurs. The bottom topography may also form deeper pockets or
pools, creating more dead space zones, resulting in a constantly changing
internal flow pattern intermediate between the ideal extremes of plug flow and
complete mixing.
All of these processes combined cause water to flow through a FWS wetland in a
shorter time period than defined by the theoretical hydraulic detention time. In
many cases, water can flow at high velocities through a small portion of the total
wetland volume, significantly lowering the hydraulic detention time, a process
often referred to as short-circuiting. For example, the theoretical detention time
for the Boggy Gut treatment wetland was estimated to be 19 days; however, the
measured value using tracer studies was approximately 2 days (Knight and
Ferda 1989). Careful consideration of the site characteristics showed that this
difference was due to large zones (volumes) of wetland (dead zones) that were
not incorporated effectively in the treatment of the influent flow.
The placement, size, and orientation of inlet and outlet works can be an
important factor in determining the hydraulic response of a FWS constructed
wetland to wastewater inputs and process withdrawals. Experience to date has
been to distribute the influent over a large portion of the inlet region and to place
relatively narrow (0.6 to 1 m) rectangular adjustable weirs along the discharge
region of the cells. To mitigate some of the short-circuiting inherent in FWS
wetlands, several strategies exist for providing a collection volume at the
terminus region of the wetland (Kadlec and Knight 1996). In one approach, a
deeper zone is created in the outlet area with the outlet weir control structure
placed away from the bank into the collection volume. Other approaches have
been to collect the influent in vegetated shallow water zones outfitted with
barriers (fenced) to minimize fish and amphibian export with the effluent.
Recently, square non-adjustable weir structures have been experimented with to
increase weir overflow rates from 225 to 500 liters per meter of weir length per
minute (L/m-min) over that of more conventional outlet weir designs (Gearheart
1998, Unpublished data).
Water Balance Effects on Wetland Hydraulics anil
Water Quality
The variability inherent in wastewater flowrates and the stochastic nature of
meteorological events controls wetland hydraulics, which in turn affects
treatment wetland performance and water quality. Impacts to wetland
hydraulics can best be described by noting the increases and decreases to the
wefland hydraulic detention time caused by water gains and losses in the
wetlands water balance. Likewise, the wetland hydraulic detention time can be
used to explain water balance impacts to wetland water quality.
Precipitation to a wetland increases inflow, which affects wetland hydraulics by
decreasing the hydraulic detention time, and affects water quality by diluting
constituent concentrations. The combination of these two influences can provide
either poorer or better performance of the wetland with regard to water quality.
In systems receiving low influent constituent concentrations, concentration
3-12
-------
SECTIONS WETLAND PROCESSES
reduction is likely to be less evident with precipitation additions; in heavily
loaded systems, concentration reductions will often be more notable. In both
cases, mass load reduction could be poorer with precipitation additions because
the added flow reduces the effective hydraulic detention time.
Evapotranspiration has the effect of increasing hydraulic detention time and
increasing constituent concentrations. The combination of precipitation and
evapotranspiration can improve concentration reduction in very lightly loaded
systems, but generally decreases concentration reduction in heavily loaded
systems. The effect of exfiltration is similar to evapotranspiration by increasing
the hydraulic detention time and increasing the potential for constituent
removal. Constituent load reduction can further be enhanced by the loss of
constituents with the water as it infiltrates into the soil.
Wetland Biogeochemistry
Free water surface treatment wetlands support a variety of sequential and often
complementary treatment processes. The predominant physical, chemical, and
biological mechanisms operative in FWS treatment wetlands are summarized in
Table 3-1. These interrelated biological, chemical, and physical treatment
processes control the transport and transformation of constituents through FWS
wetlands. Specific processes controlling total suspended solids, biological
oxygen demand, chemical oxygen demand, dissolved oxygen, nitrogen,
dissolved organic phosphorus, pH, organic pollutants, and metals within FWS
wetlands are more explicitly described in subsections below.
A hypothetical partitioning of treatment processes throughout the wetland
volume is shown in Figure 3-4. Wetland treatment processes are generally
associated with vertically and horizontally differentiated zones within the
wetland volume. These zones are linked both hydro-dynarnically and through
sequential physical, chemical and biological reactions. In the inlet zone, the
physical process of sedimentation dominates treatment and quickly removes the
easily settleable solids and their associated constituent. Finer particulates are
removed slowly by flocculent settling farther into the wetland.
The location of various aerobic biological processes operating within the wetland
is partially determined by the dissolved oxygen concentration. The oxygen
demand from degradable carbon compounds is met near the surface of the open
water zones of the wetland where oxygen transfer from the atmosphere and
release to the water column by photosynthesis is greatest. Oxidation of
ammonia nitrogen (nitrification) in a wetland occurs where carbonaceous BOD
has been generally satisfied and sufficient dissolved oxygen is present in the
water column.
Denitrification, or reduction of the nitrate nitrogen species, has been shown to be
a significant process in FWS constructed wetlands. The combination of anoxic
conditions, physical substrates for microbial attachment, and internal carbon
sources provide ideal conditions for nitrate conversion to dinitrogen gas. The
dissolved organic carbon produced as a by-product of detrital decomposition
supplies the carbon for this microbial process. Because most denitrifying
3-13
-------
SECTIONS WETLAND PROCESSES
TABLE 3-1
Mechanisms and factors that affect the potential for removal or addition of water quality constituents
in FWS wetlands (Adapted from Stoweil et al. 1980).
Mechanism Water Quality Constituent* Description
BOD TSS N P DO Bacteria Heavy
Virus Metals
Physical
Absorption
S P/S Gas transfer to and from water
surface
Adsorption/
desorption
Emulsification
Evaporation
Filtration
Impaction
Flocculation
Photochemical
reactions
S
S
Sedimentation
Thermal
Volatilization
P
I
I I
P S
P
I Interparticle attractive force (van de
Waals force); hydrophilic interaction
S Suspension of low solubility
chemicals
S Volatilization and aerosol formation;
thermal moderation
I Particulates filtered mechanically as
water passes through substrate and
plants
S Interparticle attractive force (van de
Waals force); hydrophilic interaction
Solar radiation is known to trigger a
number of chemical reactions.
Radiation in the near-ultraviolet (UV)
and visible range is known to cause
the breakdown of a variety of organic
compounds. Pathogenic bacteria
and virus attenuation.
P Gravitational settling of larger
particles and contaminants
Autoflocculation; natural coagulants
Similar process to gas absorption,
except that the net flux is out of the
water surface.
Chemical
Adsorption
Chelation
Chemical
reactions
P
S
Decomposition
Oxidation/
reduction
reactions
Precipitation
S On substrate and plate surfaces
P Formation of complex metal
compounds through ligands
Hydrolysis, for example, is an
important chemical reaction that
occurs in the environment, by which
proteins are converted into amino
acids and other soluble compounds.
Organic nitrogen can also be
converted to ammonium.
Decomposition or alteration of less
stable compounds by phenomena
such as UV irradiation, oxidation,
hydrolysis
P Anoxic condition; metal speciation;
organic acid production
P Formation of co-precipitates with
insoluble compounds
3-14
-------
SECTIONS WETLAND PROCESSES
Mechanism Water Quality Constituent*
BOD TSS N P DO Bacteria Heavy
Virus Metals
Description
Biological
Algal synthesis
Assimilation,
plant
Bacteria/
Metabolism
Aerobic
Anaerobic
Plant
adsorption
Predatlon
S S
C C S P/S I/C
P/C
S
P/C
C
S
C
S
The synthesis of algal cell tissue
using the nutrients in wastewater.
Uptake and metabolism by plants;
root excretions may be toxic to
enteric organisms; transpiration
concentrates effluent; dissolved
oxygen supply
Removal of colloidal solids and
soluble organics by suspended,
benthic and plant supported
bacteria; bacterial nitrification,
denitrification; microbial mediated
oxidation
Under proper conditions, significant
quantities of contaminants will be
taken up by plants.
Zooplankton and aquatic insect larva
particles; odonata and fish-aquatic
insect
Notes: *P = primary processes, S = secondary processes, I = incidental effect (occurring with
removal of other constituent), C = contributory effect, S/P = depends on influent and design
conditions, N = negative.
bacteria are obligate anaerobes, oxygen must be suppressed to less than
0.5 mg/L in the water column. Both nitrification and denitrification processes
are temperature dependent, and enzymatically shut down at temperatures less
than 5-7 °C.
This brief introduction illustrates how FWS constructed wetlands incorporate a
similar sequence of physical, chemical, and biological treatment processes to
those commonly employed in conventional wastewater treatment. FWS
constructed wetlands can be designed to emphasize some treatment processes
over others by altering the geometry, hydraulics, and plant types, densities, or
locations. A more detailed discussion of the role of unique features of FWS
constructed wetlands and the processes controlling specific constituents of
interest follows.
3-15
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-4
Conceptual partitioning of treatment processes through a FWS wetland (Gearheart, 1998).
Litter Layer *',._.' ^ \
Soluble BOD Removal
DOC Accumulation
7 ^^^7VS» " *\* • * \ i i i i i
3lubilization /NSv \ \J Denitrification
imonif.cation< XvJ" ' 'X " "Hi I
Discrete.'
Settling *
Flocculent Settling
Anaerobic Decomposition!
Nitrification
Hydrogen '•
\
m Methanogenisisl
\
•"""} ''2 '• if :".*•"•"•"" Sedimentation - Detrital Buildup - Peat Development"-" -J;'.." -'.'.'i
Anoxous Oxous
Hydraulic Retention Time
Anoxous
Total Suspended Solids
Processes
Total suspended solids (TSS) are both removed and produced by natural
wetland processes. During treatment, settleable incoming participate matter
usually has ample time to settle and become trapped in litter or dead zones.
Once there, soluble organic constituents are reduced to carbon dioxide and low
molecular weight organic acids and inorganic constituents can become bound as
sulfide complexes or become buried through sediment accretion. Wetland
scientists generally refer to the combination of removal processes as filtration,
although stem and litter densities are not typically high enough to act as a filter
mat. As shown in Figure 3-5, a number of wetland processes produce participate
matter including: death of invertebrates, fragmentation of detritus from plants
and algae, and the formation of chemical precipitates such as iron sulfide.
Bacteria and fungi can colonize these materials and add to their mass.
In wetlands, velocity-induced re-suspension is minimal, but gas lift and
bioperturbation can reintroduce solids into the water column. Wetland
sediments and micro-detritus are typically near neutral buoyancy, flocculent,
and easily disturbed. Bioperturbation by fish, mammals, and birds can re-
suspend these materials and lead to additional TSS in the wetland effluent. The
oxygen generated by algal photosynthesis or methane formed in anaerobic
processes can cause flotation of floe assemblages. Re-suspension due to fluid
shear forces on bed solids is not usually a significant process except in the
vicinity of a point discharge into the treatment wetland.
3-16
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-5
Wetland TSS removal, re-suspension, and internal generation processes.
Filtration
Impaction
Adsorption
Sedimentation Invertebrate
litterfall
AutofiocculaUon
Macrophyte
iitterfall
Periphyton
litterfall
Dissolution
& chemical
precipitation
The magnitude of wetland participate cycling is large, with high internal levels
of gross sedimentation and re-suspension, almost always overshadowing TSS
influent loading. TSS effluent concentrations rarely result from an irreducible
fraction of the influent TSS, and are often dictated by the wetland processes that
generate TSS within the wetland. Most FWS constructed wetland designs are
determined by an effluent limitation other than TSS and are sufficiently large
that effluent TSS approaches background levels. Large expanses of open water
not followed by vegetation can however lead to excessive algal growth and
subsequent high effluent TSS.
High incoming TSS or high nutrient loads that stimulate high TSS production
may eventually lead to a measurable increase in bottom elevation (van Oostrom
and Cooper 1990). In lightly loaded FWS wetlands typical accretion rates ranged
from 2 to 10 mm/yr (Richardson et al, 1994). Solids accretion rates were a
function of distance from the inlet and vegetation density for measurements
obtained in 12 experimental marshes receiving oxidation pond effluent in
Arcata, California, from September 1979 through September 1982. In year 3 of
operation, the solids bank had extended approximately 10 to 15 meters (12.5 - 20
percent) into the cells from the influent point. When measured next to clumps of
vegetation, the depth of settled solids varied from 20 to 36 cm, while in open
areas, this was reduced to 4 to 10 cm after the 3 years of operation (Gearheart et
al., 1983). As yet, no treatment wetland has required maintenance because of
normal solids accumulation, including some that have been in operation for 20
years or more, but this is unlikely to last indefinitely. In situations of high
incoming non-volatile solids, a settling basin can be designed to intercept a large
3-17
-------
SECTIONS WETLAND PROCESSES
portion of the solids, thus providing for easier cleanout and extending the life of
the inlet region of the wetland.
Settleable Solids Reduction-Anaerobic Decomposition
The benthic decomposition of accumulated solids from the influent and from the
plant litter produced in the wetland has a delayed effect on the oxygen budget
and biochemical oxygen demand (BOD) concentrations in a FWS treatment
wetland. The accumulated material compacts and increases in density as
anaerobic processes release aerobically degradable by-products to the sediment
and organic layer pore water. These aerobically degradable by-products
subsequently diffuse into the overlying water column and add to the BOD.
Accumulated organic debris degrades at different rates depending on the source
and composition of the material. As the degradability of the material decreases,
the decomposition rates slow and the nature of the soluble by-products change.
The implication of this degradation rate and its relationship with the BOD in the
water column is significant. The half-life of soluble BOD is approximately 3 days
while the half-life of organic sediment, which is temperature-dependent, is
closer to 4 months. Earlier observations by sanitary engineers of the role that
benthic organic deposits played in the oxygen budget in streams is analogous in
many ways to conditions in a FWS wetland. In streams, the oxygen
requirements of benthic organic deposits are limited by the rate at which
production of diffusible degradable material enters the overlying water column
and not by the rate at which anaerobic breakdown occurs (Phelps 1944). For a
given solids accumulation rate and temperature regime, a steady-state condition
of organic sediment decay and release of soluble BOD to the water column
should develop. This release of soluble BOD from the litter layer is one
component contributing to the background BOD.
Biochemical Oxygen Demand
For FWS treatment wetlands receiving municipal wastewater, some fraction of
the influent carbon compounds are dissolved while the rest enters in the form of
particulate matter. Particulate settling provides one removal mechanism, and
typically occurs in the inlet region of the wetland (Figure 3-6). Microbial
communities process the dissolved carbon compounds. Microbial removal
processes include oxidation in the aerobic regions of the wetland and
methanogenesis in the anaerobic regions. Active microorganisms are usually
associated with solid surfaces, such as litter, sediments, and submerged plant
parts.
In addition to microbial decomposition, dissolved carbon is fixed into new
biomass during photosynthesis. The decomposition of this biomass, litter and
sediments produces a return flux of BOD to the water column. The balance
between removal of influent BOD and the decomposition processes contributing
BOD determines the wetland effluent concentration of this constituent.
3-18
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-6
Simplified portrayal of wetland carbon processing. Incoming BODs is reduced by deposition of
participate forms and by microbial processing in floating, epiphytic, and benthic litter layers.
Decomposition processes create a return flux.
• Settleable solids (particulate
BOD and suspended solids)
Plant litter
Settled suspended/flocculated soids and detritus (peat building) -/
^ Release of soluble BOD from destruction of volatile settleable solids
Tf Release of soluble BOD from destruction and decomposition of solids and detritus
JL Release of soluble BOD from decomposing plant litter
Chemical Oxygen Demand
The chemical oxygen demand (COD) measures the concentration of oxidizable
compounds using a strong chemical oxidant. Thus, the COD test measures the
sum concentration of two distinct fractions of oxidizable compounds: easily
biodegradable compounds and oxidizable but not easily biodegradable
compounds. The concentration of easily biodegradable compounds is often
represented as BOD5, the biological oxygen demand measured after 5 days of
incubation, with the difference between the COD and BOD5 representing the
concentration of compounds that is not easily biodegradable. Some of these
non-BOD compounds are degradable under anoxic conditions via anaerobic
decomposition, or under aerobic conditions in periods of longer than 5 days.
Physical and microbial processes remove COD while other processes produce
COD in FWS constructed wetlands. Effluent COD concentrations from the
Arcata wetland cells only varied from 60 to 66 mg/L while the influent BOD
ranged from 45 to 92 mg/L (Gearheart et al., 1983). Consistent COD effluent
concentrations from the pilot wetland cells, even with a ten-fold range in
hydraulic/organic loading, indicate that the effluent concentrations are more
closely associated with the amount and type of aquatic plants decaying within
the wetland than the influent BOD load. The COD/BOD ratio averaged 3.7 for
the influent (oxidation pond effluent) while the wetland effluent COD/BOD
ratio varied from 3.1 at the beginning of the study to 28 at the end of the study.
3-19
-------
SECTIONS WETLAND PROCESSES
In another study, a pilot cell was loaded at 50 kg/ha-d for 15 months, after which
time the influent was switched to tap water for 9 months. The concentration of
BOD and COD before and after the addition of tap water (no addition of influent
BOD) is shown in Figure 3-7. The COD/BOD ratio was 3.9 during the BOD
loading period.
FIGURES-?
BOD and COD effluent concentration before and during tap water loading to Arcata Pilot Project
wetland.
~. 100 x
D
O
U
P
O
53
U-l
w
0)
U
Tap water used as influent second year
-BOD
-COD
4-
Fall Winter Spring Summer Fall Winter Spring Summer
Quarter (1980-1982)
After the switch to freshwater, the COD/BOD ratio increased to 17 with COD
and BOD concentrations of 30 and 1.7 mg/L, respectively. It appears, based on
these observations, that the detrital material contributed about 1.7 mg/L of BOD
at the wetland cell hydraulic loading rate of 240 mm/day.
Dissolved Oxygen
Dissolved oxygen is depleted to meet wetland oxygen requirements in four
major categories: sediment/litter oxygen demand, respiration requirements,
dissolved carbonaceous BOD, and dissolved nitrogenous oxygen demand NOD.
Sediment oxygen demand is the result of decomposing detritus generated by
carbon fixation in the wetland, and the decomposition of precipitated organic
solids that entered with the wastewater. NOD is exerted primarily by
ammonium nitrogen, but ammonium may also be contributed by the
mineralization of organic nitrogen. Decomposition processes in the wetland also
contribute to NOD and BOD, further increasing the oxygen demand and
reducing the dissolved oxygen in the wetland water.
Plant roots also require oxygen, which is normally transported downward
through passages (aerenchyma) in stems and roots. Some surplus of oxygen
3-20
-------
SECTIONS WETLAND PROCESSES
may be released from small roots into their immediate environs, but it is quickly
consumed by the local oxygen demand (Brix 1994a). Wetland soils are typically
anoxic or anaerobic (Reddy and D'Angelo 1994).
Wetland open-water areas can be aerated via oxygen transfer from the
atmosphere at the air-water interface. Reaeration mechanisms include
dissolution and diffusion (O'Connor and Dobbins 1958), as well as turbulent
transfer associated with rainfall and wind induced surface mixing (South Florida
Water Management District [SFWMD] unpublished data). In un-shaded open
water areas, photosynthesis by algae within the water column produces oxygen,
sometimes creating dissolved oxygen concentrations in excess of the saturation
limit (Schwegler 1978). Photosynthesis stops at night, and respiration, which
consumes oxygen, then dominates. The result is strong diurnal variations in
water column DO for lightly loaded, algae-rich, open water wetlands.
In vegetated regions of the wetland, shading prevents high algal concentrations
and DO levels are typically low near the surface. Anoxic or anaerobic conditions
persist throughout the remainder of the water column. The effect of vegetation
on DO level in the Arcata Enhancement Marsh is shown with the DO in the non-
vegetated zones (Figure 3-8) significantly higher than that in the vegetated zone
(Figure 3-9).
FIGURE 3-8
Vertical distribution of DO in a submergent plant zone (depth = 1.0 m) of the Arcata Enhancement
Marsh.
o
Q
Gearheart-5
Gearheart-3
Gearheart-1
middle
bottom
3-21
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-9
Vertical distribution of DO in an emergent plant zone (depth = 1.0 m) of the Arcata Enhancement
Marsh,
top
Qearheart-6
Gearheart-4
Gearheart-2
middle
bottom
Nitrogen
Nitrogen is a key element in biogeochemical cycles and occurs in a number of
different oxidation states in natural and constructed treatment wetlands.
Numerous biological and physiochemical processes can transform nitrogen
between its various oxidation states (Figure 3-10). The dominant nitrogen
species entering a FWS treatment wetlands depends on the level and type of
wastewater pretreatment, but may include organic, ammonia, nitrate, and nitrite
nitrogen, and nitrogen gases (di-nitrogen gas [N2] and di-nitrogen oxide [N20]).
A fraction of the organic nitrogen is readily mineralized to ammonia nitrogen in
aquatic and wetland environments. Ammonia nitrogen is distributed between
the ionized form (ammonium, NH4+) and a smaller percentage as un-ionized
ammonia (NH3). The distribution of total ammonia between NH4+ and NH3
depends on water temperature and pH. Un-ionized ammonia is volatile and
may be lost directly to the atmosphere.
Ammonium nitrogen can be oxidized in open, aerobic zones to nitrite and nitrate
nitrogen through an aerobic microbial process called nitrification. Free dissolved
oxygen and carbonate alkalinity are consumed in this process. Ammonium
nitrogen may also be biologically assimilated and reduced back to organic
nitrogen in the plants, or may be removed from the water column by adsorption
to solid surfaces, such as wetland sediments. Adsorbed ammonium is readily
released back to the dissolved ammonia state under anaerobic conditions.
Nitrate nitrogen is readily transformed to di-nitrogen gas in treatment wetlands
by the mierobiologically mediated anaerobic process, denitrification.
Denitrification occurs most readily in wetland sediments and in the water
3-22
-------
SECTIONS WETLAND PROCESSES
column below fully vegetated growth where dissolved oxygen concentrations
are low and available organic carbon is high. Organic carbon is consumed in
this microbial process and alkalinity is produced.
FIGURE 3-10
Nitrogen transformation processes in wetlands (Gearheart 1998, unpublished data).
Atmosphere
Blue-green
algae and
Azotobacter
Phytoplankton,
bacteria,
and aquatic
macrophyte
Azotobacter
•y
-------
SECTIONS WETLAND PROCESSES
Phosphorus
New constructed and natural wetlands are capable of adsorpting and absorbing
phosphorus (P) loadings until the capacity of the soils and new plant growth are
saturated. Phosphorus interacts strongly with the wetland soils and biota
resulting in short-term removal and long-term storage (Reddy 1984, Reddy and
D'Angelo 1994). The potential for P removal is most easily illustrated by the
seasonal uptake and release by plants of soluble reactive phosphorus. The
effects of two different phosphorus loadings on the effluent soluble reactive
phosphorus during the growing season (Figure 3-11) were evaluated in Arcata's
Pilot Project I. The difference between the lower phosphorus loading (Cell 3)
and the saturated phosphorus loading (Cell 5) represents the mass of
phosphorus taken up by macrophytes and epiphytes. The majority of the
phosphorus taken up by the wetland plants is released as soluble reactive
phosphorus in the late summer and fall as the vegetation senesce and
decomposes.
FIGURE 3-11
Influent and effluent phosphorus in the Arcata Pilot Project IFWS wetlands, Second Pilot Project,
1982. Cell 5 was loaded at 0.75 kg/ha-d, and Ceil 3 at 0.15 kg/ha d (Gearheart 1993).
10
9-
8-
I 6'
"a 5 -
4-
3-
2-
1 -
a.
in
0
8- 15- 22- 29- 5- 12- 19- 26- 3- 10- 18- 25- 1- 8- 15-
Mar Mar Mar Mar Apr Apr Apr Apr May May May May Jun Jun Jun
Date (1982)
Sustainable P removal processes involve accretion and burial of phosphorus in
wetland sediments. Uptake of P by small organisms, including bacteria, algae,
and duckweed, act as a rapid-action, partly reversible removal mechanism
(Figure 3-12). Cycling through growth, death, and decomposition returns most
of the microbiotic uptake back to the water column, but a significant residual is
lost to long-term accretion in newly formed sediments and soils. Macrophytes,
such as cattails and bulrushes, perform a similar function, but on a longer time
scale of months to years. The detrital residual from the macrophyte cycle also
contributes to the long-term storage in accreted solids. Direct settling and
3-24
-------
SECTIONS WETLAND PROCESSES
trapping of participate P may also contribute to the accretion process (Reckhow
and Qian 1994), There can also be biological enhancement of mineralogical
processes, such as iron and aluminum uptake and subsequent P binding in
detritus and the algae-driven precipitation of P with calcium.
FIGURE 3-12
Conceptual cycling of phosphorus forms in FWS constructed wetlands. SRP: Soluble reactive
phosphorus; POP: particulate organic phosphorus; TSS-POP: form of POP in terms of a fraction of
the total suspended solids.
Emergent plant
Plant litter
™_,™™-. - *™,^ -s'«i^3*f™-.>
^^^S^^U^XSA.
Hydrogen Ion
Natural wetlands exhibit pH values ranging from basic in prairie wetlands (8-9),
to slightly basic in alkaline fens (pH = 7 to 8), and to quite acidic in sphagnum
bogs (pH = 3 to 4) (Mitsch and Gosselink 1993). Natural freshwater marsh pH
values are generally slightly acidic (pH = 6 to 7). Treatment wetland effluent
hydrogen ion concentrations are typically neutral to slightly acidic. Data from
an open water, unvegetated treatment "wetland," displayed high pH during
some summer periods (pH > 9), with circumneutral influent (7.0 < pH < 7.4)
(Bavor et al. 1988). Algal photosynthetic processes peak during the daytime
hours, reducing dissolved CO2 concentrations, creating high pH during the day,
followed by a night-time sag with low pH as respiration replaces
photosynthesis.
Organic substances generated within a wetland via growth, death, and
decomposition cycles are a source of natural acidity. The resulting humic
substances are large complex molecules with multiple carboxylate and phenolate
groups. The protonated forms have a tendency to be less soluble in water and
precipitate under acidic conditions. As a consequence, wetland soil/water
systems are buffered against incoming basic substances. They are less well
3-25
-------
SECTIONS WETLAND PROCESSES
buffered against incoming acidic substances as the water column contains a
limited amount of soluble humics.
The net result of the processes described above is that treatment wetlands can
maintain their effluent pH at approximately pH 7 (Gearheart et alv 1983).
Listowel, Ontario constructed treatment wetland No. 3 received lagoon water,
which periodically exhibited high pH due to algal activity in the lagoon (Figure
3-13). During the first year of operation, little or no buffer capacity was evident.
As the vegetation spread to cover the wetland and litter formation and
decomposition became operative, high incoming pH values were neutralized
effectively by the wetland.
FIGURE 3-13
Hydrogen ion (pH) buffering in system 3 at Listowel (Herskowitz 1986).
8.5
8-
7.5-
Q.
7-
6.5 -
co CM m oo i-
i- i- i- «M !M CM CO
Months of Operation
CO
CO CD
Metals
Metals removed from the water column by settling are bound to particles and
may eventually be buried in the anoxic sediments. As shown in Figure 3-14, in
the sediments below open water zones, many metals of concern are chemically
reduced and bound as metal sulfides, a form that can minimize their biological
mobility. If sediments are disturbed, the potential exists for the chemically
reduced and sequestered metals to be oxidized and dissolve, thus becoming
biologically mobile again. Metal actions in sediments below vegetated zones
behave similarly, except that the aerobic zone is extremely shallow.
Metals are also incorporated into biomass via primary production processes
occurring in a wetland. For macrophytes, metals are taken up via the roots and
distributed throughout the plant. The extent of uptake and distribution within
the plant depends on the metal species and plant type.
3-26
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-14
Metal sulfide burial processes in a wetland (Meyers 1998, personal communication).
Atmosphere
CO2
Air-Water Interface
Organic Particles j
2CM2O -f NOj +
SO4~+2CH20 +
H2S~
. 4
X
CO2 +
> CH2O
2^-^
ff"-^
*.
<* i>- 1^§ -
f
1
i
#20 -*- CW2O + O2
+ 02 -*~ CO2 + H2O
\
CO2 + H20 + Nt/4
HS~ + 2H20 + 2CO2
•*H»-S2"
Water-Sediment Interface
.. 2+
Me ->v^
DiffiJsiotN- Me2++ S2'-**
2CH2O
-*~CH4
MeS "Permanent"
MetaJ-Sulfide Burial
+ CO2
m
c
fl
JO
t
'01
0)
1
.S
3
2
CU
C
Thermal Effects in Wetlands
The temperature of wetland waters influences both the physical and biological
processes within a FWS constructed wetland. Under winter conditions, ice
formation may also alter wetland hydraulics and limit oxygen transfer. Under
severe conditions, freezing may even result in system failure. Decreased
temperatures are known to reduce the rates of biological reactions, the extent of
which, however, varies with the constituent. In FWS constructed wetlands, BOD
removal does not always appear to exhibit temperature dependence.
Temperature dependent BOD removal may be masked by other processes, such
as internal loads due to decomposition that are also temperature dependent, or
the removal may be primarily due to non-biological mechanisms. Nitrogen
removal has consistently been observed to decrease with temperature, indicating
that it is controlled by biological mechanisms.
Predicting and understanding the influence of water temperature within a FWS
wetland is an essential step in identifying the limits of its operation.
Temperatures can be estimated using an energy balance that accounts for the
gains and losses of energy to the wetland over time and space. The important
gains and losses in the energy balance will vary seasonally. At minimum, a
winter and summer energy balance will be needed to predict the range of
3-27
-------
SECTIONS WETLAND PROCESSES
operating water temperatures, and thus the corresponding range in temperature
dependent pollutant removal rates.
In summer, large amounts of energy are supplied by solar radiation. A small
portion of this recharges the soil energy storage, but most is lost via evaporation
and transpiration. In winter, energy gains are from soil storage, and loss is to the
cold ambient air. If snow or ice is present, radiation, convection, and
sublimation create a balance that dictates the snow surface temperature.
When ice cover is absent, the energy balance is typically such that the gains and
losses of energy are balanced, and the water temperature approaches
equilibrium with the mean monthly air temperature, T (Figure 3-15).
FIGURE 3-15
Correlation between wetland water temperature and air temperatures. Both northern (Listowel) and
southern (Orlando Easterly) systems show effluent water temperatures that follow the mean daily air
temperature during warm months from nearby weather stations (Kadlec and Knight 1996).
30
25-
20-
B 15
£
3
I 10-
-------
SECTION 3 WETLAND PROCESSES
balance as in summer, and temperature decline will typically proceed
throughout the flow path.
The amount of ice formation is determined by dimatological conditions that
vary greatly from one winter to another, Wetland vegetation is effective in
trapping snow to greater extents than unvegetated areas. Therefore, ice
thickness in wetlands may be much less than in adjacent lakes or frost depths in
nearby uplands. The Listowel, Ontario, wetlands experienced ice thickness of
about 100 to 150 mm during flow conditions for a climate typified by a mean
January air temperature of -9 °C (16 °F). Ice or frost depths in the Houghton
Lake, Michigan, wetland range from zero (for copious early snow) to 200 mm for
unvegetated pond zones with little snow. The mean January temperature is -8°C
(18°F), and there is no winter water flow. Kadlec and Knight (1996) and Reed et
al. (1995) provide a thorough discussion of FWS wetland temperature and ice
formation prediction.
Constituent Characteristics
The characteristics (size, density, solid or dissolved), of wastewater constituents
are of major importance in analyzing the performance of any wastewater
treatment processes. These characteristics change as the wastewater proceeds
through various processes in wastewater treatment systems. Wetland processes
can play a role in the separation and solubilization of various wastewater
constituents.
In effect, a FWS constructed wetland replicates a full wastewater treatment train
in terms of types and linkages of the physical, chemical, and biological
processes. It is important in the design and operation of a FWS wetland to
determine the particulate/soluble distribution of the constituents. Settleable
organic solids separated by settling process will serve as an internal load of
dissolved and colloidal solids upon anaerobic decomposition. Biodegradable
dissolved organic solids (VSS) are broken down, releasing ammonia, soluble
reactive phosphorus, dissolved organic carbon, and gases (CO2 and CH4).
Colloidal solids are also released in the decomposition process and include the
heterotrophic bacteria responsible for the decomposition, as well as organisms
and/or viruses of public health significance. The latter two particle types are
adsorbed or affected in the settleable solids and are released to the water column
upon decomposition.
As FWS constructed wetlands are placed further into conventional treatment
trains; i.e., secondary and tertiary applications as opposed to primary and/or
advance primary applications, the physical characteristics (size, density, etc.) of
the constituents must be taken into consideration.
Soluble forms of COD and BOD dominate in the effluent from a wetland. An
example of the partitioning of the various particle size of the constituents can be
seen in Figure 3-16. In the case of an oxidation pond effluent, the majority of the
BOD, for example, is supracolloidal. In the case of the wetland effluent, the
majority of the BOD is soluble. It is the removal of the settleable and
3-29
-------
SECTIONS WETLAND PROCESSES
supracolloidal BOD through the wetland, which accounts for the majority of the
BOD removed in the wetland. The soluble BOD removed is also significant and
represents a net removal since the decay of the settled solids and plant detritus
add to the soluble BOD in the system. This can be seen in the COD values in the
oxidation pond and wetland effluent. The COD values are about the same for
both systems. The BOD/COD ratio has changed significantly through the
system as refractory and more complex organic compounds are formed in the
decomposition of the plant material.
FIGURE 3-16
Distribution of BOD and COD concentration by form (settleable, supracolloidal, or soluble) in
oxidation pond effluent and treatment marsh effluent from Arcata, California (Gearheart 1992).
_o
I
c
u
o
o
90.0 1
80.0
70.0 -
60.0 -
50.0 -
40.0 -
30.0 -
20.0 -
10
D Soluble
D Supracolloidal
• Settleable
Ox. Pond
BOD
Wetland
BOD
Ox. Pond
TSS
Wetland
TSS
Ox. Pond
COD
Wetland
COD
Source and Constituent
Aquatic Vegetation
Types of Wetland Vegetation
Of the thousands of vascular plants on earth, only a relatively limited number
are adapted to the conditions of continual submergence and waterlogged soils.
FWS wetlands may consist of a variety of different emergent, submerged, and
floating aquatic vegetation species, distributed primarily based on water depths.
In general, emergent species are found in shallow water depths, while
submerged species occupy deeper water zones; floating species of vegetation can
occur in both shallow and deeper water areas.
In FWS constructed wetlands, the most common vegetation species have
typically been emergent species such as bulrush, cattails, rushes, and reeds
(Pullin and Hammer 1989, Reddy and Smith 1987). In the past, general practice
was to use either a mono-culture of one species, or a combination of two or more
species in FWS constructed wetlands used primarily for the treatment of
3-30
-------
SECTION 3 WETUND
wastewater. Constructed FWS wetlands that are used as habitat or enhancement
wetlands, will typically be planted with a variety of emergent, submerged, and
floating species. Some of the more common wetland plants used in FWS
constructed wetlands, either for treatment or enhancement, and the species type
and typical water depths of occurrence are listed in Table 3-2.
TABLE 3-2
Some common wetland plants and depths of occurrence used in FWS and floating aquatic
constructed wetland.
Plant type
Emergent
Submerged
Floating
Species name
Typha spp.
Scirpus spp.
Juncus spp.
Carexspp.
Phragmites spp.
Potamogeton spp.
Vallisneria spp.
Ruppia spp.
Nupharspp.
Elodea spp.
Lemna spp.
Elchhomia crassipes
Hydrocotlye umbellata
Azolla spp.
Wolffia spp.
Common name
Cattail
Bulrush
Rushes
Sedges
Reeds
Pond weeds
Tapegrass, wild celery
Widgeongrass
Spatterdock
Waterweed
Duckweed
Water hyacinth
Water pennywort
Water fem
Watermeal
Range of depths (m)
> 0.1 to < 1
>0.1 to<1
> 0.1 to < 0.3
> 0.1 to < 0.3
> 0.1 to < 1
>0,5
>0.5
>0.5
>0.5
>0.5
Flooded
Flooded
Flooded
Flooded
Flooded
Vegetation Patterns
Treatment wetlands develop large amounts of emergent vegetation in areas with
water depth less than about 60 cm deep. In general, larger nutrient supplies
produce larger standing crops. These plants influence treatment performance in
many ways, including;
« Uptake and cycling of nutrients and other elements
• Providing substrate for microbes and epiphytes, which process pollutants
• Creating drag on the flowing water, thereby creating head loss
• Occupying some of the water column, thus excluding liquid volume
3-31
-------
SECTION 3 WETLAND PROCESSES
» Increased plant biomass can increase the background concentrations of
COD and BOD, which can amplify nutrient pulses from the effluent as a
result of the seasonal decay of the vegetation in temperate climates with
distinct growing seasons.
Above-ground macrophyte biomass may be separated into three compartments:
live (green), standing dead (brown, upright), and litter (brown, broken,
prostrate). Different compartments dominate the above-ground vegetation
structure during different seasons. In northern climates, the end-of-season
standing live crop converts to standing dead, and subsequently to litter. In
warmer climates, such phases are shorter and less pronounced, but there are
dormant periods at all latitudes.
Constructed wetlands do not initially possess all vegetative compartments;
typically many months to a few years are required for the vegetative and litter
compartments to fully develop (Figure: 3-17 and 3-18). In Figure 3-18, grass and
duckweed that were predominant during the first year were relatively
uncommon in the second year as cattail and hardstem bulrush grew taller and
either shaded or filled in the open water areas. During this developmental
period, carbon- and plant-dependent wetland functions may not be operating at
their full potential.
FIGURE 3-17
Newly constructed wetlands require a startup period to attain full vegetative cover. Ground level and
aerial reconnaissance were used to follow this process for the Tarrant County project (Alan Plummer
Associates Inc. [APAI] 1995). The litter layer developed subsequently.
100
I
Q
1
1
1
g
90-
80-
70
60-
50-
40-
30
20
10-
0
Typha
Scirpus
Tarrant County, TX
River Treatment Wetlands
0 10 20 30 40 50 60 70 80 90 100 110 120
Days From 7/1/93
3-32
-------
SECTION 3 WETLAND PROCESSES
FIGURE 3-18
Coverage of plants during the startup period of the Arcata Pilot Project wetland.
D)
s
0)
g
o
* *
JliUSn p <*** ff ^ -.'^v^jJ^**1?--. ? ',^^ ^ ' Si ^ t1" &a
Q
Study Period
The amount of biomass is climate and species-specific, as is the stem density.
Cattail (Typha spp.) has a relatively large basal diameter, and occurs at about 40-
50 stems per square meter in treatment wetlands. In contrast, bulrushes (Scirpus
spp.) have smaller stems and may occur at hundreds per square meter. Stem
density is additionally constrained by the growth requirements of the plant in
question.
The space occupied by submerged plant parts acts to reduce detention time
compared to an empty basin of the same depth. Plants block a small fraction (0
to 5 percent), and standing dead and litter can add a comparable fraction of
blockage, leading to a total of 0 to 10 percent. In combination this can result in
wetland porosities from 0.65 to 0.75 where the lower value is associated with
dense mature vegetation (Reed et. al. 1995).
In contrast to the deleterious effect on detention time and head-loss, more
submerged surfaces have the potential to house more microbes and epiphytes,
and thus potentially enhances treatment. The amount of submerged area
contributed by stems and leaves has been measured to range from 1.0 to 7.6
times the bottom area (Table 3-3). Dead plant parts are comparable in biomass
and may contribute a comparable surface area, as does the un-vegetated wetland
bottom in the absence of litter.
3-33
-------
SECTIONS WETLAND PROCESSES
Role of Aquatic Plants in Controlling Treatment Processes
Aquatic macrophytes play an important role in the treatment processes active
within FWS constructed wetlands. The plants, unique to the wetland
environment, both influence the pollutant removal processes and act as sources
and sinks of certain dissolved and particulate water quality constituents.
Wetland plants also play an important role in preventing incoming radiation
from entering the water column. Interception of incoming radiation significantly
reduces algal growth, which can add carbon back to the system via
photosynthesis. The shading of the water surface also can moderate the water
temperature of a wetland. A distinguishing characteristic of FWS constructed
wetlands is that the water temperature profile is buffered from abrupt changes
in the ambient temperature. The cooling potential for any one site is dependent
upon the range of temperatures found at that site, the ET rate, and the extent of
the canopy. While the magnitude of thermal buffering is unique to a site, in
certain locations this effect can be taken advantage of to meet receiving water
temperature standards.
Well-developed stands of vegetation also reduce the natural reaeration process
by influencing the micrometeorology within the wetland and limiting wind
induced turbulent mixing. Lower rates of oxygen transfer, combined with low
algal concentrations and the dissolved oxygen consumed within the water
column to satisfy oxygen demands, usually results in low dissolved oxygen
concentrations in FWS constructed wetlands. Surface level dissolved oxygen
concentrations at 20 to 40 percent of saturation are commonly observed.
While debate surrounds the potential for in-situ reaeration via emergent
macrophytes, no debate exists concerning the ability of submergent plants to
contribute dissolved oxygen. In most cases, emergent and submergent plants are
not found in the same wetland zones. Submergent aquatic macrophytes can
thrive in un-shaded regions of FWS constructed wetlands. These plants
contribute dissolved oxygen directly to the water column while affording a
physical substrate for periphytic and epiphytic bacteria and algae. Plants such
as Potamogeton pectinatus (sago pondweed), are commonly planted in FWS
constructed wetlands to support the nitrification of ammonia and serve as a food
source for aquatic waterfowl.
Floating aquatic macrophytes are subject to being moved by the wind over the
surface of the open water. It is not uncommon to have plants such as Lemna spp.
windrowed amongst and against emergents or a berm, resulting in nearly
complete inhibition of normal photosynthetic reaeration processes. Proprietary
processes have been developed to keep floating aquatic macrophytes from being
redistributed by the wind through various anchoring mechanisms.
Wetland vegetation is also a source of dissolved and particulate material that
combines with the influent wastewater to produce a mixture of biodegradable
compounds similar to the production of BOD via algal growth and degradation
in an oxidation pond. A broad range of heterotrophic and autotrophic
organisms is capable of degrading these compounds.
3-34
-------
SECTIONS WETLAND PROCESSES
Many of the biochemical transformations that occur in treatment wetlands are
mediated by a variety of microbial species residing on solid surfaces such as
those areas provided by plant leaves, stems, and litter. Examples of these
processes include the decomposition of organic matter, periphyton fixation,
nitrification-denitrification, and sulfate reduction. For example, maximum
biofilm production of 1500 mg/m2-d (dry-wt) has been measured in wastewater
treatment wetlands at 60 percent of maximum sunlight (Tojimbara 1986). In
turn, these processes are directly responsible for the water quality improvement
potential of treatment wetlands. The submerged surface area of vegetation in a
wetland is a function of plant type, plant density, and water depth. Reported
submerged plant surface areas for various typical wetland plants are given in
Table 3-3.
TABLE 3-3
Submerged surface area of wetland vegetation, normalized for a depth of 0.5 m.
Site
Arcata, CA
Benton, KY
Houghton Lake, MI
Pembroke, KY
Dominant Submerged Area Depth
Vegetation (n^/m2) (m)
Scirpus acutis
Typha latifolia
Scirpus cyperinus
Typha latifolia
Carex spp.
Typha angustifolia
Typha latifolia.
Scirpus validus
Typha angustifolia
7.6
2.6
1.8
1.0
2,4
2.7
2.1
1.2
1.5
0.6
0.6
0.25
0.25
Unknown
0.3
0.3
0.2
0.2
Normalized
Submerged
Area (m2/m2)
6.5
2.2
3.6
2.0
Unknown
4.5
3.5
3.0
3.7
Source: Kadlec and Knight 1996, Kadlec 1997, Pullin and Hammer 1991, Gearheart et al. 1999
(publication in progress).
Depending on the dominant plant type, plant surface area may or may not be a
function of wetland depth. If the primary contribution to surface area in the
wetland is the bottom litter layer, then the surface area available for attached
growth does not increase significantly with depth once the litter layer is
submerged. In a scenario such as this, effluent quality may be largely
independent of water depth (Kadlec and Knight 1996). For example, data from a
sedge meadow at Houghton Lake indicated that very little additional surface
area was observed for water depths greater than 250 mm (Kadlec, 1997).
In contrast, the surface area of live and dead plant material and litter for a Typha
zone of the wetland at Houghton Lake is still showing a significant increase at
3-35
-------
SECTIONS WETLAND PROCESSES
0.3 m (Figure 3-19). The change in leaf and stem (not litter) surface area with
depth in a bulrush (Scirpus acutis) and cattail (Typha latifolia) zone of the Arcata
Treatment Marsh are shown in Figure 3-20. In this example, the leaf and stem
surface area continues to increase significantly up to the maximum depth
measured of 1 meter. From, these results, it can be concluded that in wetlands
supporting plants that grow in deeper water (e.g., cattails and bulrush), the
surface area for attached growth does increase with depth.
FIGURES-IS
Stem, leaf and litter cumulative surface area for Typha spp. in Houghton Lake discharge zone
wetland (Kadlec, 1997).
3.0
IS
Q>
O
I
(0
2.5 -
2.0-
0.5 -
o
0.0
^ Typha latifolia
O Typha angustifolia
0.05 0.1 0.15 0.2
Water Depth (m)
0.25
0.3
0.35
Wetland vegetation also has an effect on the hydraulic characteristics of the
wetland, which directly influences water quality constituent removal processes.
Wetland vegetation can
» Increase water losses through plant transpiration,
» Decrease evaporation water losses by shading water surfaces and cooling
water temperatures,
• Create friction on the flowing water and, thereby, creating head-loss and
floeculation of colloids,
• Provide wind blocks, thus promoting quiescent water conditions and
protection for floating plants such as duckweed,
• Provide complex water column flow pathways, and
* Occupy a portion of the water column, thus decreasing detention time
3-36
-------
SECTIONS WETLAND PROCESSES
In summary, it is the vegetation, specifically the emergent and submergent
vegetation that gives a FWS constructed wetland its capability to treat
wastewater effectively in a passive manner. Free water surface constructed
wetlands are unique in that they grow their own physical substrate for
periphytic microorganisms and epiphytic plants while minimizing incoming
radiation addition. The fact that fixed-film biological reactions, sedimentation,
and anaerobic digestion can all occur in an aquatic system can be attributed to
the ecosystem created by the aquatic macrophytes. Without this vegetated
component, the same physical conditions would result in an oxidation pond
producing a large amount of total suspended solids (algae) in the effluent.
FIGURE 3-20
Stem and leaf surface area for Scirpus acutis (hardstem bulrush) and Typha latifolia (cattail) in Arcata
Treatment Wetland (Gearheart et al., 1999, publication in progress).
n
o>
•o
It
co
3
|
O
10.0--
9.0-
8.0-
7.0-
6.0-
5.0-
4.0-
3.0-
2.0-
1.0-
0.0--
0.0
• Scirpus acutis
• Typha latifolia
H 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 H
0.2 0.4
Stem Depth (m)
0.6
0.8
3-37
-------
SECTION 4
Performance Expectations
Approach to Performance Evaluation
Free water surface constructed wetlands tend to function as a sequence of
coupled processes: discrete settling, flocculent settling, and benthic
decomposition (ammonification and release of soluble degradable organics,
soluble BOD removal, nitrification, phosphorus uptake, denitrification, etc.).
The contribution and even presence of each process within a FWS constructed
wetland is highly dependent on the design and operation of the treatment
wetland.
The performance and permit compliance of operating FWS constructed wetland
treatment systems reveals the range of effluent quality and the variability in
performance, possible with these types of systems. Evident in this analysis is the
range of conventional treatment strategies provided to wastewater, and thus the
range of constituent loads, to which FWS treatment wetlands are subjected.
Further, many of the wetlands systems with sufficient operational and design
data represented in the NADB could be characterized as constructed wetlands
receiving high quality effluent from advanced wastewater treatment processes.
This allows in-situ contributions of BOD, TSS, and nutrients to dominate the
wetland effluent and hence add to the variability of the wetland effluent quality.
This section describes and compares the performance of a subset of operating
FWS constructed wetlands for which sufficient data and information were
available.
In addition to the performance assessment, an analysis of permit compliance for
those sites that had both permit limits and operational data of comparable
frequency available in the NADB is included. Actual system operational flows
were compared to design flows as a measure of the loading to each system
during the period evaluated. For sites with adequate data, the length of the data
record, the percent compliance, and the average and maximum effluent
concentrations during that period are reported. Because of limitations of the
NADB, a subset of the operational data from most of these systems is included in
this analysis.
The performance assessment and permit compliance analyses presented are
organized by constituent, with subsections for BOD, TSS, nitrogen, phosphorus,
fecal coliform, metals, and organics. The chapter concludes with a discussion of
background concentrations and stochastic variability.
4-1
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Methodology of Performance Evaluation
The FWS constructed wetlands used in this technology assessment were the
systems which best met the minimum criteria for inclusion for analysis (see
Section 2 and Table 2-5). This Technology Assessment Database (TADB)
includes selected systems from the NADB and additional systems for which
operational data are available. The wetland sites with sufficient water quality
and operational data to be included in the TADB and used for this technology
assessment are reported in Table 4-1. For some systems, data on all water
quality constituents were available while for other systems only select
constituents were available. Permit compliance was evaluated using systems in
the NADB, but not necessarily in the TADB.
The performance evaluation of FWS constructed wetlands has been analyzed at
three different levels. The first level includes a summary analysis of all available
data for systems listed in Table 4-1, calculation of the mean influent and effluent
concentrations and their range of values. The mean and range of mass loadings
for each water quality constituent are given in Table 4-2. This first level of
assessment is useful in the context of summarizing the range of operating
conditions of FWS constructed wetlands and the range of response in terms of
effluent concentration. At this level, the wide range of applications and expected
performance for operating FWS treatment wetlands are summarized. No
accounting for differences in upstream waste treatment processes, geometric
configuration, planting strategy, inlet/outlet works, climate, etc. has been made
even though, each of these factors can significantly affect the effluent quality of a
FWS constructed wetland.
In the second level of performance data analysis, those systems with the most
extensive monthly influent/effluent data for the constituents of interest are
compared. This level of analysis is presented in terms of cumulative probability
over the data collection period. The third level of analysis is designed to
determine how individual systems perform in terms of effluent concentrations
over the range of their loadings. In the third level of analysis, monthly loadings
versus effluent concentrations for a single site are compared, thus demonstrating
the expected variability within a single system.
4-2
-------
SECTION 4 PERFORMANCE EXPECTATIONS
TABLE 4-1
Water quality constituent data availability for the FWS constructed wetland systems included in this
assessment (TADB), as identified in Table 2-6.
Water Quality Parameter
Wetland System BOD TSS NH.-N TKN NO,-N TN OrgN TP DP FC
Areata Pilot I Cell 8 ...
Aroata Pilot II * • « »
Arcata Treatment • •
Arcata Enhancement Allen • •
Arcata Enhancement « • •
Beaumont • » •
Benton Cattail ••»• *•».
Benton Woolgrass •••• •»**
Brookhaven Meadow Marsh •••••• •
Cannon Beach « • « »
Central Slough •••• *•**
Clermont Plot H • » • • •
Columbia *
Fort Deposit « • » • « • »
Gustine (89-90) 1A » • » • »
Gustine (89-90) 1B • • » • »
Gustine (89-90) 1C * • • • .
Gustine (89-90) 1D • • • • •
Gustine (89-90) 2A • • « • *
Gustine (89-90) 2B » • «
Gustine (89-90) 6D • • » • »
Gustine (94-97) • •
Houghton Lake » »
Iron Bridge »••• ****
Lakeland » • •
Listowel 4 ***•*•••••
Manila • •
Minot « • »
Mt. Angel • • •
Orange County «••• *•**»»
Ouray •
Pembroke FWS 2 ••••»•••
Poinciana Boot •«»• *•**
Reedy Creek WTS1 »•». ***.
Reedy Creek OFWTS .»•••• *
Sacramento • • • • • •
Sea Pines Boggy Gut ••«• *••*
Tres Rios Hayfield « • « • » •
Vereen Bear Bay ••«• *•••
West Jackson County • • • • • • • •
4-3
-------
SECTION 4 PERFORMANCE EXPECTATIONS
TABLE 4-2
Summary of performance data and loadings for TADB systems analyzed in this assessment (listed in
Table 4-1).
Constituent
Influent (kg/ha d)
Min Mean Max
Biological Oxygen Demand (BOD) 0.04
Total Suspended Solids (TSS)
Ammonia (NH4-N)
Total Kjeldahl Nitrogen (TKN)
Nitrate (NOa-N)
Total Nitrogen (TN)
Organic Nitrogen (OrgN)
Total Phosphorus (TP)
Dissolved Phosphorus (DP)
Fecal Coiiform (FC) (col/1 OOmL)
0.07
0.02
0.04
0.05
0.12
0.02
0.01
0.01
31
22
3
5
0
3
1
1
0
.5
.8
.9
.0
.8
.2
.6
183
92
16
20
3.5
9.9
5.7
4.4
1.3
Influent (mg/L)
Min Mean Max
1
1
.7
.0
0.63
1
.3
0.31
2.1
0.
0.
0.
1
74
27
23
.7
70
69
8.7
18
3
12
5.6
4.1
2.6
73,000
438
588
29
51
13
32
18
11
5.7
360,000
Effluent (mg/L)
Min Mean Max
1.2
1.1
0.07
0.82
0.01
0.85
0.71
0.09
0.04
47
15
15
6.8
11
1.2
4.0
2.1
2
1.5
1,320
69
40
23
32
3.5
9.8
3.2
4.2
3.7
9,800
BOD Performance
Database Assessment
The relationship between average BOD loading and average BOD effluent
concentration for TADB systems shown in Figure 4-1. There is a general linear
trend between increased BOD loading and increased effluent concentration over
the loading range of 0.1 to 180 kg/ha-d. Considering the wide range of
conditions, wetland design, and data quality, a general trend exists between
increased loading and decreasing effluent quality. Specific systems have BOD
effluent versus BOD loading curves, which are better correlated and predict
lower effluent quality compared to the general trend observed.
As shown in Figure 4-1, considerable effluent variation exists for a given BOD
loading. At a BOD loading of 25-kg/ha-d, the effluent concentrations vary from
9 to 35 mg/L. At lower BOD loading rates, the effluent BOD varied from 1 to 8
mg/L (BOD loading rate of 0.1 to 8 kg/ha-d). The effect of the background BOD
due to plant decomposition is evident in systems with low loading rates. In
addition to plant decomposition, relatively small changes in the inlet/outlet
region, levels of animal activities, or weir location and operations, can all
significantly affect the effluent BOD concentration under low loading rates.
4-4
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-1
Average BOD loading rate versus effluent BOD concentration for TADB sites.
O)
E,
Q
O
m
0)
—
iy
OU 1 ' " """ '
70 -
60-
50-
40-
30-
20 -
10-
*
^
^
*
*** * *
fP
** * *
^ 1r** *
<•» •* - -- - -
50 100
BOD Load (kg/ha-d)
150
200
Temporal BOD Performance
A summary of BOD loading versus effluent BOD concentration for the treatment
and enhancement wetlands at Arcata, California, is given in Figures 4-2 and 4-3,
respectively. Seven years of monthly data shows the normal variation observed
in two systems in the same location receiving different loads. As seen in Figure
4-2, the treatment marsh effluent BOD concentration is sensitive to influent BOD
while the enhancement marsh effluent is not as sensitive to the influent BOD
concentration.
FIGURE 4-2
Monthly influent and effluent BOD values for Arcata's treatment wetland.
Q
O
m
120
100-
80-
60-
40-
20-
— Influent
•Effluent
0
Jan-90 Jan-91
Jan-92 Jan-93 Jan-94
Date
Jan-95 Jan-96 Jan-97
4-5
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-3
Monthly influent and effluent BOD values for Arcata's enhancement wetland.
45
-s 40
13»35
E 30
O 25
? 20
% 15
I 10-
** 5-
Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
Date
Effluent cumulative probability BOD levels from West Jackson County,
Mississippi, are shown in Figure 4-4, This particular system shows effluent
concentrations between 2 and 20 mg/L over influent BOD levels ranging from 8
to 48 mg/L with a mean effluent BOD of 4 mg/L and a mean influent value of
24 mg/L.
FIGURE 4-4
Influent and effluent monthly BOD cumulative probability values for West Jackson County,
Mississippi.
99.00
95,00
^ 90.00
re
70.00
* 50.00
I 30.00
= 10.00
5.00
1.00
W. Jackson County
Influent
Effluent
i i
i i
I
10
20 30
BOD (mg/L)
40
50
4-6
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Influent and effluent BOD data for Lakeland, Florida, are given in Figure 4-5. In
this system, the majority of influent values are less than 5 mg/L and during one
12-month period, the effluent BOD is greater than the influent. In this case, the
internal processes producing total suspended solids and dissolved BOD from
aquatic plant and epiphytic primary production and decomposition increase the
effluent BOD above the influent BOD.
FIGURE 4-5
Influent and effluent monthly BOD for Lakeland, Florida.
o
o
m
Date
The Fort Deposit influent and effluent BOD data are presented in Figure 4-6. As
shown in the figure, this system exhibits consistently effective BOD removal.
Effluent BOD concentrations are almost always low, between 2 and 15 mg/L,
while the influent concentration varies from 18 to 100 mg/L.
FIGURE 4-6
Influent and effluent monthly BOD cumulative probability for Fort Deposit, Alabama.
>t
•ft-*
1
a.
_>
«•*
JS
3
E
o
aa.uu
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 nn
- / '*J
} /
~\ /• Fort Deposit
J / Influent
I "^
-j- / tniueni
I i I f I I I I i I I I I | | I 1 I i I i I I
20 40 60 80
BOD (mg/L)
100
120
4-7
-------
SECTION 4 PERFORMANCE EXPECTATIONS
The relationship of BOD loading to effluent BOD concentration for the Arcata
Treatment Marsh is shown in Figure 4-7. The BOD loading ranged from 76 to
605 kg/ha-d, with an average of 180 kg/ha>d. As might be expected, better
relationships between loading and effluent concentrations were found on a site-
by-site basis than observed by lumping data from all the sites together or even
when comparing two systems at the same site.
The Arcata Treatment Marshes have removed BOD at a constant rate of 68,000
kg/ha-yr, for the last 7 years. These three treatment wetlands with a total area of
1.86 ha operate in parallel and remove approximately 30 percent of their influent
BOD. This constant removal rate can be seen in Figure 4-8, in which the
accumulated BOD mass in and out of the treatment wetland is plotted.
Effluent BOD concentration from the Arcata Pilot Project can be predicted using
Equation 4-1:
Ce = 3.42 + 0.262 Q (4-1)
Where: Ce = effluent BOD (mg/L)
Q = influent BOD (mg/L)
This equation fit the 3 years of experimental data for cells with hydraulic
residence times ranging from 6 to 12 days, with an R2 of 0.91, indicating a fairly
constant relationship.
FIGURE 4-7
Monthly BOD loading rate versus BOD effluent concentration for Arcata Treatment Marsh.
100
'BJ ^o -
Q 60 H
o
m 40 -
4-i
I 20 n
5=
m 0
*
•
100 200 300 400 500 600 700
BOD Load (kg/ha-d)
4-8
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-8
Cumulative monthly mass influent and effluent BOD for the Arcata Treatment Wetland. The area
between the two curves is representative of the mass of BOD removed.
I
Q
O
ffl
ffi
J&
"3
E
O
1.4E+06
1.2E+06-
1.0E+06-
8.0E+05 '
6,OE-i-05 '
4.0E+05 '
2.0E+05 -
O.OE+00
'Influent
•Effluent
Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
Date
BOD Permit Compliance
Enough information was available from the NADB to evaluate BODS permit
compliance for 12 FWS treatment wetland systems. Effluent BOD5 permit limits
varied from 5 to 30 mg/L on a monthly average basis. Summer BOD5 limits
were more restrictive than winter limits for five of these systems. In general,
FWS constructed wetlands have been very effective at meeting BOD5 effluent
limits, even as low as 5 mg/L.
Only four of the 12 FWS constructed wetlands had less than 100 percent
compliance with BOD5 permit limits during the analyzed period of record. The
Central Slough, South Carolina, natural treatment wetland exceeded its effluent
permit limit of 30 mg/L just once in 69 months of operational data. Flow to this
system was about 40 percent of design capacity during that period. The Fort
Deposit, Alabama, constructed treatment wetland exceeded the summer BOD5
limit of 10 mg/L one month out of seven with a concentration of 13 mg/L. Flow
at that time was only about 54 percent of design capacity. The Norwalk, Iowa,
system exceeded its BOD5 limit of 30 mg/L six times during the 35-month record
analyzed. The maximum-recorded effluent BODS during this period was 70
mg/L. Flow averaged about 58 percent of design flow during that period. The
Pembroke, Kentucky, constructed wetland exceeded its 10 mg/L limit about 67
percent of the time during a 9-month period. The maximum, recorded effluent
value was 24 mg/L at an average flow of 84 percent of design.
4-9
-------
SECTION 4 PERFORMANCE EXPECTATIONS
TSS Performance
Database Assessment
The effectiveness of FWS treatment wetlands to remove TSS is recognized as one
of their principal advantages. The relationship between TSS loading and
effluent TSS levels for the entire data set is shown in Figure 4-9. Over a range of
loadings from 0.5 to 180 kg/ha-d, there does not appear to be any relationship
between loading and effluent quality with this data set. What is apparent is that
under a fairly wide range of solids loadings, relatively low effluent TSS
concentrations can be attained.
Because physical processes dominate the removal of TSS, it is expected that, to a
point, TSS effluent levels are not affected by hydraulic or solids loading rates.
The dominant TSS removal processes occur within the first 1 to 2 day hydraulic
residence time (HRT) period. This effect can only be seen in transect data with 1
to 2 day increments. Most of the wetlands in the wetland database have
detention times in excess of 2 days, which allows the removal of TSS to be
masked by subsequent internal generation of TSS. The variation in the effluent
TSS shown in Figure 4-9 is most likely related to internal TSS sources such as
algal growth, sloughed epiphytes, animal sources, re-suspension, or detrital
particles.
In the case of TSS effluent cumulative probability distribution, there are
examples of systems that are consistently effective and systems in which the
background levels are sometimes greater than the influent. For example, the
Fort Deposit, Alabama, influent TSS levels varied from 18 to 183 mg/L, with an
average loading of 7.4 kg/ha-d, while the effluent TSS levels varied from 3 to 39
mg/L, representing a significant TSS removal rate (Figure 4-10). In contrast,
Orange County, Florida, influent TSS ranged from 1 to 4 mg/L, while the
effluent ranged from 1 to 17 mg/L, with an average effluent of 4 mg/L, 2.6
mg/L greater than the average influent concentration. Based on data from sites
like Orange County, it can be concluded that wetlands generally will not reduce
TSS concentrations below 3 mg/L.
Removal of TSS is most pronounced in the inlet region of a FWS constructed
wetland. Transect data from pilot project studies at Arcata show this pattern of
removal (Figure 4-11). Generally 50-60 percent of the TSS from oxidation pond
systems is removed in the first 2-3 days of nominal hydraulic detention time.
Gravity settling processes account for most of this removal, and the overall
removal efficiency is a function of the terminal settling velocity of the influent
biosolids. Within the TSS loading range of 50 to 200 kg/ha-d, the removal of the
settled total suspended solids does not require any routine solids handling
operation. The separated solids undergo anaerobic decomposition, releasing
soluble dissolved organic compounds and gaseous by-products, carbon dioxide,
and methane gas, to the water column.
Long term studies from individual sites have shown low and stable effluent
concentrations from a relatively wide range of TSS loading rates. The TSS
4-10
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-9
Monthly TSS loading versus effluent TSS concentration for TADS wetland systems.
45
-. 40"
^35-
£. 30-
(I) 21
t 20-
| 15-
i= 10-
* 5i
0
50 100 150
Solids Load (kg TSS/ha-d)
200
FIGURE 4-10
Cumulative probability distribution of monthly influent and effluent TSS concentration for Fort
Deposit wetland.
£
3
(0
O
£
1
"•§
3
E
£3
O
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 00
-
~~ (^ ' ^-* — """""*"
r ''"^
— / --X
j ^r'~'
~l ~-^"
} ^'X/
I x-^
T" x-
Fort Deposit
Influent
Effluent
I , I , I
0 50 100 150 200
TSS (mg/L)
effluent concentrations rates from the Arcata Enhancement Wetland are
consistently low, less than 5 mg/L, 90 percent of the time, with an annual
average loading of 16 TSS kg/ha-d (Figure 4-12). The Arcata enhancement
marsh has continued to remove TSS at a constant rate of approximately 90
percent for the last 6 years. An operational change in January of 1991 increased
the BOD removal rate, while TSS removal has remained constant. An increase in
hydroperiod (depth increase from 0.25 to 0.5 m) coupled with no alteration in the
outlet weir setting over the year has stabilized the effluent TSS and BOD levels.
The effluent TSS concentration has not tracked the influent levels with the
operational strategies used over the last 6 years.
4-11
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-11
Weekly transect TSS concentration for Arcata's Cell 8 Pilot Project, with theoretical retention time of
6 days, receiving oxidation pond effluent.
160
140-
120-
i> 100-
H 80 -
03 60-
40-
20-
Average
Detention Time
FIGURE 4-12
Weekly Influent and effluent TSS concentration for Arcata Enhancement Wetland,
60.00
j" 50.00 H
-E 40.00 -
CO
[2 30.00-
w
a 20.00 -
< 10.00-
0.00
influent
Effluent
Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
Date
Temporal TSS Performance
The Arcata Treatment Marshes have the highest average TSS loading in the
TADB (180 kg/ha-d, average influent TSS of 60 mg/L), yet the removal has
continued at a more or less constant rate of about 50 percent over the last 6 years
(Figure 4r-13). The TSS effluent levels from the treatment marsh are less than 27
mg/L, 50 percent of the time.
4-12
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-13
Cumulative yearly mass influent and effluent TSS for Arcata Treatment Wetland.
en
P
E
3
u
1.6E+06
1.4E+06 -
1.2E+06-
1.0E+06
8.0E+05
6.0E+05
4.0E+05 '
2.0E+05
O.OE+00
Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
Date
TSS Permit Compliance
Thirteen FWS constructed wetland systems with permit and effluent data were
available in the NADB that could be used to evaluate permit compliance.
Effluent TSS permit limits varied from 10 to 30 mg/L on a monthly average
basis. One system (Reedy Creek, Florida) also had an annual average TSS limit.
Only one of these systems had seasonal limits for TSS (Vermontville, Michigan).
In general, the FWS constructed wetlands were able to meet effluent TSS limits.
The cases where limits were exceeded resulted from poor vegetative cover and
the subsequent growth of phytoplankton or solids re-suspension. Of the 13
systems in the NADB, 8 had 100 percent compliance with TSS effluent limits.
Five FWS constructed wetlands had less than 100 percent compliance with TSS
permit limits during the period of record in the NADB. The Central Slough,
South Carolina, natural wetland exceeded a 30 mg/L effluent limit twice during
24 months of operational data, and had a monthly maximum of 66 mg/L during
this period. Benton, Kentucky, Cell 2 exceeded its 30 mg/L permit level twice
during 20 months, with a maximum during this period of 53 mg/L. Average
flow to this cell was about 65 percent of design flow. Benton Cell 1 exceeded its
permit limit of 30 mg/L three times during the same 20-month period. Average
flow in this cell was also about 65 percent of design. The Norwalk, Iowa,
constructed wetland was in compliance with the 80 mg/L permit limit about 69
percent of the time during the 35 months of record.
4-13
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Nitrogen Performance
Effluent concentration data for nitrogen species shows considerable variation in
response to the nitrogen loading. Total nitrogen (the sum of all nitrogen species)
and total Kjeldahl nitrogen (organic plus ammonia nitrogen) effluent
concentrations are generally correlated to their respective loadings. However,
individual forms of nitrogen, ammonia, nitrate, and organic nitrogen, may
exhibit very little correlation between effluent concentrations and influent
loadings. This latter set of nitrogen species has both sources and sinks within
FWS wetlands and a speciated nitrogen balance for a specific system is necessary
to analyze removal performance.
In a number of cases, effluent concentrations of ammonia or nitrate N have been
found to be higher than influent concentrations. This concentration increase is
rarely the case for organic or total N. The conclusion from these observations is
that the sequential nitrogen transformation processes result in an overall uni-
directional conversion of elevated total and organic nitrogen forms to oxidized
or gaseous nitrogen forms in treatment wetlands. However, these processes can
also lead to increasing concentrations of intermediate nitrogen forms due to
temporal and spatial differences in conditions necessary to support
denitrification (alkalinity/carbon concentrations, and redox potential).
Distribution of various species of nitrogen within a wetland indicates that the
nitrogen dynamics are affected by the influent loading, the degree of plant
coverage and maturity of emergent vegetation (Sartorius et al. 1999).
Organic Nitrogen Performance
Nearly all the FWS treatment wetlands that have been studied have reported
reductions in total nitrogen and organic nitrogen. The transient nature of
organic nitrogen is a consequence of the balance of sources and sinks active at a
given site. Organic nitrogen is produced by anaerobic degradation and is
converted to ammonia nitrogen by ammonification processes making it difficult
to determine the relationship between organic nitrogen loading and effluent
concentration. Analysis of performance data requires a complete nitrogen
balance for a particular site; it is somewhat meaningless to use data from
different sites. A better relationship between influent and effluent organic
nitrogen was found for individual sites. For example, a consistent removal of
organic nitrogen from influent mean values of 25 mg/L to effluent mean value
of 8 mg/L is shown in the data from West Jackson County (Figure 4-14).
4-14
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-14
Cumulative probability distribution of influent and effluent organic nitrogen for West Jackson
County, Mississippi.
.0
a
A
O
<0
"5
3
u
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 nn
i •
/ j"
— f /
I /
— I /
I , I
W. Jackson County
Influent
• n Effluent
!
0.0 5.0 10.0
Organic N (mg/L)
15.0
20.0
Ammonia Nitrogen Performance
The ammonia effluent concentrations observed for the range of loading rates in
the TADB is shown in Figure 4-15. Ammonia nitrogen effluent concentrations
are poorly correlated with ammonia loading rates, due to the internal ammonia
contribution from organic nitrogen (org N) associated with the TSS. Ammonia
nitrogen shows considerable variability for a given loading. At loadings
between 2.0 and 3.0 kg/ha-d, effluent ammonia concentrations ranged from 0 to
20 mg/L. Systems represented in the lightly loaded region generally showed
low effluent ammonia levels.
FIGURE 4-1S
Ammonia nitrogen loading versus effluent ammonia concentrations for TADB systems.
25
S
0 2 4 6 8 10 12 14 16 18
NH4 Loading (kg/ha* d)
Presentation of ammonia loading versus effluent concentration data for a
number of different systems tends to mask the relationship between the various
forms of nitrogen, the influent concentrations of ammonia, the water
temperature, and the detention time of the wetland. The Beaumont, Texas, FWS
constructed wetland is an example of a system that showed very consistent
ammonia nitrogen removal (Figure 4-16). Over a 4-year period, the 8 cell system
4-15
-------
SECTION 4 PERFORMANCE EXPECTATIONS
of the Beaumont wetland had an average hydraulic detention time of 17.4 days,
an average water temperature of 22.5 °C, and an average ammonia loading of 4.3
kg/ha-d. As shown in Figure 4-17, the average ammonia removal was nearly 90
percent.
Ammonia nitrogen levels in constructed wetlands can increase within the
wetland as decomposing particles become soluble. This increase mirrors the
contribution of dissolved organic carbon as settled solids decompose in the inlet
zone of the wetland.
FIGURE 4-16
Cumulative probability distribution of monthly influent and effluent ammonia nitrogen from
Beaumont, Texas.
>, 95.00
£ 90.00
5 70.00
£ 50.00
> 30.00
| 10.00
§ 5.00
0
1 on
0.
_ J X
_/" . J Beaumont
/ r / Influent
— . • EfHuttnl
I , I , I
0 5.0 10.0 15.0 20
NH4 (mg/L)
FIGURE 4-17
Ammonia nitrogen removal for Beaumont, Texas, through 8 cells with a total HRT of 17 days.
5 10 15
Hydraulic Detention time (days)
20
Total Kjeldahl Nitrogen Performance
Total Kjeldahl nitrogen (TKN) loading versus effluent levels for TADB systems
shows general trends of increased loading producing increased effluent
4-16
-------
SECTION 4 PERFORMANCE EXPECTATIONS
concentrations (Figure 4-18). Because TKN is the sum of the organic nitrogen
and the ammonia, the correlation between influent and effluent TKN is expected
to be higher than for the individual components because analyzing TKN
eliminates the effects of internal conversion reactions between the organic and
ammonia nitrogen. Generally, those systems with an influent TKN
concentration less than 2 mg/L had effluent ammonia concentration
significantly less than 1 mg/L, indicating that in treatment wetlands, the
background level of TKN is attributable to the organic nitrogen. The cumulative
probability distribution of the influent and effluent TKN concentration for the
Central Slough wetland is shown in Figure 4-19. The Central Slough system had
an average influent concentration higher than the TADB average (17 versus 12
mg/L), and an average removal rate of 75 percent, slightly higher than the
TADB average of 67 percent.
FIGURE 4-18
Total Kjeldahl nitrogen loading versus effluent ammonia concentrations for the TADB.
35
^ 30-
|> 25-
Z 20
o>
I 10 - ^
W sJ t „
0|L±_i_±_
10 15
TKN Load (kg/ha-d)
20
25
FIGURE 4-19
Cumulative probability distribution of monthly influent and effluent TKN from Central Slough, South
Carolina.
99.00
95.00
S" 90.00
" 70.00
50.00
Q.
_>
1 30.00
3
E
o 10.00
5.00
1.00
Central
influent
Effluent
10 15 20
TKN {mg/L)
25
30
4-17
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Nitrate and TIN Performance
Nitrates are also transient nitrogen species in FWS wetlands. The extent of
nitrate removal or production depends on the presence and distribution of
aerobic (nitrification produces nitrate from ammonia) and anoxic (denitrification
in which nitrate is converted to nitrogen gas) regions within a FWS wetland. As
shown in Figure 4-20, essentially no relationship exists between nitrate loading
and effluent quality in the TADB systems. Only in the case of a highly nitrified
effluent would one expect to see a relationship between nitrate loading and
effluent nitrate concentration.
FIGURE 4-20
Nitrate nitrogen loading versus effluent nitrate concentrations for the TADB.
H.U -
..j.3.5-
"& 3.0 -
E
•S 2.5 -
i 2.0-
1 1-5-
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-21
Cumulative probability distribution of monthly influent and effluent nitrate concentrations for Orange
County, Florida.
o
CL
0)
2
3
E
o
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 nn
T i~~~~~~~~~~~~
— ; Orange County
/ *~ Influent
/ Etlluent
i < i , i , i ,
0,0
0.5
1.0 1.5
NO3 (mg/L)
2.0
2.5
FIGURE 4-22
Monthly influent and effluent of total inorganic nitrogen (TIN) for the Arcata Enhancement Wetland.
Jan-94 Feb-94 Mar-94 Apr-94 May-94 Jun-94 Ju!-94 Aug-94 Sep-94 Oct-94
Date
Total Nitrogen Performance
Total nitrogen, the sum of the organic and inorganic forms, in FWS constructed
wetlands shows a correlation between increased loading and increased effluent
concentrations (Figure 4-23). However, within the range of 0.1-6 3 kg/ha-d
considerable variation exists in the effluent concentrations.
4-19
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FlSURi4-23
Total nitrogen loading versus effluent total nitrogen concentrations for TADB wetland systems.
E 8H
NMK*
f 6H
ui 2 -
0
0
468
TN Loading (kg/ha-d)
10
12
The typical range of inlet and outlet TN concentrations for the first 12 cells of the
FWS constructed wetland at Iron Bridge, Florida, is illustrated in Figure 4-24.
Individual maximum monthly outlet concentrations are more than two times
higher than the long-term average.
FIGURE 4-24
Range of monthly inlet and outlet TN concentrations for cells 1 through 12 at the Iron Bridge FWS
wetland near Orlando, Florida.
Aug-87 Feb-88 Sep-88 Apr-89 Oct-89 May-90 Nov-90 Jun-91 Dec-91 Jul-92 Jan-93
4-20
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Nitrogen Permit Compliance
Ammonia Nitrogen
Ten FWS constructed wetland systems with NH4-N permit and effluent data
were available for evaluation from the NADB, The NH4-N effluent permit limits
varied from 1 to 20 mg/L on a monthly average basis. Six out of ten of these
systems had seasonal limits for NH4-N.
Effluent NH4-N limit compliance continues to be a challenge for FWS
constructed wetlands. Of the ten systems in the NADB, only four had 100
percent compliance with NH4-N effluent limits.
Six FWS constructed wetlands had less than 100 percent compliance with NH4-N
permit limits during the period of record in the NADB. Benton Cells 1 and 2 had
100 percent compliance with winter NH4-N effluent limits of 10 mg/L, but they
exceeded summer limits of 4 mg/L 83 percent of the time during their 6 months
of record. Maximum outlet NH4-N concentrations were about 12.5 mg/L at an
average flow of approximately 69 percent of design flow. The Fort Deposit,
Alabama, constructed wetland exceeded its NH4-N effluent limit of 2 mg/L only
once out of 25 months with a maximum monthly value of 4.84 mg/L. The
Norwalk, Iowa, wetland exceeded its summer limit of 8 mg/L only one time out
of 20 months of record in the NADB. The maximum monthly value was 16.3
mg/L for Norwalk. The West Jackson County, Mississippi, system missed its
NH4-N permit limit of 2 mg/L 6 months out of 33 months of record with a
maximum value of 3.92 mg/L. Average flow during this period was about 96
percent of the design flow.
Total Nitrogen
Only four FWS constructed wetlands had TN permit limits and associated data
in the NADB. The permit limits for TN varied from 2.0 to 2.5 mg/L for these
wetlands. The Reedy Creek, Florida, natural wetland systems had annual
average limits in addition to monthly limits. A few treatment wetlands
receiving highly pretreated (fully nitrified) wastewater have been able to attain
low TN effluent limits.
Two out of four systems in the NADB had 100 percent compliance with their TN
effluent limits. The Iron Bridge, Florida, constructed treatment wetland met its
TN effluent limit of 2.3 mg/L during all of the 63 months of record in the NADB
at an average flow about 61 percent of design. Maximum TN outlet
concentration recorded during this period was only 1.7 mg/L. The Orange
County, Florida, hybrid treatment wetland (both constructed and natural cells in
series) met a TN permit limit of 2.2 mg/L 86 percent of the 37 months of record.
The maximum recorded TN value during this period was 2.6 mg/L at an
average flow of about 48 percent of design. The Reedy Creek System 1 exceeded
TN effluent permit limits of 2 to 2.5 mg/L about 15 percent of the time during
the period reported in the NADB. The maximum recorded annual average TN
outflow value for this system was 8.2 mg/L and was the result of a 6-month
upset in the activated sludge conventional treatment system preceding the
natural wetland.
4-21
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Total Phosphorus Performance
Database Assessment
Total phosphorus removal in wetlands has been of great interest to system
operators and researchers, thus the amount of data and analysis is much greater
than for many other constituents. There are hundreds of wetland-years of
performance data for phosphorus, spanning two decades. The majority of these
studies focused on non-domestic wastewater phosphorus sources. While
comparisons can be made, it is important to separate the inorganic participate
phosphorus performance from the organic particulate phosphorus performance.
Because of the great amount of study conducted regarding phosphorous in
treatment wetlands, Table 4-3 is provided to illustrate the range of hydraulic
loading rates and TP concentrations and resulting outlet concentrations (annual
averages) for natural and constructed wetlands in the NADB. For the NADB
sites considered the average TP annual average removal ranged from as low as
9.7 percent to greater than 98 percent. Overall, the mean average annual
removal rate for this collection of sites was 61 percent with a standard deviation
of 30 percent
TABLE 4-3
Total Phosphorus Removal Rates for Non-Forested Treatment Wetlands (NADB, 1993).
No. of
Site Wetlands
Hidden Lake, Florida
Des Plafnes, Illinois
ENR, Florida
OCESA, Florida
Iron Bridge, Florida
Cobalt, Ontario
Listowel, Ontario
Great Meadows, Massachusetts
Houghton Lake, Michigan
Pembroke, Kentucky
Sea Pines, South Carolina
Fontanges, Quebec
Benton, Kentucky
Leaf River, Mississippi
Lakeland, Florida
Clermont, Florida
Brookhaven, New York
1
4
4
4
5
1
5
1
1
2
1
1
2
3
7
1
1
Data
Years
3
7
1
6
8
2
4
1
18
2
8
2
2
5
7
3
3
HLR
cm/day
0.59
4.55
2.75
0.97
1.21
7.71
2.41
0.95
0,44
0.77
20.20
5.60
4.72
11.68
7.43
1.37
1.50
TPIn
mg/L
0.100
0.106
0.125
0.212
0.252
1.678
1.909
1.996
2.983
3.015
3.940
4.150
4.540
5.167
6.540
9.140
11,075
Site
TPOut
mg/L
0.045
0.022
0.025
0.042
0.069
0.774
0.717
0.507
0.100
0.115
3,360
2.400
4.098
3.964
5.690
0.150
2.325
Average
TP Removal
%
55.0
79.2
80.0
80.2
72.6
53.9
62.4
74.6
96,6
96.2
14.7
42.2
9.7
23,3
13.0
98.4
79.0
60.7+30.2
Source: NADB 1993
4-22
-------
SECTION 4 PERFORMANCE EXPECTATIONS
The relationship between the total P loading and effluent concentration for the
TADB data set is shown in Figure 4-25. Over a range of loading from 0.5 to 4.5
kg/ha-d, total phosphorus effluent concentration increases with loading. At
lower loading rates (<0.5 kg/ha-day), however, the effluent phosphorus
concentration ranged from 0.1 to 1.5 mg/L. Mean site specific data from Central
Slough for influent and effluent total phosphorus were 4.5 and 2.2 mg/L,
respectively (Figure 4-26).
Thirteen TADB and 39 NADB sites reported dissolved phosphorous data that
were grouped into four categories based upon the analytical method used; (1)
orthophosphate (ORP), (2) soluble reactive phosphorous (SKP), (3) total
dissolved phosphorous (TDP) and unknown (UNK). At sites represented in the
TADB and NADB databases, both phosphorous loading and wetland treatment
performance varied. At the Iron Bridge, Florida, site, the mean influent and
effluent dissolved phosphorus values (ORP) were 0.35 and 0.1 mg/L,
respectively, while removal efficiency ranged from -16.3 percent to 73.6 percent.
The long-term average total dissolved phosphorous removal efficiency based
upon inlet and outlet concentration for the Houghton Lake, Michigan, system
was 96.6 percent. In Listowel, Ontario, alum addition was part of the lagoon
pretreatment process. The wetland treatment systems there also exhibited both
negative (concentration increase) and positive (concentration reduction) soluble
reactive phosphorous removal efficiencies ranging from 21.5 percent to 32.5
percent at Listowel 1 and 3, respectively.
FIGURE 4-25
Total phosphorus loading versus effluent phosphorus concentrations for the TADB FWS systems.
3.U "
S4'5'
» 4.0 -
E3.5-
£ 3.0-
f 2.5-
t 2.0-
S 1.5-
1 1.0 -
UJ 0.5 -
n n i
t
* *
* * *
» •
^ * *
^ *
0.0
1.0 2.0 3.0 4.0 5.0
Total P Loading (kg/ha«d)
4-23
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-26
Cumulative probability distribution of monthly influent and effluent total phosphorus concentrations
for Central Slough, South Carolina.
99.00
Central
Influent
Effluent
2345678
Total P (mg/L)
Temporal Phosphorus Performance
Phosphorus removal in FWS constructed wetlands follows a seasonal pattern in
most temperate climate conditions. The form of phosphorus, type and density of
the aquatic plants, phosphorus loading rate, and climate determine the amount
of phosphorus removed in FWS constructed wetlands. Aquatic plants serve as
seasonal reservoirs for phosphorus as they take up soluble reactive phosphorus
(SRP) during the growing season, however, only a finite amount of SRP can be
incorporated in the aquatic plants and plankton in the water column. In those
temperate climates where senescence of aquatic plants occur in the fall, the
majority of the biologically incorporated phosphorus is released back to the
water column upon decomposition of the participate organic phosphorus (POP)
and detrital plant material.
Figure 4-27 shows an example of the pulsing of SRP for the conditions in Arcata,
California. In this example SRP was loading at a rate of 0.15 kg/ha/day for a
year (Marsh 3). A separate control cell, Marsh 1, was fed tap water (no
phosphorus load, at the same HRT) at the beginning of the growing season (late
January and early February). At a loading rate of 0.15 kg/ha-d, 1 to 2 mg/L of
SRP was taken up by the aquatic plants and associated microbes through mid-
summer. The stored phosphorus in the plant material was being released as the
plants stopped growing and began to senesce, in late July. By early August,
effluent SRP from Marsh 3 is 1-2 mg/L higher than the influent to the marsh cell.
A cell received effluent for one year, with the same standing crop as Marsh 3,
then received tap water for one year. This cell, Marsh 1, showed a significant
contribution of SRP in the late summer as phosphorus was released from the
plant material and the detrital layer.
4-24
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-27
Phosphorus pulsing, as illustrated in a pilot cell in Arcata, California, Marsh 1 received tap water
until June 1982 (no phosphorus load), while Marsh 3 received oxidation pond effluent (Gearheart
1993).
12/1/81
3/2/82
6/2/82
9/1/82
Date
Marsh cell 1 also shows that about 0.5 mg/L of SRP is always in solution even
with no phosphorus inputs. The SRP is moving between various biological
compartments, with relatively short half-lives, as different microbial
communities dominate. The standing crop in this particular wetland was
approximately 15,000 kg/ha-yr above-ground material.
Total Phosphorus Permit Compliance
Only five FWS wetlands had TP permit limits and associated data in the NADB.
Permit limits for TP varied from 0.2 to 1.0 mg/L. The Reedy Creek, Florida,
natural wetland systems had annual average TP limits in addition to monthly
limits. Based on these limited data, it appears that FWS constructed wetlands
can comply with very stringent TP effluent limits.
Four of the five systems in the NADB had 100 percent compliance with their TP
effluent limits. The Iron Bridge, Florida, system met the most stringent limit, 0.2
mg/L, every month out of 63 recorded in the NADB with an average effluent TP
concentration of 0.09 mg/L and a maximum of 0.16 mg/L during that period.
The Orange County, Florida, hybrid wetland exceeded its monthly limit of 0.2
mg/L 5 months out of 37 months of record. The maximum TP value recorded
during this period was 0.39 mg/L.
4-25
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Fecal Goliform Performance
Database Assessment
As shown in Figure 4-28, there does not appear to be any general relationship
between the influent and effluent concentrations of fecal coliform from the
TADB systems. In general, the correlation between influent and effluent
conditions was better for specific sites (Gersberg et al. 1989). For example, a
consistent 2 to 3 log removal with a 6-day hydraulic residence time was
measured in Cell 8 in the Arcata Pilot Project. The mean influent (from an
oxidation pond) fecal coliform was 5,000 cfu/100 mL and the mean effluent
concentration was 35 cfu/100 mL. The cumulative probability distribution for
influent and effluent fecal coliform is shown in Figure 4-29. Fecal coliform
removal was also found to be correlated with TSS removal in this system. In
studies performed with MS-2 bacteriophage, virus removal appears to follow the
removal of fecal coliforms (Ives 1988).
FIGURE 4-28
Influent FC versus effluent FC for the TADB systems.
_i iu,uuu
o
o
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-29
Cumulative probability distribution of influent and effluent fecal coliform from Arcata Pilot Project
Cell 8, California (Gearheart et al. 1986).
95.00
90.00
.a
a
t
a.
70.00
50.00
« 30.00
3
3
u
10.00
5.00
1 10 100 1000 10000
Arcata Pilot Project Cell 8 FC (cfu/100ml)
Estimates of the internal production of background fecal coliforms in treatment
wetlands is provided by those systems that receive disinfected influent. For
example, the Arcata Enhancement Wetland receives chlorinated effluent, and
during the period 1990-1997, the effluent FC was less than 500 MPN /100 mL
about 80 percent of the time (Figure 4-30). A similar study on the same system
during 1995-1996 showed that the effluent FC had a mean of 40 cfu/100 mL, was
less than 300 cfu/100 mL more than 90 percent of the time, and that no sample
exceeded 500 cfu/100 mL. While some of the differences between these two
sampling results can be attributed to comparing MPN versus membrane filter
results, they also indicate the variations that can occur over time at a single site.
FIGURE 4-30
Cumulative probability distribution fecal coliform from Arcata Enhancement Wetland, California
(Gearheart 1998, unpublished data).
SSL
'£
is
.o
o
.1
3
£
o
1.00
0.50
1 10 100 1000
Arcata Enhancement Marsh Effleunt FC (MPN/100ml_)
4-27
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Temporal Fecal Coliform Performance
The considerable temporal variability in the effluent organism counts produced
by treatment wetlands and conventional treatment technologies suggests the use
of geometric averaging to determine monthly mean values from daily or weekly
measurements. Even with geometric means, individual monthly values are
frequently 10 times larger or smaller then the long-term mean for many
treatment wetlands. As indicated by the preceding discussion, organisms in the
wetland effluent did not necessarily originate with the incoming wastewater.
Fecal Coliform Permit Compliance
Only four FWS constructed wetlands had fecal coliform permit limits and
associated data in the NADB. In each case, monthly effluent permit limits were
200 colony forming units (cfu)/100 mL; only one system met this limit 100
percent of the time (Apalachicola, Florida, with only 2 months of data). Percent
compliance for the other four systems ranged from 22 to 83 percent. A maximum
value of 27,000 cfu/100 mL was reported for one month from the Benton,
Kentucky, constructed wetland, and maximum values of 2,600 to 5,800 cfu/100
mL were reported for Central, South Carolina, and Pembroke, Kentucky,
respectively. Based on these limited data, it appears that most FWS constructed
wetlands will have problems consistently meeting fecal coliform limits of 200
cfu/100 mL.
Metals
While some metals are required for plant and animal growth in trace quantities
(barium, beryllium, boron, chromium, cobalt, copper, iodine, iron, magnesium,
manganese, molybdenum, nickel, selenium, sulfur, and zinc), these same metals
may be toxic at higher concentrations (Gersberg et al. 1984, Crites et al. 1995).
Other metals have no known biological role, and may be toxic at even very low
concentrations (e.g., arsenic, cadmium, lead, mercury, and silver).
Information from FWS treatment wetlands indicates that a fraction of the
incoming metal load will be trapped and removed effectively through
sequestration in plants and sqjls (Crites et al. 1995). A summary of published
treatment wetland inlet/outlet metal concentrations from a variety of sites is
presented in Table 4-4. For many metals, the limited data indicate that
concentration reduction efficiency (EFF) and mass reduction efficiency (RED)
correlate with inflow concentration and mass loading rate (Kadlec and Knight
1996). Wetland background metal concentrations and internal profiles are not
well established.
4-28
-------
SECTION 4 PERFORMANCE EXPECTATIONS
TABLE 4-4
Metal removal data from free water surface treatment wetlands.
Metal
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
Wetland Type
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Natural
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
Natural
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
Natural
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
Natural
Concentration (ug/L)
In Out
0.45
2.41
0.58
43
0.10
160
3.4
1.57
1,510
8
7.87
20.4
6,430
205,000
241
1.7
2.2
1.28
2.0
210
7,400
<0.2
0.0112
35
7.5
6.26
17.0
0.68
0.36
0.40
2,200
36
36.85
20.6
0.20
2.47
0.05
0.6
0.05
20
1.5
1.13
60
3
3.48
6.1
2,140
6,300
766
0.4
1.63
0.25
5.5
120
3,900
0.21
0.0042
10
3.8
7.10
9.1
0.71
0.53
0.11
230
11
6.71
5.6
Mass Removal
(kg/ha-yr)
0.6
-0.1
1.25
2.4
0.1
7.9
4.5
1.0
82
11
10.4
0.21
243
29,900
-4.3
3.1
0.085
2.4
-0.03
5.1
526
0.0001
0.017
1.4
0.8
-2.0
0.14
-0.07
-0.0005
0.7
112
60
71.3
0.22
Reference
Nolte & Associates 1 998
Nolle & Associates 1998
Nolte & Associates 1 998
Hendryetal. 1979
Nolte & Associates 1 998
Hendryetal. 1979
Crites et al. 1 995
Nolte & Associates 1 998
Hendryetal. 1979
Crites et al. 1995
Nolte & Associates 1 998
CH2M Hill 1992
Hendryetal. 1979
Edwards 1993
CH2M Hill 1992
Hendryetal. 1979
Edwards 1993
Nolte & Associates 1998
CH2M Hill 1992
Hendryetal. 1979
Edwards 1 993
CH2M Hill 1992
Nolte & Associates 1 998
Hendryetal. 1979
Crites et al. 1995
Nolte & Associates 1998
CH2M Hill 1992
Nolte & Associates 1998
CH2M Hill 1992
Nolte & Associates 1 998
Hendryetal. 1979
Crites et al. 1995
Nolte & Associates 1 998
CH2M Hill 1992
4-29
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Other Performance Considerations
Wetland Background Concentrations
Wetland ecosystems typically include diverse autotrophic (primary producers
such as plants) and heterotrophic (consumers such as microbes and animals)
components. Most wetlands are more autotrophic than heterotrophic, resulting
in a net surplus of fixed carbonaceous material that is buried as peat or is
exported downstream to the next system (Mitsch and Gosselink 1993). This net
production results in an internal release of particulate and dissolved biomass to
the wetland water column, which is measured as non-zero levels of BOD, TSS,
TN, and TP. Enriched wetland ecosystems are likely to produce higher
background concentrations than oligotrophic wetlands because of the increased
biogeochemical cycling that result from the addition of nutrients and organic
carbon.
Background concentrations are not constant, but have a cycle of release that is a
function of the biogeochemical cycle rates and external (other than wastewater
inputs) factors. An example of this cycling can be seen in Figure 4-31 from the
Arcata Enhancement Wetland. Six years of weekly BOD measurements show?
that for this system the background concentration varies between 1.3 and 4.0
mg/L. The higher values (3.5 to 4.0 mg/L) occur in the fall while the lower
values occur in the summer. This variation is attributed to the accelerated
decomposition of the vegetative material and to increased bird activity in the
fall. Lower values in the summer are correlated with low decomposition rates
(low recent litter production) and decreased bird activity.
FIGURE 4-31. Variation in effluent BOD at the Arcata Enhancement Marsh.
o>
.§
O
o
ffi
3=
UJ
July-91 July-92 July-93
Ju!y-94
Date
July-95 July-96 July-97
Treatment wetland background concentration ranges can be estimated from
systems that are loaded at a low enough rate to result in an asymptotic
concentration profile along a gradient of increasing distance from the inflow
(several examples exist in the NADB). Long-term average annual outflow
4-30
-------
SECTION 4 PERFORMANCE EXPECTATIONS
constituent concentrations for this selected group of FWS treatment wetlands are
summarized in Table 4-5. Wetland systems typically have background
concentrations within the ranges listed in Table 4-6.
TABLE 4-5
Long-term average annual outflow concentrations for lightly loaded FWS wetlands in the NADB.
System
Eastern Service Area, FL
Iron Bridge, FL
Bear Bay, SC
DesPlaines, IL
Hidden Lake, FL
BOD.
1.2
2.0
1.9
--
3.0
Concentrations, mg/L
TSS NH,-N TN
3.0
2.8
2.7
5.2
13.0
0.07
0.18
0.27
0.03
0.05
1.45
0.95
2.35
1.34
0.66
TP
0.09
0.08
0.40
0.02
0.16
Source: NADB 1993
TABLE 4-6
Expected range of background concentrations for constituents of interest.
Constituent
5-day biochemical oxygen demand (6005)
TSS
Organic N/TKN
Fecal coliforms (FC)
TN
Ammonium N
Nitrate N
Total Phosphorus
Unit
Mg/L
Mg/L
Mg/L
MPNAIOOmL
Mg/L
Mg/L
Mg/L
Mg/L
Concentration
1 to 10
1 to 6
1 to 3
50 to 500
1 to 5
less than
less than
less than
Range
0.1
0.1
0.1
Natural Variability
Free water surface treatment wetlands demonstrate the same type of water
quality variability typical of other complex biological treatment processes.
While inlet concentration pulses are frequently dampened through the long
hydraulic and solids residence times of a treatment wetland, there is always
significant spatial and temporal variability in constituent concentrations. The
stochastic character of energy inputs, rainfall, and the periodicity and seasonal
fluctuation in ET contribute to the variable constituent concentrations often seen
in treatment wetland effluents as can be seen in Figure 4-31, which shows the
variability in effluent BOD concentrations over 7 years for the Arcata
4-31
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Enhancement Marsh. Such variation can and should be accounted for by
treatment wetland designers, operators, and regulators alike. If it is, FWS
treatment wetlands can be utilized successfully and confidently in a
communities overall wastewater management strategy.
4-32
-------
SECTION 5
System Planning and Design
Considerations
Planning Considerations
Like other wastewater treatment processes, FWS constructed wetlands perform
within definable limits. These limits must be identified and summarized to allow
designers to size FWS constructed wetlands that consistently achieve pollutant
reductions from a known influent to a desired effluent concentration. Regression
equations, areal loading rate methods, and simple first-order models are the
most common tools used to summarize constructed wetland performance. With
a general knowledge of performance expectations, the experienced designer can
use these tools to specify characteristics such as wetland area, water depth, cell
configuration, and plant selection to achieve desired treatment efficiency.
Consideration must also be given to specific constraints associated with the
living, autotrophic ecosystems such as those that exist in FWS constructed
wetlands. The natural processes that occur in FWS wetlands result in
background concentrations for some constituents that may be higher than the
influent concentrations of the same constituent. Knowledge of these background
concentrations is important to avoid overly optimistic expectations for
constructed wetlands performance. Additionally, a certain amount of statistical
variability is inherent in wetland effluent concentrations, some of which is due
to environmental factors (such as seasonal temperature and plant community
changes) outside the control of the wetland designer and operator. Unless
discharge permits are written to include this natural variability, the inevitable
scatter associated with wetland effluent quality must be factored into design to
avoid permit violations.
Some of the modeling tools and general considerations that are important to
wetland planning, design, and sizing are described in this section. The models
presented were developed with input output data collected from selected
wetland treatment systems, which may, or may not be representative of the
myriad of potential treatment wetland applications. Not unlike activated sludge
or other conventional wastewater treatment process design, rate constants used
in wetland models "lump" together the mechanisms and responses taking place
to improve water quality because of the present constraints in data availability
and quality control.
However, treatment wetland scientists, engineers, and practitioners are now in
the process of refining existing relationships and exploring new sizing methods
as new information is collected and made available. Expect models in the near
future that consider the non-idealities of FWS wetland flow and/or utilize
retarded rate constants to more accurately describe the principal
5-1
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
removal/transformation mechanisms taking place within constructed wetlands.
Once high quality data necessary to develop these relations are available, these
models should provide more accurate insight into predicting the performance of
FWS wetlands for a given source water, treatment volume, and/or treatment
area. For now, it is paramount that individuals or entities wishing to design and
implement FWS treatment wetlands for wastewater treatment utilize competent
professionals experienced and abreast of the technology.
Role of Wetlands in the Watershed
The first step in assessing the feasibility of FWS constructed wetland is to
identify the goals and objectives of the wetland within the watershed. Natural
wetlands are an integral part of their watershed; functioning as water storage
areas, nutrient sinks, and wildlife habitat. Once a minimum water quality is
achieved, which protects public health and addresses ecosystem concerns, FWS
can be used to provide considerable benefits beyond water quality improvement.
These additional objectives should be integrated into the feasibility and planning
process and ideally, incorporated into an overall master plan establishing
restoration goals for the entire watershed and its receiving waters.
The process used to evaluate the feasibility of FWS constructed wetlands for
water quality improvements and to function as landscape units on a watershed
requires a sequence of assessments. The process is similar to the evaluation of
conventional wastewater unit treatment processes because FWS constructed
wetlands function similarly to them in terms of their ability to convert, remove,
and store specific constituents. However, the process steps are dissimilar in that
FWS constructed wetlands fulfill other functions and values as landscape units
within a watershed. The procedure described below (Steps 1 through 12)
incorporates evaluation of the possible additional functions of FWS constructed
wetlands. The type of information required at each step and its relationship to
the decision process is depicted graphically in Figure 5-1.
Step 1 - Identify the goals and objectives of the project. In this initial step, the
role the wetiand will play in maintaining, restoring, or enhancing the beneficial
uses in the receiving system is established.
Step 2 - Characterize the wastewater(s) entering the FWS constructed wetland.
Each type of wastewater or non-point water source has its own unique physical,
chemical, and biological characteristics. A thorough characterization of the
constituents and their concentrations combined with identification of pathogen
indicators or pathogens should be conducted. This step should also include a
thorough literature review and may require laboratory and mesocosm testing.
Step 3 - Determine the discharge requirements and limitations. The discharge
constraints coupled with the constituent properties determined in Step 2 would
dictate the required effectiveness of treatment.
Step 4 - Determine the ability for wetland processes to reduce, retain, and
transform constituents. Mesocosm and bench scale treatability studies might be
required prior to proceeding to the next step. Wetland treatability studies
5-2
-------
5 AND SITE
usually require more time than most biological treatment systems because of the
time it takes to develop the aquatic macrophyte standing crop.
Step 5 - Identify the roles the wetland can fulfill in the watershed given the
constituent concentrations and treatment goals imposed upon it. Certain
wetland roles may not be appropriate due to factors such as loading variations,
types of constituents, and site location. The function and value of wetlands such
as ecological (habitat/production), hydrological, biogeochemical, and
educational can be important in determining the economic costs or benefits of
the system.
Step 6 - Evaluate the site characteristics and constraints. The planning and
design of a system is site specific. Once the type of system and the treatment
goals have been established, the soil, vegetation, and hydrologic conditions
necessary to achieve these goals are identified. The inherent characteristics of
the site should be evaluated and compared to these requirements to determine
the need for modifications and additions.
Step 7 - Determine the FWS wetland area required to achieve the treatment
objectives. For planning purposes, the methods described in this section can be
used as a preliminary estimate of the area required to achieve the treatment
objectives identified in Step 3.
Step 8 - Evaluate alternate sites. The land capacity in terms of quantity and
quality must be compared between alternate sites and technologies based upon
constraints and capabilities.
Step 9 - Estimate the total cost of the system. The life cycle cost is a function of
capital cost, and operational/maintenance cost distributed over a predetermined
time base. The computed life cycle cost can be compared with alternative
treatment systems or can be used to determine cost effectiveness and
benefit/cost analyses. The value of the additional benefits, such as real estate,
habitat, recreation, flood control, and water resource, should be included in the
development of a total cost for the system.
Step 10 - Prepare construction and wetland system development plans. Wetland
systems have several major differences from the construction of a conventional
wastewater treatment plant. The primary difference is that aquatic macrophytes
take time to develop into the requisite standing crop to support the treatment
processes. Soils that support these plants are also critical to the start-up of the
system. Preparing bid documents for the planting and maintenance of aquatic
macrophytes should include the skills of a landscape architect and/or botanist
with related experience. FWS constructed wetlands must also include flexible
hydraulic controls for operational tasks such as isolating and draining cells.
Inlet and outlet location and configuration should be considered at this step, as
this can be critical to maximize treatment efficiency.
Step 11 - Plan the system start-up. The start-up of a wetland system might
require changes in the hydrodynamics and density of vegetation. The start-up
period for a FWS constructed wetland is regionally variable and can take from 18
to 36 months because it takes time for the plants to reach operational density.
5-3
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
FIGURE5-1
Diagram of a methodology for determining the appropriateness of the use of a constructed wetland
and the factors necessary for the design of a multi-use constructed free surface wetland.
iterature 2a
i
f
T
i r
Alternative ^J-fe
Treatment 4a ~""Ap
4
i
'ublic Access
Employment 5
i
Vegetation «^
Requirement 6 a ~~"
Not Suitable
Construction
CosF~"9a
t
Constituents/P
Fate and Effects
i
i
Discharge Stt
Permit Requlr
I
¥
roperties A ^ ^
5/Mass Row "
indards 3 ^
sments
Testing, 2b
Studies
i
i
Bench snale
)t Amenable to Wetlands ^ unknown ^ ^
prop Treatment Processes 4 ~~" ~~ p
j
I
esocosms 4b
lot Studv
|
Function/value ot wetlands
In this Application 5
i
i
rreatmentsb Ecological So Hydrofogical gd
^
i
i
Biogeochem.Se
i
System Planning
Basis for Design 6
(Loading Criteria)
1
'
•"" System Hydrology 6b ^" "^ Soil Requi
1
'
'ement $c
Land Requirement 7
'
'
Site Constraints 8 "" " "Modify tta **
^
^ Cost of $
'
Life Cycle 9 costs 9b
4 ,
Costs 9c
Construct
System ' 0
^
Start-i
^
'
P 1 1
f
| FuIIscale Operations 1 2
1
r
5-4
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Discharge permits for a wetland must reflect the lag time necessary to develop
the requisite standing crop of vegetation to support treatment processes.
Step 12 - Full-scale operation requires determination of the placement and
density of aquatic plants, inlet and outlet control structures, design hydroperiod,
and design HRTs. The full-scale operation should have established background
levels of soluble BOD, COD, ammonia nitrogen, etc. Full-scale operation could
include determining procedures to store and/or drawdown the wetland system
in anticipation of discharge constraints and/or peek monthly flow conditions.
Procedures for control of vectors and nuisance mammals, vegetation
management, etc., should also be developed and ready to implement.
Step 13 - Daily monitoring of influent flow and effluent flow should be
conducted, and monthly average (weekly samples) BOD, TSS, coliform, and
other (ammonia, nitrates, etc.) pollutant concentrations tracked. Vegetation
coverage should be monitored annually along with the detrital accumulation
(TSS, plant detritus, and floating litter layer). An inspection of the hydraulic
integrity of berms, inlet and outlet works, and bottom (if required) should be
performed annually. Under certain scenarios, monitoring for mosquito larvae
and adults might be required during the mosquito-breeding season. The
activities of other potential nuisance organisms such as nutria, beavers, and
muskrats need to be monitored monthly as they can have a negative effect on
effluent quality and wetland performance, in which case they might require
management.
Additional Benefits/Habitat Considerations
Designers interested in providing habitat value in FWS treatment wetlands can
turn to the ample literature on wildlife management to find clues to optimizing
wildlife use. However, there is a significant amount of published and
unpublished literature on habitat richness and wildlife populations in FWS
treatment wetlands. Although these data have not yet been assembled and
correlated to wetland design criteria, a treatment wetland habitat database is
currently being prepared with funding from EPA's Environmental Technology
Initiative to begin to fill this information void (Knight, in preparation, 1999). In
the interim, a document published by EPA (USEPA 1988b) provides a general
description of the habitat features of 17 treatment wetlands in the United States.
USEPA has also published a book on Created and Natural Wetlands for
Controlling Nonpoint Source Pollution that has chapters on habitat
considerations (USEPA, 1993). Lastly, the habitat quality of two FWS
constructed wetlands was evaluated by the EPA's Environmental Research
Laboratory located in Corvallis, Oregon (McAllister, 1993).
Effluent Quality Considerations
Free water surface constructed wetlands produce a wide range of effluent
qualities, depending on the influent characteristics, constituent operational
loading rates, climate, and areal extent of the system. When designed and
operated properly, FWS constructed wetlands perform within a predictable
5-5
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
range of effluent values and meet their permit limitations. However, a limitation
to using a FWS constructed wetland as a wastewater treatment system is the
background concentration of constituents produced by external loading and
internal wetland processes.
Natural background concentrations of BOD, COD, turbidity, total phosphorus,
total nitrogen, and total and fecal coliform will control the effluent quality
achievable using FWS constructed wetlands. The natural variation in the
effluent from FWS wetlands is unique to each site and dependent upon the
inlet/outlet configuration, hydroperiod, and seasonal factors controlling detrital
decomposition, wildlife activity, and constituent influent loading.
The natural cycle of nutrients and the potential re-release of constituents
incorporated in the wetland biomass must be considered in the effluent permit
requirements for FWS constructed wetlands. In most cases, nutrient cycling and
release follows seasonal patterns. The seasonal cycle of decomposition release or
reduced microbiological conversion can be asynchronous with the critical water
quality requirement for the receiving waters. For example, seasonal ammonia
standards are often specified to protect receiving waters during periods of warm
temperatures and low flow conditions. These receiving water conditions can
occur during periods of high biological ammonia uptake in the wetland,
resulting in the highest rates of ammonia removal and hence discharge limits
can be attained.
In the case of coliform effluent standards, seasonal increases in coliform bacteria
(total and fecal) may result from high bird populations in the wetland. If
disinfection is required, the potential increase in wildlife populations in and near
FWS constructed wetlands needs to be taken into consideration and may require
seasonal permit exceptions. The extent and placement of open water, prime
habitat for migrating waterfowl, is an important factor in minimizing increased
coliform counts in the effluent.
Wetland Treatment System Objectives
The required effluent quality from a FWS constructed wetland is specified, in
most cases, by the state water quality control regulatory agency. Effluent
limitations are based upon (1) receiving water beneficial uses and, to some
extent, by the receiving waters hydraulic and biogeochemical assimilative
capacity and, (2) by reuse and reclamation guidelines specified for various reuse
options. While FWS constructed wetlands have been shown to be effective
wastewater treatment units, they do have treatment limitations due to factors
such as seasonal nutrient cycling, plant decomposition, and bird activity. These
limitations must be considered in both the design and the permitting of these
systems.
Another critical treatment objective consideration is the wetland effluent
discharge point. Most FWS constructed wetlands discharge to surface waters,
but a leaky FWS constructed wetland can be designed to serve as both a
treatment and disposal system (Nahar et al. 1998). Infiltration wetlands are
designed to combine the horizontal processes in the FWS constructed wetland
5-6
-------
5 FEASIBILITY CONSIDERATIONS AND SITE
with the vertical processes through the sediment and soil to meet water quality
objectives for either groundwater infiltration or surface water discharge.
Examples of infiltrating FWS constructed wetlands performance can be found in
the Hillsboro, Oregon, data and the Orange County Water District, Florida, and
Tres Rios, Arizona, wetland demonstration projects.
Permitting
A major constraint on the use of many natural marshes is the fact that they are
often considered part of the receiving water by regulatory agencies (Reed et al.
1995). Consequently, wastewater discharged to a natural wetland has to meet
discharge standards prior to application. In Arcata, this obstacle was avoided by
taking advantage of the "enhancement" clause of California State law regarding
water quality, in which wastewater application can be allowed if enhancement
of the existing wetlands can be shown. The distinction between natural and
constructed FWS wetlands is not always clear, and the barriers to using natural
FWS wetlands for treatment may also be applied to constructed FWS wetlands.
Historically, the use of natural wetlands in wastewater management in the
Southeast occurred because of convenience or the lack of other reasonable
alternatives. Only in the past decade have wastewater management systems
incorporated design elements to optimize the wastewater renovation capabilities
of wetlands. The use of natural wetlands for wastewater management may not
be appropriate in many cases. Most situations will require site-specific analyses
to determine site feasibility and acceptability based on existing natural wetland
type, size, condition and sensitivity.
In general terms, the use of natural wetlands should be avoided when:
• The wetlands being considered are pristine wetlands and representative
of unique wetland types;
» Projected impact to the wetlands would result in changes that would
threaten the viability of the system; and/or
• Conflicts with other uses could not be mitigated adequately such as
adjacent land use activity, availability, and cost of land.
Most natural wetlands are designated as "Waters of the United States." Such
wetlands are either adjacent to other Waters of the U.S., or upon use,
degradation, or destruction could affect interstate or foreign commerce, and as
such, are afforded protection under the programs of the Clean Water Act. In
Addition, other wetland protection programs must be considered when
evaluating the use of natural wetlands.
Under the Clean Water Act, the four programs that can directly or indirectly
affect wetland wastewater management decisions are:
• Construction Grants (Section 201)
• Water Quality Standards (Section 303)
5-7
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
• National Pollutant Discharge Elimination System (NPDES) Permits
(Section 402)
• Discharge of Dredge/Fill Permits (Section 404).
For each program area, there are currently existing specific program regulations,
guidance and procedures. However the use of wetlands for wastewater
management has not been addressed specifically by any program and clear
guidelines do not exist. Minimum criteria relating to waters of the U.S. that can
be applied to wetland effluent discharges require that:
• Water quality standards be maintained;
• A minimum of secondary treatment be attained prior to discharging from
municipal treatment facilities to natural wetlands considered waters of
the U.S.;
* An NPDES permit for each discharger or discharge point; and
• A 404 Permit for the discharge of dredge and fill material into
jurisdictional wetlands.
Regulations for the U.S. Environmental Protection Agency's (USEPA) three
major wastewater management programs (Water Quality Standards, NPDES
Permits, and Construction Grants) are designed for facilities discharging to
lakes, streams, rivers, estuaries, or other free-flowing surface waters. Wetlands
are different from these aquatic systems due to their role as a transition between
fully terrestrial and fully aquatic systems. As such, wetlands are often
hydraulically slow moving systems, as opposed to the free-flowing nature of
most streams and rivers. Additionally, the functions and use of wetlands cover a
broader range of ecological, water quality, and hydrological values. Because the
regulatory guidelines and programs developed under the Clean Water Act's
wastewater management programs did not acknowledge or address specific
wetland considerations, they are usually not directly applicable to wetland
wastewater management systems.
Although wetlands that are Waters of the U.S. cannot be classified for "waste
treatment," they can be used in wastewater management as long as established
uses are protected. Many wetland functions and values, (e.g., storm buffering,
and water storage), however, are not covered by existing use classifications.
Additional qualitative or quantitative criteria addressing wetland characteristics
(e.g., hydroperiod, water depth, and seasonal influences) may be necessary and
appropriate to protect wetland uses. Entities that choose to build constructed
treatment wetlands for helping to meet advanced treatment requirements (e.g.,
the Tres Rios Project for meeting "excursions" by the 91st Ave. plant) that are
also designed to provide high value wetland habitat for wildlife and public use
may find themselves facing CWA §404 issues if they locate their system in
existing wetlands or waters of the U.S. On the other hand, if they seek formal
recognition of the habitat values for potential eligibility and use as wetland
mitigation areas, they may also create long-term responsibilities to maintain
these areas. Opportunities do exist, especially in the arid West, for such projects
5-8
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
to involve the use of pretreated effluents to help restore degraded or former
wetland systems. Further guidance on this and other policy and permitting
issues associated with constructed wetlands can be found in "Guiding Principles
for Constructed Treatment Wetlands: Providing for Water Quality and Wildlife
Habitat, prepared by the Interagency Workgroup on Constructed Wetlands"
(this document is available on-line at
www.epa.gov/owow/wetland/constructed/guide.html").
Section 402 of the Clean Water Act authorized EPA and delegated to the states,
administration of the NPDES Permit Program. This program requires a permit
to discharge pollutants from any point source into waters of the U.S. Therefore,
the discharge to wetlands considered as Waters of the U.S., or from treatment
wetlands into a Waters of the U.S. requires the issuance of an NPDES permit.
Important elements of the permitting process include the application process,
establishing effluent limits, permit conditions and requirements, permit
issuance, and compliance monitoring.
Alternatives which accompany the application for the NPDES permit for
wetland wastewater systems include the use of a tiered approach for information
requests and monitoring requirements based primarily on wetland types and
hydraulic loadings. The use of performance criteria as a permit requirement to
monitor wetland and downstream water quality is also suggested.
An important step in establishing effluent limits is determining whether the
stream segment (or in this case the wetland) to which a discharge is proposed is
classified as effluent limited or water quality limited as defined by EPA (1985).
A stream segment that is effluent limited requires best available technology or
secondary treatment. A stream segment that is water quality limited requires
greater than secondary treatment. The task of establishing effluent limits in
water quality limited situations is not straightforward. The use of water quality
models may not adequately predict wetland responses to wastewater discharges
and the use of an on-site wetland assessment will likely be necessary. The
qualitative results of an on-site assessment then need to be related to
quantitative or qualitative effluent limits.
Public Access
The ancillary benefit of wetland and riparian habitat associated with free surface
constructed wetlands has given some communities the opportunity to allow total
or limited public access to the wetland treatment facility after sufficient water
quality improvement has been achieved. These ancillary (or value-added)
benefits have allowed some communities to extend the public and
environmental services of the wetland to other uses. Ancillary benefits include
but are not limited to passive recreation, environmental education, green belts,
mitigation wetlands, etc. Various states have their own guidelines and
regulations concerning public access to wastewater treatment facilities. The
addition of a passive recreation and/or an environmental education facility has,
in many cases greatly enhanced the local and regional visibility of the project.
This visibility and usage has in most cases resulted in community support
5-9
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
related to the wetland treatment concept as well as for the community
environmental service efforts, i.e., watershed planning, stormwater
management, riparian/wetland corridors, etc.
Requirements placed on public access vary considerably from site to site even
within the same state. In California for example, Arcata allows 24 hour, 365-day
access to the Arcata Marsh and Wildlife Sanctuary, while Hayward restricts
public access to environmental education visits, by appointment only. Much of
these differences are due to the demographic and geographic settings of the two
sites. Hayward is a highly urbanized area with no direct community
management. Arcata, on the other hand, is a mostly rural area where intensive
volunteer involvement and management efforts exist.
Public access, which does not disturb wildlife, is generally considered a
favorable component of a project. Careful planning and design of a system can
minimize human disturbances while maximizing the habitat value.
Hydrological Considerations
FWS constructed wetlands have been utilized successfully in a wide range of
hydrologic, climatic/ and geographic settings, establishing the general utility of
FWS constructed wetland systems over an array of locations and conditions.
Although these systems are robust enough to operate under a variety of
scenarios, consideration must be given to the effects of local conditions on the
performance. When possible, these local condition effects can be mitigated by
design constraints.
Precipitation and Evapotranspiration
As described in Section 3, precipitation and evapotranspiration affect the
performance of FWS constructed wetlands by altering the concentration of
constituents in the wetland and by changing the volume of water transported
through the wetland. In areas of high rainfall or during periods of high rainfall,
the precipitation accumulated in the wetland can dilute effluent concentrations
and reduce the hydraulic residence time (HRT). High evapotranspiration rates
act in the opposite manner, concentrating the water quality constituents and
increasing the HRT.
In arid regions of the United States, monthly net loss due to evapotranspiration
can be as much as 25-400 mm. At typical hydraulic loading rates of 50 to 80
mm/d, the loss of water can concentrate the dissolved constituents 10 to 25
percent. At the same time, the nominal hydraulic residence time would increase
proportionally. The opposite effect is observed in the wetter regions of the
United States.
In regions with long dry periods, dramatic increases in coliform bacteria, total
suspended solids, ammonia, and turbidity can occur at the start of the wet
season. These increases in water quality constituents are due to bird fecal
5-10
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
material and other partieulates washed off the plants and into the water column
at the beginning of the rainy season.
Groundwater
Free water surface constructed wetlands are normally designed to be isolated
from underlying aquifers. For site design, the elevation of the seasonal high
groundwater table and direction of predominant flow should be determined to
ascertain potential problems with interception or berm failure. In the case of
unlined FWS constructed wetlands designed to discharge through infiltration,
groundwater monitoring will be necessary to measure constituent
concentrations and hydraulic effects of the discharge.
Ice and Snow
In areas of significant snow cover or thick ice formation, free board is made
available in FWS constructed wetland design and operation to allow the ice
cover to serve as insulation over the water column. Ice formation requires an
increase of 300 to 500 mm in the operating depth to maintain the design water
column depth. In some cases, better effluent quality is obtained in the colder
months due to the lack of external factor effects (wind, wildlife, etc.) and
seasonal low contributions from internal sources such as plant litter and solids
decomposition.
Engineering Considerations
Pre-Treatment Requirements
FWS constructed wetlands have pre-treatment requirements similar to other
biological wastewater treatment processes. In Europe and the United States, this
minimum appears to be that equivalent to primary and/or septic tank effluent.
Floatable solids and large settleable solids should be removed from the influent
wastewater. Excessive levels of oil and grease should also be avoided. Specific
constituents or constituent loadings that may upset biological processes should
receive pre-treatment. The wastewater delivery system should be designed to
distribute influent evenly across the wetland cross-section to maximize the
treatment volume available to remove settleable and suspended solids.
Also important to a FWS constructed wetlands are the incoming metal
concentrations. While a FWS constructed wetland can remove and immobilize
many heavy metals, if the system is designed for habitat enhancement, the
potential for metals accumulation in the biota exists. In cases of high metal
concentrations in the wastewater, a source reduction program and an industrial
waste pretreatment ordinance may be more appropriate than a multi-use FWS
treatment wetland.
5-11
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Soils, Slope, and Subsurface Geology
The principal soils considerations in siting and implementing a FWS constructed
wetland are the infiltration capacity of the soil and its suitability for berm
construction. In most cases, FWS constructed wetlands are required to meet
stringent infiltration restrictions. Specifications for infiltration losses from
wastewater ponds and wetlands range from 1x10"' to 7 x 10"6 mm/s depending
on the state regulations for construction and groundwater protection. Systems
designed to incorporate infiltration as part of the treatment and discharge
process of the plant are an exception. In these cases, the underlying soil must
have infiltration rates compatible with the design rates of discharge. In both
cases, the native soils may need amendment or restructuring.
An additional soil consideration for FWS constructed wetlands is its suitability
to establish and grow wetland plants. Aquatic macrophytes generally reproduce
asexually by tuber runners. Soils with high humic and sand components are
easier for the tubers and runners to migrate through, resulting in rapid plant
colonization and growth.
FWS constructed wetlands can be built on sites with a wide range of topographic
relief. Construction costs are lower for flat sites as highly sloped sites require
more grading and berm construction. With proper design, however, high slope
sites can possibly reduce pumping costs by taking advantage of the existing
hydraulic gradients.
Percolation and Use of Liners
If the native soil does not have sufficiently low infiltration rates, amendment
with day or soil binders can be used. Another option for minimizing infiltration
is installation of a geosynthetic membrane beneath the system (Kays 1986). Both
of these requirements can add significantly to the construction cost of a FWS
constructed wetland.
Inlet/Outlet Types and Placement
The hydraulic response of a FWS is dependent on several factors: vegetation
type, amount and location, geometry of the system (especially as it might relate
to dominant wind directions and velocities), and the type and location of the
inlet and outlet works.
The inlet works should ensure a uniform distribution of the influent normal to
the direction of flow. This can be accomplished in several ways. One technique
is a manifold which extends across the inlet zone with adjustable ports located
every meter or so. Another technique is to have several large inlet weirs
(controls that allow to shut off the flow) which discharges into a mixing volume,
which extends across the entire entrance of the wetland. This area tends to be
deeper to minimize emergent plants and to ensure an even distribution of the
influent through the aquatic plants in the wetland.
5-12
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
The outlet works can also be of several types. The outlet works serve both as
water level controls and as collection points for the effluent. Not much work has
been done comparing the various types of weir structures and locations as they
relate to effluent quality. Geometry of the wetland cell has a lot to do with the
number and type of outlet weir structures. In general, weir structures are placed
every 8 to 25 m along the effluent collection zone located at downstream point(s)
in a FWS constructed wetland. Similar to the influent collection/distribution
zone, some systems have effluent collection volumes that direct flow to a weir
collection/control structure. This type of system can produce variable TSS and
coliform as both algal population and wildlife are attracted to this deeper clear
water volume. Best successes have been observed where the aquatic plant
communities are more, or less, contiguous with the effluent zone/control
structure. Extremely high weir overflow rates in this type of system suggest that
increasing total weir length might assist in improving effluent quality.
Wildlife/Habitat Consideration
A FWS constructed wetland utilized for treating municipal wastewater can also
function as wildlife habitat, and in some cases where water quality permits it,
constructed wetlands are being designed with wildlife habitat creation as a
secondary or primary goal. This approach is similar to the role in which
oxidation ponds are used by waterfowl and wildlife. Constructed FWS wetlands
can provide incidental support of wildlife, or it can be enhanced by considering
factors that encourage and support a wide range of wildlife communities. In the
case of FWS constructed wetlands, the amount of open-water area and the types
of submergent and floating macrophytes are positive habitat factors. The
proportion and location of open water areas can also affect wetland effluent
water quality. Based on pilot project work performed and subsequently used to
design enhancement wetlands in Arcata, California (1986), it was shown that
having 25 to 70 percent of the water surface dominated by submergent and
floating macrophytes allowed optimal water quality and habitat enhancement
objectives to be achieved (EPA 199b).
Another important design consideration for wildlife habitat is the inclusion of
islands with low-sloped sides. Waterfowl and shorebirds can use the islands for
feeding, nesting, and rest areas. Slopes of 1:4 to 1:10 around the island will
encourage shallow zoned aquatic plants, while allowing easy access for aquatic
fowl. Islands have been used effectively in many wetlands to support resident
and migrating bird populations.
5-13
-------
SECTION 8 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Environmental Impact
The following planning level considerations for the possible use of FWS
constructed wetlands are important in communicating advantages and
disadvantages of these wastewater treatment systems to clients, community
members, and regulatory officials.
Land Use
The first major consideration for the use of FWS constructed wetlands is the land
requirement and issues associated with general plans and zoning restrictions.
Depending on the size of the community and the land uses adjacent to it, these
could represent constraints or time-consuming requirements. Several possible
strategies could be employed to expedite these issues. One successful strategy is
to highlight the major advantages of FWS constructed wetlands; the multiple
land use activities that can be assigned to their footprint. Overlays of land use
activities, such as parks, passive recreation, wetland habitat, environmental
education, green belts, possible wetland mitigation, open space, and viewshed
corridors increase the public value of FWS constructed treatment system. These
beneficial impacts can assist in mitigating the cost of the land for wastewater
treatment.
Insect Vectors
Potential problems with insect vectors, particularly mosquitoes, are another
major concern. Wetlands are prime habitats for mosquitoes and black flies, and
are habitat for most of their major predators. Proximity of the wetland to houses
and areas of intense use can become a siting constraint. For the most part,
mosquitoes do not fly more than 400 m from their breeding area. Certain
species, however, under the influences of wind direction and speed can disperse
mosquitoes and black flies much farther. Regardless of location, mosquitoes will
be present at some time of the year in any FWS constructed wetland. Serious
consideration should always be given to implementing integrated pest
management to control mosquito populations. Integrated pest management
requires measures such as maintenance of an ecosystem that attracts and
sustains viable populations of natural adult mosquito predators (dragonflies and
damselflies, bats, swallows, frogs), and larval predators
(carnivorous/omnivorous fish, and aquatic macroinvertebrates). Consideration
should also be given to management of mosquito larval populations through the
use of mosquito-specific larvicides such as those derived from the bacterium
Bacillus thuringiensis or from a strain of Bacillus sphaericus. Chemical adulticide
plans should also be formulated in case of mosquito generated public heath
threats.
5-14
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Odors
A FWS constructed wetland will have a seasonal odor associated with the
normal decomposition of plant material and incoming settled solids. These
odors will be more or less concentrated around the wetland as a function of
micrometeorological factors such as wind speed, humidity, and lapse rate dose
to the surface. The odors associated with a FWS constructed wetland are not the
same type or magnitude as the odors associated with a conventional wastewater
treatment plant. Hydrogen sulfide is the predominant odor mixed with gaseous
by-products of actinomycetes. The large area over which the odor is released
tends to keep the concentration low, easily diffused, and dispersed. Odors can
also develop if the influent wastewater is not properly introduced into the
wetland.
Wildlife and Ecological Attractive Nuisances
While one of the major potential objectives of a FWS constructed wetland is to
provide habitat value, some concerns are often voiced about the potential for
attracting endangered species. At present, there is mixed information related to
this issue. There is no state or federal law that exempts constructed wetlands
from Endangered Species Act issues. There are examples where wastewater
discharges support the habitat for endangered or listed species (e.g., pupfish in
China Lake, California). For the most part, it is considered a net gain if a FWS
constructed wetland becomes habitat for an endangered species. Oxidation
ponds function similarly in many arid regions.
All FWS constructed wetlands provide habitat whether intended or not. FWS
constructed wetlands that incorporate habitat features by design can attract large
numbers of wildlife. One major potential problem is attracting too large a
population of migrating birds. If the wetland supports large bird populations
and water quality conditions are conducive to pathogen survival, then potential
disease problems can develop (vibrio, clostridium).
The disease potential is highlighted at several wetlands in the San Francisco Bay
area. For example, Hayward Marsh is the only source of freshwater on the Bay
perimeter and as such, attracts large bird populations. It has limited vegetation
cover that results in large open areas for resting, watering, and feeding. This
provides large numbers of birds with opportunity to share common food sources
and to come into close contact, effectively transmitting disease throughout the
population.
The potential for introduction and spread of disease in migratory bird
populations can be minimized. This is achieved by using a diverse assemblage
of aquatic and riparian habitats, and by having the flexibility to manipulate the
hydroperiod and flow rate into and out of the wetland.
Another major problem associated with constructed wetlands is the intentional
release of domestic aquatic fowl and other domestic animals such rabbits, cats,
and dogs. In the case of domestic ducks, their interbreeding with wild aquatic
fowl presents a major wildlife problem. Feral cats may also be a significant
5-15
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
problem, as they feed on birds at the wetland. The issue of domestic species
management requires advance plans be developed and implemented before
problems escalate.
FWS Wetlands and Bird Strike Issues
Because of the great potential FWS wetlands have for attracting wildlife,
specifically avian species, there exists a potential for conflict between animals
and aircraft. In most cases, if siting a FWS constructed wetland within 5 miles
from an airport, the habitat features must be in compliance with criteria set forth
in 14 CFR Part 139. In brief, guidelines exist which govern the placement of
habitat features within 10,000 feet and 5 miles from an airport that require
developing a plan addressing wildlife hazards.
Wetland Sizing
As FWS constructed wetlands became recognized as a viable wastewater
treatment process, a need arose for FWS design models. These models aid
engineers in the process of FWS wetland design and performance assessment
(e.g. wetland area requirements and effluent quality predictions).
Approaches to Sizing
The current trend in wetland design modeling is the development of simple
mass balance or input/output models. These simplified models do not explicitly
account for the many complex reactions that occur in a wetland, either in the
water column or at interfaces such as the water/sediment interface. Instead, all
reactions are lumped into one, overall reaction rate that can be estimated from
FWS wetland input/output data. At this stage of wetland model development,
more complex and theoretical wetland models—in which the kinetics of known
wetland processes are described explicitly—are not possible, due to limitations
in the existing wetlands data.
To date, a number of wetland design methods have been proposed for predicting
constituent removals in FWS wetlands. The methods include fundamentally
equivalent design relationships and equations presented by Reed et al., (1995),
Kadlec and Knight (1996), Crites and Tchobanoglous (1998). The design
relationships and methods have been used to predict the reactions (degradation
or generation) of BOD, TSS, TN, NH4, NO3, TP, and coliform. An estimate of the
wetland surface area can also be made by rearranging the relationships to solve
for wetland area given constituent removal goals.
Design relationships are summarized in Table 5-1 along with new relationships
proposed by Gearheart et al. (1998). The reader should keep in mind that none
of the relationships presented in Table 5-1 are developed in this Technology
Assessment to the extent needed to design a successful FWS treatment wetland.
The reader is further encouraged to seek additional design information/insight
5-16
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
TABLE 5-1
Equations used to compute the performance of FWS constructed wetlands
Formula
Type
Definition of Terms
Reed etai. (1995)
Volumetric
Kadlec and Knight (1996)
(Ce-C*) f k,
—2 = exp| _
CC0-C')
Retardation Model (discussed in Crites
and Tchobanoglous, 1998, Gearheart,
1999 [in preparation])
Areal
Volumetric
BOD only
Ce=C0e
LM)J
'D
Sequential Model (Gearheart, 1999 [in
preparation])
Volumetric
Two-rates
BOD only
a = Delaying constant, temperature-
dependant
Co = background BOD concentration
Ce
Co
C*
contributed by decaying plants (g/m)
= effluent concentration (g/m3)
= influent concentration (g/m3)
= BOD concentration due to
solubization of TSS and residual total
BOD (1 to 65 days)
= background concentration (g/m3)
curve-fitting parameter
= temperature corrected first-order
areal reaction rate constant (m/yr)
= temperature-dependent first-order
rate volumetric reaction rate constant
Kvt = volumetric based solids/particulate
BOD removal rate
Kva = volumetric based dissolved BOD
removal rate - temperature-
dependent
q = nominal hydraulic loading rate (m/yr)
t = theoretical hydraulic detention time
from other sources such as those provided in Tables 2-2 through 2-4, or from
competent professionals currently practicing in the field.
Regression equations have also been used to summarize system performance for
a variety of constituents and physical parameters. General loading relationships
have been used to predict removals for TSS, BOD, nitrogen, phosphorus and
coliform.
To utilize one of the FWS design relationships or methods, it will be necessary to
estimate or assume various parameters. Generally, the influent concentration,
the expected or desired effluent concentration, and the flow rates are known
from project goals and/or previous work. However, the remaining parameters
5-17
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
need to be estimated from pilot project studies or assumed from literature
values.
Assessment of Predictive Equations
In most of the existing FWS wetland design relationships, it is assumed that the
hydraulics of FWS wetlands can be approximated by a plug flow reactor (PFR)
model and the reactions of constituents are described by first order reaction
kinetics. The use of the PFR to approximate the wetland hydraulics appears to
be generally accepted by the wetland modeling community. However, there is
ongoing debate over the appropriate form of the first order reaction rate
constant.
The general relationship, assuming steady-state plug flow hydraulics and first
order constituent removal, is:
^ = -kappC (5-1)
where: C = pollutant concentration (m/L3),
t = mean hydraulic detention time (t), and
kapp = apparent first-order rate constant (t"1).
This differential equation has the exact solution:
C— f i»vn "W1 (^JJ\
t ~™ V-'QWA.J/ \^J £.}
where: C0 = initial pollutant concentration at t = 0 (m/L3).
The apparent first order reaction rate constant (kapp) can be a function of
temperature so values are generally reported at 20° C. The kapp value can be
adjusted to the desired temperature using a modified form of the van't Hoff-
Arrhenius relationship:
* (5-3)
where: k,. = apparent first order reaction rate constant at T degrees C [t"1],
kg, = apparent first order reaction rate constant at 20° C [t"1],
8 = empirical temperature coefficient, and
T = temperature at which k,. is adjusted.
The FWS wetland predictive equations presented in Table 5-1 are derived from
the general PFR model (Equations 5-1 to 5-3). However, each of the models uses
5-18
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
different concepts and approaches in defining the general PFR parameters
(i.e., k and t).
The Reed et al. (1995) and Crites and Tchobanoglous (1998) relationships
incorporate the adjusted nominal hydraulic detention time (t) through the
wetland, and an apparent first order volumetric reaction rate constant. To utilize
these equations, the depth, porosity and average flow through the wetland is
required. The background pollutant concentration (C*) is not directly
incorporated into these equations, but is included as a boundary condition
(implied lower limit on effluent concentration) of the model.
The relationship proposed by Kadlec and Knight (1996) is based on the nominal
hydraulic loading rate (q) to the wetland, and a temperature dependent first-
order areal reaction rate constant. For some constituents, such as BOD, Kadlec
and Knight report that the areal reaction rate constant is not temperature
dependent. In this model the depth, porosity and water losses and gains
through the wetland are not required, but lumped into the first order areal
reaction rate constant. The background pollutant concentration, C*, is directly
incorporated into the model equation.
Areal Loading Rate Method
In the areal loading rate method, a maximum loading rate per unit area for a
given constituent is specified. The use of loading rates is common in the design
of oxidation ponds and can be used to give planning level surface area estimates
for FWS constructed wetlands from projected pollutant mass loads. Areal
loading rates are also used to check a FWS wetland designed using one of the
above mentioned design models to ensure that the wetland is not overloaded. A
range of typical influent concentrations, target effluent concentrations, and
constituent areal loading rates for FWS wetlands are listed in Table 5-2. The
suggested values given in Table 5-2 are based on data from the FWS wetland
systems listed in Table 2-5. These rates can also be used to give a preliminary
estimate of the FWS wetland surface area required for a given constituent
loading, and to check wetland areas determined from equations in Table 5-1.
A typical areal loading design curve based on the long-term average
performance of systems listed in Table 2-5 is shown in Figure 5-2. By knowing
the areal loading rate, constituent effluent concentrations can be estimated from
or compared to the long-term average performance data of full-scale operating
systems.
5-19
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
TABLE 5-2
Range of areal loading rates for FWS constructed wetlands derived using data from FWS systems
listed in Table 2-5 (Hydraulic Loading Rate for these systems ranged from 10-100 mm/day).
Constituent
BOD
TSS
TN
NH4
N03
TP
Typical Influent
Concentration (mg/L)
5
5
2
2
2
1
-60
-60
-20
-20
-10
-10
Target Effluent
Concentration (mg/L)
O "™
5-
1 -
1 -
0.5
0.5
20
20
10
10
-3
-3
Loading Rates
(kg/had)
10-
10-
2-
2-
1 -
1 -
50
60
10
10
5
5
FIGURE 5-2
Annual average areal BOD loading rate vs. annual average effluent BOD concentration for TADB
systems.
100
BOD Loading (kg/ha*d)
150
200
Design Approach to Sizing
The approach to design of free water surface constructed wetlands should
consider a wide range of local factors as well as general operational experience
gathered on these systems. Design equations used to size FWS constructed
wetlands summarized in Table 5-1 require the estimation of one or two
parameters. Based upon observed data summarized in Chapters 3 and 4, "best-
fit" parameter values vary greatly from site to site. The use of statistically
derived, national parameters suffer from the inadequacies present in the data
5-20
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
used to derive them as previously discussed. The equation parameters
incorporate many factors and should be applied carefully when the setting and
condition are different than those used to generate them. As discussed in
Chapter 4, most of the systems in the database were underloaded and, therefore,
are over-designed with respect to certain constituents in terms of areal
requirement. None of the design formulas presented in Table 5-1 and used to
determine wetland surface area requirements include the effect of inlet/outlet
type and location, and vegetation type and distribution, which are potential
determinants in wetland treatment effectiveness.
The first-order decay constant in all of the design equations is an apparent "k"
value since it incorporates many factors including hydrological factors,
temperature, solubilization factors, and removal/transformation processes.
Over-designed wetlands mask the effect these factors have because the removal
or transformation of a constituent is complete prior to the outlet. Since the
performance of most of the wetlands in the database has been estimated from
inlet and outlet samples this fact is reflected in the state of technology. As more
experience is gained from multiple celled and/or systems in which samples are
collected at intermediate points, a more useful database for the estimation of
removal rate values can be developed. At present, the approach to design could
include the use of the equations given in Table 5-1 with the resulting area
checked against the empirical areal loading rates given in Table 5-2. This design
approach is detailed in the EPA Wetland Design Manual (EPA, 1999).
Though plug flow is assumed for the purposes of FWS constructed wetland
design, the actual wetland flow hydraulics will not follow an ideal model. The
degree of deviation from plug flow of an existing FWS constructed wetland can
be determined by tracer tests. One of the important results of a tracer test is the
determination of the tracer detention time, defined as the centroid of the
response curve. Tracer detention time is equal to the active water volume
divided by the volumetric flow rate, and thus represents a direct measure of
actual detention time. Comparison of the theoretical to the actual detention time
is an important tool for evaluating the performance of existing FWS constructed
wetlands.
The plug flow assumption is conservative in design if the degree of non-ideality
(represented by the ratio of the theoretical plug flow to actual hydraulic
detention time) in the designed system is less than that in the wetlands from
which model parameters were determined (Hovorka, 1961; Kadlec and Knight,
1996). Because actual detention times are always less than the theoretical (plug
flow) detention time, apparent removal rate constant estimates based on the
plug flow assumption will be lower than the actual removal rate.
Using an apparent removal rate constant from one system for a different
wetland with a different degree of actual to theoretical detention time can lead to
serious over or under-design. For example, using tracer data developed at
Treatment Marsh 1 (TM1) in Arcata, the observed hydraulic detention time was
84 hours. The theoretical detention time for this marsh is about 200 hours, nearly
250 percent longer. If an apparent removal rate constant computed based on the
theoretical detention time is used for sizing a new system where the ratio of
5-21
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
theoretical to actual hydraulic detention time is higher (say 3.5 ;l), the new
system will not meet performance expectations due to the relatively shorter
actual detention time. The degree of non-ideality should be similar in wetlands
with similar geometry, vegetation patterns, and hydraulic loadings. Note, the
treatment wetland literature typically provides only apparent plug flow k
values.
5-22
-------
SECTION 6
Lessons Learned and
Recommendations
A successful FWS treatment wetlands project requires that a number of other
considerations be addressed which are just as important as the wetland process
and design issues (e.g. water conveyance and wetland area) discussed earlier.
Issues and considerations that are important in the implementation process for a
FWS treatment wetland are described in this section. Items discussed include
the need for more high quality wetland performance data and updating of
wetland databases, potential nuisance conditions, open water/emergent
vegetation areas, major components of wetland civil design and construction,
issues surrounding wildlife enhancement wetlands, multiple benefits and public
access, and general operation and maintenance considerations.
Information Management
New information from free water surface treatment wetlands is accumulating at
a rapid pace. Between existing projects with ongoing monitoring programs and
new projects that incorporate updated design features, the amount of useful
information that could be applied to resolving technology issues is greater than
can be accumulated and analyzed by individual wetland designers. Coordinated
state or federal activities have not proven to be an effective method for keeping
up with this accelerating information supply either. The most useful
information has been generated by well documented moderately to highly
loaded systems with cell-by-cell flow, depth, and constituent data.
Databases can provide a convenient method of accumulating and analyzing
large amounts of treatment wetland design and operational data. Expansion,
maintenance, and analysis of a FWS constructed wetland database is presented
as a priority for future technology assessments. Research-level pilot studies
provide the best method for testing the effectiveness of new treatment wetland
design criteria. However, many pilot studies have failed to address new issues,
and most have had such short operational periods that drawing general
conclusions about the performance of a mature wetland from their data is
difficult. New treatment wetland research efforts should consider focusing on
some of the key technology issues that have been identified in this report.
Database Maintenance and Analysis
The initial NADB project began in 1991 and ended before completion in
September 1993. This project captured a significant fraction of the wetland
design and performance information available at that time, however;
approximately 100 additional treatment wetlands in the U.S. and Canada were
6-1
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
tentatively identified during that effort. It is now likely that up to 200 additional
North American treatment wetland systems are not currently described in the
database. In addition, the coverage and quality of data for those systems
included in the NADB are suspect and can be incomplete in terms of using the
data to evaluate system performance.
The depth of existing information displayed in portions of this technology
assessment is testament to the potential value of an extensive design and
operational performance database for wetland treatment systems. Intra-system
data analysis allows determination of the effects of design variables on
performance for major constituents of interest. Inter-system data comparisons
allow the designer the opportunity to detect regional differences and differences
due to variable water sources. The "data cloud" figures presented in this report
reassure wetland practitioners that they can expect certain reasonable
performance from treatment wetlands.
In spite of the limitations of the NADB, it can be used for a variety of purposes.
One is to provide an inventory of how many treatment wetlands are "out there"
and how they were built. This knowledge provides an understanding of how
important this technology has become and to assess how rapidly it is growing,
but does not require detailed operational data. A second goal, more in line with
the purpose of a technology assessment, would be to assess accurately, wetland
performance under a variety of design conditions. The NADB, as it is presently
formulated and implemented, falls short of meeting this goal, in that insufficient
information exists to optimize design of free water surface treatment wetlands.
Since all of the presently available design models are approximations of system
performance based upon the presently available limited data, the variability in
empirical design relationships cannot be reduced until sufficient wetland data
are available to document the effect of all design variables. More complex, multi-
parameter design models can only be supported by analysis of detailed
information from a number of long-term, research-oriented treatment wetlands.
Lastly, ensuring that complete flow measurements are included for all systems is
critical to the utility of the database in evaluating performance.
Additional funding should be sought for reformulating, updating, balancing,
and editing the existing NADB. Most of the systems evaluated in the NADB are
lightly loaded systems. Some of these systems have influent BOD and TSS
values dose to background values, resulting in periods of net negative pollutant
removal. Efforts should be made to identify sites with higher loading rates to
provide a more balanced view of the potential of the technology to treat
wastewater. An initial effort could be completed over a 2-year period. The
resulting updated NADB should be analyzed thoroughly and the results widely
published. Practitioners in this field should be encouraged to maintain their own
project data in an electronic form compatible with the reformulated NADB to
allow rapid entry of new information.
6-2
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Planning
Multiple Benefits and Public Access
The general public rather than individual landowners primarily receive benefits
produced by wetland areas. After an appropriate level of pretreatment,
wastewater introduced into a constructed wetland can sustain the wetland
ecology and provide for multiple benefits, including public access (education,
birdwatching, walking, jogging, and picnicking), two of the strongest
endorsements for the use of this treatment process. The advantages of a multiple
benefit investment in landuse can be a positive aspect of any FWS constructed
wetlands project. These landuse types could include (1) parkland, (2) wildlife
habitat, (3) environmental education, (4) open space, (5) greenways, (6) water
reclamation storage, and (7) landuse set aside for future public use and
treatment. These overlays of uses increase the societal value of the land
investment made for treating wastewater. Public access is essential for
communication and maintaining the multiple benefits of a FWS constructed
wetlands project.
As a wastewater treatment system, FWS constructed wetlands have introduced a
unique management opportunity. If the wetland system has multiple benefits,
such as education, recreation and research, a public access policy needs to be
developed specifying public use guidelines. Public access to a wastewater
treatment facility is normally restricted due to the potential risk associated with
pathogens present in the wastewater. Many states have specific regulatory
constraints concerning public access to wastewater treatment facilities.
Clearly, public contact with raw or untreated wastewater is a potential human
health threat that must be eliminated from both conventional and wetland
treatment systems. If appropriate pre-treatment is provided a responsible
public-use / access policy can be developed which allows for many of the
potential ancillary benefits of a wetland treatment system to be realized. In
assessing the upstream treatment processes with respect to potential health risks
in wetland systems placed further downstream in the treatment train, it is
important to take into consideration that there is a distinction between
secondary treatment processes with respect to pathogen removal. For example,
there is a measurable difference in the potential public health risk to public
access between a lagoon secondary process with 20 to 40 days of HRT and an
activated sludge system where HRT's are typically 0.3 to 0.5 days. This scenario
results in ratios of 60:1 versus 120:1 in the difference in exposure to natural
disinfection processes (Gearheart, Personal Communication).
The level of pretreatment necessary for public contact may be achieved at the
end of the conventional process or at some point within the treatment wetland
complex. In the latter scenario, restricting public access may be more
challenging but is not impossible. In either case, the goal to eliminate the
likelihood of human pathogen transmission to those visiting the facility should
be paramount.
6-3
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Some communities have successfully convinced regulatory agencies to allow full
and/or limited public access to the wetland component of the wastewater
treatment facility after adequate pretreatment has been attained. Public access is
provided for or encouraged at a number of treatment wetland sites in the
communities of Arcata, Hayward, and Martinez, California; Cannon Beach,
Oregon; Incline Village, Nevada; Phoenix and Tucson, Arizona; and Iron Bridge
and Everglades National Park, Horida. Limited published data concerning
public use of these sites are available, including a thesis (Benjamin 1993) in
which it was reported that there are about 90,000 visitors per year over the
2,000-acre Hayward Marsh, and 140,000 visitors per year at the Arcata Marsh
and Wildlife Sanctuary. In 1998 the visitors to the Arcata, California, system had
increased to approximately 180,000 (Gearheart, Personal Communication).
Examples of projects with significant wetland habitat values and wildlife usage
are featured in Constructed Wetlands for Wastewater Treatment and Wildlife Habitat
-17 Case Studies [EPA 832-R-93-005,1993.] Available information on such
benefits is summarized in Treatment Wetland Habitat and Wildlife Use
Assessment: Executive Summary [EPA 832-S-99-001; June 1999]
Environmental Education and Interpretation Centers
Wetland treatment systems present an excellent focus and facility for
implementing community wide environmental education dealing with water
conservation, pollution prevention, wastewater treatment, water reclamation,
wetland ecology, watershed management, and energy conservation. The
wetland site can be designed to incorporate public access (limited or full),
esthetically pleasing viewsheds, riparian and upland fringe areas, and physical
structures for interpretative purposes. All of these components can complement
the wastewater treatment objectives of a city and increase the public stewardship
of water resources through awareness, protection, and participation in their
natural surroundings. One of the strongest cases for incorporation of these
benefits can be seen in subsequent support for water quality and watershed
protection requirements.
Some communities have constructed Interpretive Centers, which are the focus of
much of the organized environmental education occurring at FWS constructed
wetlands. Examples of interpretive centers can be found at Hayward Marsh,
California (East Bay Park District), and the Arcata Marsh and Wildlife Sanctuary,
California (Friends of the Arcata Marsh). Many other wetland systems have
incorporated informational signs into trail system(s) surrounding the wetlands
for environmental education. Local educational institutions can use FWS
constructed wetlands as a field trip site for biology, wildlife, and engineering
classes. In some communities, the wastewater utility forms partnerships with
school districts to allow use of the wetland and interpretative center for
environmental education. This component of a FWS constructed wetland allows
for unique and creative sharing of resources and spaces to meet larger
community needs.
6-4
-------
SECTION 8 AND
Open Water/Emergent Vegetation Ratio
Providing adequate open water areas (open water /emergent vegetation ratio) is
an important, but often overlooked, component in the design and
implementation of FWS constructed wetlands. Historically, many FWS
constructed wetlands were designed and built as fully vegetated basins with no
open water areas. Some systems configured in this manner experienced
problems with very low water column dissolved oxygen levels, incomplete
nitrification, odor production, and vectors, primarily mosquitoes.
Many natural wetlands contain a mix of open water and emergent vegetation
areas and they are as important for water quality reasons as they are for wildlife
purposes. Open water areas provide many functions such as reoxygenation of
the water column from atmospheric reaeration and algal photosynthesis, and
habitat and feeding areas for waterfowl, as well as allowing for the predation of
mosquito larvae by fish and other animals. Open water areas in FWS
constructed wetlands also reduce BOD concentrations and improve nitrification
of ammonia in wastewater because of the increased oxygen levels. In most
cases, it is recommended that a FWS constructed wetland incorporate a mix of
shallow vegetated and deep open-water areas that should result in a more
complex, dynamic, and self-sustaining wetland ecosystem that more closely
mimics a natural wetland.
The ratio of open water to emergent vegetation depends on the function and
goals of the FWS constructed wetland project. For constructed wetlands whose
primary function is wastewater treatment, the location and amount of open
water is a function of the nitrification requirement for that system. Open water
(submergent and floating aquatic plants) supports nitrification processes while
minimizing the internal carbon load. If land area is at a minimum, and/or costs
are to be kept low, then a minimal amount of open water area should be
provided. However, if land availability is not an issue, then a maximum amount
of open water area can be provided. Recommended open water to emergent
vegetation requirements range from 0 to 30 percent for treatment wetlands and
40 percent or greater for enhancement wetlands.
While higher open water may be desirable, treatment wetlands can operate
successfully at the suggested lower limits if constrained by land availability
and/or construction costs. Generally, enhancement wetlands will be designed
with larger open water areas for waterfowl and other wildlife than treatment
wetlands with water quality improvement as its only performance criterion.
Two methods can be used for creating open water areas: (1) excavate zones that
are deep enough to prevent vegetation growth, and (2) periodically raise water
levels to a depth that limits vegetation growth. Thus, wetland design and
operation can be used to control the types of plant communities that exist in FWS
treatment wetlands. The type of macrophytes (i.e. emergent, submergent, and
floating) can be controlled to some extent by the design operating water depth.
Water column depths of 1 to 1.5 m planted with submergents, such as
Potamogeton spp,, will not be encroached upon by emergent macrophytes like
Scirpus spp. and Typha spp. If the water column depth is between 0.2 to 0.6 m and
6-5
-------
SECTION 6 LEARNED AND RECOMMENDATIONS
planted with emergent vegetation, such as like Scirpus spp. and Typha spp., they
will prevail over submergents and fill in the surface area through rhizome and
tuber propagation. Alteration of water depth is a determining factor in
establishing various aquatic macrophyte communities to meet both water quality
and habitat objectives. A list of common wetland vegetation species and typical
growing depths was provided in Section 3.
Large open water zones that are not shaded by emergent or floating
macrophytes can allow significant blooms of phytoplanktonic or filamentous
algae to establish in FWS wetlands. However, if the open water areas are
designed for less than 3 to 4 days open water travel time, then algal growth
should not occur, as the growth cycle of algae is approximately 7 days. Finally,
if open water zones are located immediately adjacent to the outlet, the wetland
may not be able to consistently meet stringent standards for BOD, TSS, or
nutrients due to the export of algal solids. For this reason, it is recommended
that a large vegetated zone exist (emergent or floating aquatic plants) at the
outlet of a FWS constructed wetland to reduce sunlight penetration of the water
column.
Site Topography and Soils
Pre-existing topographic, geological, and soil chemistry conditions can greatly
affect wetland cost and performance. Excessive site relief creates large
earthwork volumes for a given wetland area, significantly increasing
construction costs. Surface and subsurface geologic conditions can also increase
costs by requiring removal of rock or by presenting the need for liner materials
to reduce groundwater exchanges. For the most part, level land with clay soils
affords the best physical setting for a FWS constructed wetland. However,
potential wetland sites with other conditions can be used, but may require more
complex engineering, earthwork, and construction techniques, and the use of
geotextile membranes.
Another consideration in the construction of a FWS constructed wetland is the
soil required to support the emergent aquatic plants. Substrate for this
vegetation should be agronomic in nature (e.g. topsoil), well loosened, and at
least 150 mm deep. If this type of soil exists at the site it can be scraped off prior
to excavation and saved, otherwise it can be imported from offsite. After the
wetland basin, berms, and other earthen structures are constructed and the liner
is installed (if required), the agronomic type soil can be placed back into the
excavated region. This pre-conditioned substrate will greatly increase the rate of
plant growth and extent of plant community coverage.
Another concern regarding soils is elevated concentrations of organic carbon,
organic nitrogen, or phosphorus, which may result in increasing concentrations
(negative removal efficiencies) between the wetland inlet and outlet following
system startup. This potential problem can be anticipated during design and
managed effectively by initial batch flooding to allow desorption and refixation
(SFWMD unpublished).
6-6
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Wetland Hydraulics
Inlet/Outlet Structures
Placement and type of inlet and outlet control structures are a critical feature in
FWS constructed wetlands. Within the general loading guidelines, control
structures are the most important feature after shape, in terms of wetland
treatment effectiveness and reliability. To minimize short-circuiting in a FWS
constructed wetland, two guidelines concerning inlet/outlet structures are
critical: (1) effective distribution of inflow across the entire width of the wetland
inlet and (2) uniform collection of effluent across the total wetland outlet width.
These guidelines will also minimize localized velocities around inlet/outlet
structures, thus reducing potential resuspension of settled solids. It is important
that any outlet structure be designed so that the wetland can be drained
completely, if required. Listed below are some of the common types of wetland
inlet/outlet systems in use today, and general guidelines regarding their design.
Two types of inlet/outlet structures are commonly used in FWS constructed
wetlands. For small or narrow wetlands perforated PVC pipe can be used for
both inlet and outlet structures. The length of pipe should be approximately
equal to the wetland width, with uniform perforations (orifices) drilled along the
pipe. The size of the pipes, and size and spacing of the orifices will depend on
the wastewater flowrate and the hydraulics of the inlet/outlet structures. It is
important that the orifices be large enough to prevent clogging with solids, but
small enough to provide uniform distribution along the length of the pipe. In
some cases, the perforated section(s) of this type of inlet/outlet structure can be
covered with gravel to provide more uniform distribution or collection of flows.
Where the local climate permits, the use of an exposed, accessible inlet and/or
outlet manifold is recommended for FWS wetlands to facilitate maintenance,
except in the cases where public exposure is an issue.
For larger wetland systems, multiple weirs or drop boxes are generally used for
inlet and outlet structures. Weirs or drop boxes are usually constructed of
concrete. These structures should be located no greater than every 15 m apart
across the wetland inlet width, with a preferred spacing of 5 to 10 m apart. The
same spacing requirements apply for the outlet weirs or drop boxes. Depending
on the source of the wastewater influent, the inlet weirs or drop boxes can be
connected by a common manifold pipe, or directly to the wastewater influent
source (a common arrangement for wetlands adjacent to oxidation ponds).
Whatever the configuration, it is important that the hydraulics of the manifold
and weirs be analyzed to ensure that uniform distribution occurs. Simple weir
or drop box type inlet structures are relatively easy to operate and maintain, but
generally provide less potential for solids settling in the inlet zone than a
perforated pipe inlet with its axially distributed load.
Depending on the type of wastewater influent, the inlet structure outflow point
can be located below or above the wetland water surface. Oxidation pond
effluent, for example, which is high in algal suspended solids, should be
introduced near the surface to allow for maximum settling, autoflocculation, and
6-7
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
predation to occur. Primary or secondary treated effluents should be introduced
below the surface if flocculated solids are expected, or if oil and grease, and/or
primary solids are expected.
Outlet structures represent an operational control feature that can affect wetland
effluent water quality. It is important that outlet structures have a wide range of
operating depths. By adjusting the outlet structure, both the water depth and
hydraulic detention time can be increased or decreased. The quality of wetland
effluent found in the upper layers of the water column is generally of higher
quality then water from lower depths, especially in terms of dissolved oxygen,
TSS, BOD, and hydrogen ion (pH). However, the differences in water quality
between water depths can be highly variable, and in some instances water from
lower depths can be of higher quality than upper layers. An outlet structure
design that allows for maximum flexibility of collection depths is recommended.
With this type of design, the outlet structure can be raised or lowered to draw
wetland effluent from the water depth with the best water quality.
Flow Measuring Devices
After analyzing the NADB, it became apparent that many existing wetland
systems do not have flow-measuring devices. Even if accurate estimates of
inflows and/or outflows to the treatment plant are known, internal flow
distribution to individual wetland cells was not known or measured. Without
accurate flow measurements to individual wetland cells, it is impossible to
determine actual flowrates and hydraulic detention times to each cell, thus
making flow adjustments difficult. It is recommended that some type of flow
measuring device be installed in all FWS constructed wetland projects and
further, separate flow measuring devices should be provided at each inlet and
outlet for multiple wetland cell configurations. Typical examples of flow
measuring devices include simple V-notch or rectangular weirs, and more
sophisticated Parshall flumes. Depending on the size and layout of the wetland,
flow measurement devices can and should be incorporated directly into
inlet/outlet structures,
Internal Drainage
In the event a FWS constructed wetland needs to be drained, the wetland bottom
should have a minimum slope of 0.1 percent to assist in drainage. Drainage may
be required for maintenance reasons such as liner repair, vegetation
management, and berm repair. Deeper channels may be required to allow for
drainage and/or continued use when serial cells are taken out of service.
Channels can also be used to connect deep-water pools, which may have been
designed into the project to afford open water for waterfowl, or to provide
refuge for fish and aquatic invertebrates during drawdown for maintenance.
6-8
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Engineering
FWS treatment wetland construction has several planning issues based upon soil
type, slope of the land, and cell configuration and shape. Other issues are
associated with the civil engineering aspect of the design, such as impermeable
barriers and liner materials, berm construction and specifications, inlet/outlet
structures, flow measuring devices, internal drainage, sediment settling zone,
and wetland planting. Many of these issues should be considered during the site
selection process, as they may become difficult or more costiy to correct later in
the actual design and construction of the FWS constructed wetland. For the
most part, the construction/civil engineering requirements are similar to other
earthen water quality management systems such as sedimentation ponds,
oxidation ponds, and sludge lagoons. The more important construction/civil
engineering design issues that need to be considered in a FWS constructed
wetlands project are as follows.
Berm Construction and Specifications
The height and width of berms or levees around FWS treatment wetlands is
important for a number of reasons. First, the berms must be able to contain all
design flows over a range of roughness conditions, including significant
headless through densely vegetated wetland cells with high aspect ratios.
Secondly, the berms must be high enough to account for normal or excessive
rates of solids deposition and peat building over the planned life of the wetland.
The third consideration is the need to hold and release peak wastewater inflows,
especially from collection systems with high infiltration and inflow rates or to
planning storage for systems that do not discharge during periods of the year
(typically winter). A fourth consideration is the need to protect berms from
damage by animals and root penetration.
Berms containing FWS wetland cells are generally built with 3:1 side slopes,
unless the soil characteristics allow for a steeper slope configuration, and a
minimum of 0.6 m of freeboard above the average operating water depth. For
wetlands that will receive high peak inflows, additional freeboard may be
required to ensure that berm overtopping does not occur. All external berms
should have a minimum top width of 3.0 m that should provide an adequate
road wide enough for most standard service vehicles to operate on. In some
cases, internal berms can have smaller top widths, as routine operation and
maintenance can be carried out by small motorized-vehicles, such as ATVs.
Road surfaces should be of the all weather type, preferably gravel to minimize
direct runoff into the wetland.
Berm integrity is critical to the long-term operational effectiveness of FWS
constructed wetlands. Common berm failure mechanisms include burrowing by
mammals such as beaver and muskrat, and holes from root penetration by trees
and other vegetation growing on or near the berms. Several design features can
eliminate and/or minimize these problems. The insertion of a thin impermeable
wall, or internal layer of gravel, installed during construction, can minimize
mammal burrowing and/or root penetration. Also, planting the berm using
6-9
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
vegetation with a shallow root system can be effective. Unlike oxidation ponds,
berm erosion in FWS constructed wetlands from wave action is generally not a
concern due to the dampening effect of the wetland vegetation.
In the design and site selection process, an important consideration is the
amount of additional area required for berms. In general, the higher the length
to width ratio for a FWS constructed wetland, the more area will be required for
the berms and for the entire wetland system. This increase in required total
wetland area to accommodate berms is more pronounced for smaller wetlands
(less than approximately 10 ha) than for larger wetlands, but in both cases
manifests itself as increased construction costs.
Wetland Configuration and Shape
There is substantial evidence, in both the design of oxidation ponds and FWS
constructed wetlands, that a number of cells in series can consistently produce a
higher quality effluent. This is based upon the hydrodynamic characteristics of
"tanks in series," where constituent mass is gathered at the outlet end of one cell,
and redistributed to the inlet end of the next cell. This process also minimizes
the short-circuiting effect of any one unit, and maximizes the contact area in the
subsequent cell. In addition, multiple treatment wetland cell configurations
operated in parallel add operational flexibility to the overall treatment process
and can facilitate maintenance activities. For treatment and water quality
purposes, a FWS constructed wetland system could consist of a minimum of 2 to
3 cells in series with the capability of taking one cell out of service; however, the
effects of headloss and inlet/outlet structures must be considered for systems
constructed in this manner.
The shape of a FWS constructed wetland can be highly variable depending on
site topography, land configuration, and surrounding landuse activities. FWS
constructed wetlands have been configured in a number of shapes, including
rectangles, polygons, ovals, kidney shapes, and crescent shapes. There is no
general data that supports one FWS constructed wetland shape as being superior
in terms of constituent removal and effluent quality, over another shape.
However, design issues such as hydraulic detention time, short-circuiting,
headloss, inlet/outlet structures, internal configurations, etc., do significantly
affect wetland effluent quality, and some wetland shapes could potentially
compound these problems over others.
Sediment Storage Zone at Inlet
Incoming settleable total suspended solids loadings are often waste stream
specific and are removed by discrete settling in the inlet region of a FWS
constructed wetland. Because a significant portion of the solids can often be
removed in the inlet area of the wetland, every effort should be made to
optimize the treatment potential of this region. It is recommended that some
type of open water (settling zone) or solids retention area is provided in the inlet
region of a FWS constructed wetland.
6-10
-------
6 AND
A settling zone could consist of an open water area that exists across the entire
width of the wetland inlet. A possible guideline is to design a settling zone such
that it provides approximately 1 to 2 days hydraulic detention time at the
average wastewater flowrate, as most of the suspended solids are removed in the
first 1 to 2 days of detention time in a FWS constructed wetland (refer to Section
4). The settling zone should be deep enough to provide adequate accumulation
and storage of settled solids, and to prevent the growth of emergent vegetation,
such as bulrush and cattails. The accumulated solids will slowly decay and
reduce in volume over time. However, at some time in the future the
accumulated solids may need to be removed from the settling zone.
Finally, inlet structure location and design will directly influence inlet velocities
in the settling zone. Velocities in the outlet zone are functions of the cell
geometry, vegetation pattern, and inlet/outlet type and location,
Wetland Planting
One of the most important considerations in the construction of a FWS
constructed wetland is the lead-time necessary to develop a fully vegetated
wetland. This factor enters into effluent compliance schedules and start-up
periods. The planting strategy can determine the length of time it will take to
reach functional densities of wetland vegetation. In general, the greater the
initial planting density, the sooner the vegetation stands are developed.
However, greater planting densities can also lead to greater planting costs. The
source and type of planting material is also a major concern. Wetland planting
success is highly dependent on the skills of the planting contractor, the type and
quality of planting material, the soil matrix, and the time of planting. At best, it
can be expected that a wetland will be producing target effluent values 2 or 3
years after completion of the planting, but this time is often waste and site
specific.
Two periods exist when wetland planting is most successful: fall and spring. In
the fall, tubers or clumps of aquatic emergent vegetation can be planted. Fall
planting allows the plants to acclimate to the new soil substrate slowly, as
wastewater is introduced at shallow depths. The hydroperiod (i.e. water depth
and duration of flooding) of the wetland should stay below the tops of any
newly planted emergent vegetation clumps or tubers. The other planting period
is spring when seeds, sprigs, tubers, and/or dumps can be introduced. Water
level control is much more critical to spring planting of sprigs, seeds, and tubers.
The most successful planting method for emergent vegetation, in either fall or
spring, is by placing clumps of 4 to 10 plants into the wetland on 0.6 to 1.0 m
staggered centers. These clumps include the native soils, along with multiple
tubers, which ensures the highest success rate of wetland planting. This type of
planting is limited to smaller systems and in areas where plant material is
available for harvest. Backhoes and dump trucks can be used to extract and
harvest the plants from acceptable and approved harvesting areas. Stems can be
cut off in late summer and fall plantings to facilitate transporting and planting of
the dumps. The cost of planting clumps is dependent on the distance to the
6-11
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
source of material. It is possible to have a fully functioning wetland in 1 to 2
years after planting with emergent clumps.
Other planting techniques include the use of purchased tuber stock and seeds
from commercial sources. Tuber stock is typically planted in a similar fashion to
transplanting seedlings and, depending on the size of the stock, can be planted
in spring or fall. For small tuber stock, spring planting is best. The use of seed is
the most risky way to vegetate wetlands. Seed treatment (acid, base, oxidizing
agents), seed placement (hand casting, hydroseeding), and water level
manipulation are all critical factors in the success of seed germination. Planting
with seed is less expensive than planting clumps or tubers, but the success and
rate of vegetation development is much less.
When planting seeds, sprigs, or tubers, it is necessary to bring to water levels up
slowly with the plant growth, starting at an initial depth of 20 to 30 mm, and
slowly increasing depth to 200 to 300 mm as the plants grow. Slow water depth
increases also ensure that wetland vegetation does not float before the roots take
hold into the soil. If the size of the wetland does not allow a 0 to 0.1 percent
bottom slope, then grading the wetland bottom into small sections separated by
shallow internal berms (200 to 300 mm in height) will be required. This
particular requirement is not needed when planting techniques incorporate
clumps of soil, roots, and stems.
Planting of the wetland should be done as soon as possible in the construction
sequence of a wastewater treatment plant. Often during wetland start-up, the
water quality is degraded due to algal growth, sediment resuspension, and
wildlife activity in the more open shallow water units. Treatment wetland
designers, owners, and regulators all need to take into account the method and
time of year of planting, as these are determinants for the time needed for
wetland vegetation to mature and hence the startup period needed between
initial planting and the production of effluent meeting discharge permit '
requirements.
Regional sources are usually able to supply relatively small amounts of plant
material, which in turn may influence initial planting densities. Planting tubers
or dumps on 0.5 m centers, for example, requires approximately 40,000 plants
per hectare. Planting on 1 m centers require 10,000 plants per hectare. If given a
planting constraint (cost or availability), it is better to place more plants in the
last half of the cell, than in the first half. It is important to ensure success of
planting of the last effluent half of the wetland. Planting 5 to 10 m wide
vegetated strips across the wetland width and perpendicular to the flow will
minimize short-circuiting, and allow for a future source of plant material for
later planting.
The emergent plants of choice for wetland treatment purpose are Scirpus species
(bulrushes) and Typha species (cattails). Of these two, Scirpus spp. appears to
have higher treatment potential based upon surface area. Hardstem bulrush, for
example, grows at higher stern densities, which affords much greater specific
surface area in the water column than does cattails. This specific surface area is a
critical growth location for attached microflora and microorganisms. Cattails are
6-12
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
generally larger in diameter than the bulrush, and have a much larger stem to
tuber transition in the water column. Further, hardstem bulrush does not
contribute as much detrital material during the dormant period as cattails,
thereby reducing potential BOD leaching back into the water column. Sdrpus
spp, wetlands have about one-third the background BOD as Typha spp. wetlands.
The seasonal change in plant community coverage in a FWS constructed wetland
is shown for the City of Arcata's Enhancement Marshes in Table 6-1. The loss of
open water to duckweed and sago pondweed coverage from spring to fall is
evident from the data given in Table 6-1.
Wetland plant growth and survival is also dependent on environmental factors
other than hydroperiod. Two of these include soil texture and soil chemistry.
Many wetland plants grow rapidly in soils of sandy to loamy texture. Soils with
excessive rock or clay material may retard plant growth and actually result in
mortality. Excessively acidic or basic conditions may limit the availability of
nutrients required for plant growth. In some cases, soil concentrations of macro
or micronutrients may not be available in the native soil for initial plant growth,
and organic fertilizers may have to be used.
TABLE 6-1
Percent of dominant plant species areal coverage of the Enhancement Wetlands of the Arcata Marsh
and Wildlife Sanctuary.
Enhancement marsh units, date
Allen Marsh Gearheart Marsh Mauser Marsh
April Sept. April Sept. April Sept.
Type of Cover 1986 1987 1985 1985 1986 1987
Open water 70.0 36.2 83.8
Common cattail 6.3 5.5
Marsh pennywort 5.6 10.0
Sago pondweed N V*
Alkali bulrush 11.9
Lesser duckweed 40.0*
Hardstem bulrush
Common spikerush
Upland grass spp. 30.0
5.0 32.5 23.0
6.0 10.5 4.3
11.8 27.0
77.2 NVb NV*
0.8
30.0° 69.6
2.3
0.7
* Duckweed coverage was too low because the wind had pushed it into windrows.
" NV = not visible because of duckweed coverage
As discussed earlier, there are several water quality reasons for balancing the
amount of open water area (submergent and floating), and the amount of
vegetated water area (emergent). Dissolved oxygen levels are maintained at
6-13
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
higher levels in open water areas, which supports aquatic organisms such as
aquatic insect larva, amphibians, and fish.
Impermeable Barrier and Liner Materials
A major concern with FWS constructed wetlands is the potential loss of water
from infiltration. While there are some wetland applications where infiltration is
desirable, the majority requires some type of barrier to prevent exchanges with
groundwater. Under ideal conditions, the wetland site will consist of natural
soils with low permeability that restricts infiltration. However, many wetlands
have been constructed or proposed on sites where soils have high permeability.
In cases where waste and site specific conditions warrant, some type of liner or
barrier can be required to restrict infiltration. Some general guidelines and
specifications for minimizing infiltration and berm storage losses are as follows.
Existing natural site soils with permeability less then approximately 10"* cm/s
are generally adequate as an infiltration barrier. For site soils with higher
permeability, some type of liner material is required. Some examples of wetland
liner materials include bentonite soil layers, chemical treatment of existing soils,
asphalt, and synthetic membrane liners. In some instances, existing in-situ soils
can be compacted to acceptable permeability. Whatever liner material is chosen,
an important consideration is to provide adequate soil cover and depth that
protects the liner from incidental damage and root penetration from the wetland
vegetation. Another consideration should be given to burrowing mammals such
as muskrats, nutria's rats, etc., which can do substantial damage by chewing and
consuming liner material.
Operation and Maintenance
The operation and maintenance of FWS constructed wetlands is much less
demanding than for mechanical wastewater treatment technologies such as the
activated sludge or trickling filter processes. Routine operation and maintenance
requirements for wetland systems are similar to those for oxidation pond
systems, and include hydraulic and water depth control, inlet/ outlet structure
cleaning, grass mowing of berms, inspections of berm integrity, wetland
vegetation management, vector control, and accumulated solids/peat
management if required.
Operation and maintenance considerations for FWS constructed wetlands are as
important as design issues in meeting regulatory requirements pertaining to
effluent water quality. The treatment effectiveness of most of the existing FWS
constructed wetlands can vary considerably depending on water depth, weir
overflow rate, plant density/plant location, and wildlife activity. Following are
some of the more important operation and maintenance considerations for FWS
constructed wetlands.
6-14
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Management of FWS Constructed Wetlands
Limited attention has been paid to the overall operation and maintenance
strategies of a FWS constructed wetland to meet water quality objectives. To be
accepted readily by regulatory agencies and owners, more effort should be
directed to developing holistic and sound management plans that cover a wide
range of issues associated with FWS constructed wetlands (Hammer 1992).
Many management issues pertaining to FWS constructed wetlands are not
mutually exclusive. Typically, one management decision or action influences
other management goals.
Listed below are considerations that need to be addressed when developing a
FWS constructed wetlands management plan:
• regulatory requirements
• hydroperiod and hydraulic retention time—water depth and flowrate
• hydraulic control—weir overflow rate/Inlet-outlet distribution
* vegetation control (planting, harvesting, and monitoring)
• proximity of airports
» wildlife management
» vector control (mosquitoes)
• structural integrity of berms
« nuisance conditions (odors)
• inlet/outlet structures
* public access
• environmental education
A set of operation and maintenance procedures needs to be developed for each
of the goals of the management plan developed above. This management
manual should be organized in a manner to assist the operator and owner in
effectively operating the wetland system under a wide range of environmental
conditions. At a minirnurri, the following categories should be included for each
goal of the management plan.
1. Objective and goal for the component
2. Startup condition/monitoring
3. Normal operating condition/monitoring/lead time
4. Abnormal operating condition/monitoring/lead time
* Problems
• Indicator
• Cause of abnormal condition
• Course of action to solve problem
5. Maintenance requirements
6. Sampling/monitoring program
6-15
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Potential Nuisance Conditions
Constructed and natural FWS wetlands are typically enriched semi-natural
wetland ecosystems. Because of their very nature, they have the potential to
create conditions that may be a nuisance to human neighbors or to the wildlife
species they harbor. Nuisances that could conceivably occur include mosquito
breeding, creation of odors, attraction of dangerous reptiles (snakes and
alligators), potential for accidental drowning, and the potential for
bioaccumulation or biomagnification of pollutants in wildlife (Hammer 1992;
Wass 1997). There is limited quantitative FWS treatment wetland data available
for these potential nuisances; however, some information is available on
mosquito and odor control. Unfortunately, there is inadequate data to date on
any of these issues to help assess all possible effects when implementing a FWS
treatment wetlands.
Wetlands and other stagnant water bodies can provide breeding habitat for
mosquitoes. Some of these mosquito species can transmit diseases to humans or
to valuable livestock. In addition, mosquitoes may be a nuisance because of
their large numbers and painful bites. Few quantitative data have been
published on mosquito population densities in treatment wetlands although a
large number of treatment wetland systems are periodically monitored for
mosquito larvae and pupae populations. General conclusions are that the
numbers of breeding mosquitoes in treatment wetlands are not higher than in
adjacent natural wetlands (Crites et al. 1995). When mosquito populations are
present, their numbers appear to be directly related to organic loadings (Martin
and Eldridge 1989, Stowell et al. 1985, Wieder et al. 1989, Wile et al. 1985, Wilson
et al. 1987).
Generally, odors in FWS treatment wetlands are associated with high organic
loadings, especially in the inlet region of the wetland. It has been observed that
most treatment wetlands have odors similar to the normal range of odors
observed in natural wetlands. No published qualitative information has been
found during preparation of this assessment on odors associated with treatment
wetlands.
Dangerous reptiles, including poisonous snakes and alligators, are attracted to
FWS treatment wetlands in some regions of the U.S. These same species are
generally a natural component of natural wetlands in those same areas, and
most citizens are aware of the need to avoid these animals when they are
encountered. No published information has been found on population densities
of these organisms in treatment wetlands or relating tihe occurrence of these
species to wetland design. Further, no data has been found indicating that
treatment wetlands are more or less likely to create risks to wildlife species than
adjacent natural wetland ecosystems. This issue is being examined further
through another EPA-funded project in progress.
Vegetation Management Implications
Routine harvesting of vegetation is not usually necessary for FWS constructed
wetlands (Reed et al., 1995). In many cases, the only routine vegetation
6-16
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
management consists of annual or biannual harvesting of emergent vegetation
from designed open water areas, and inlet/outlet structures. Over some period,
whose exact length is unknown, some removal of accumulated plant material
and detritus may be required in FWS constructed wetlands. Studies at Arcata
have indicated that detrital/litter has reduced the wetland volume by about 50
percent in 12 years with no apparent change in performance. However,
harvesting may be required if vegetation growth cycles significantly reduces
pollutant removal efficiencies, restricts water flow, affects habitat goals of the
project, and or inhibits wetland operation and maintenance activities.
If routine harvesting is required, it is recommended that the vegetation be
removed in 5 to 10 m strips perpendicular to the direction of flow. This strip
should be replanted and allowed to grow, before the next adjacent strip of
vegetation is harvested. This process can be repeated over a number of years.
The primary goal of this type of vegetation harvesting is that the wetland is
never completely devoid of vegetation at any one time. The harvested
vegetation can be transplanted, composted or burned; harvested wetland
vegetation has also been used for the production of methanol. It is also
important to consider potential effluent water quality impacts during vegetation
harvesting. Typically, the wetland cell being harvested is taken off line during
this time.
One problem that can be very difficult to manage for is the potential for animals,
in particular nutria and muskrat, to use the emergent wetland vegetation as a
food source. Some FWS wetland systems have had these creatures consume all
the emergent vegetation. If this occurs, the only action possible is to trap and
relocate the animals, and re-vegetate the damaged wetland cells.
Mosquito Control
Mosquitoes are common in any wetland or open water environment. However,
in some cases, especially urban environments, a FWS constructed wetland can
produce mosquito populations that are viewed as a nuisance by the public or
pose a true public health threat. Mosquito populations appear to be controlled
effectively in FWS treatment wetlands by small fish, such as the mosquito fish
(Gambusia affinis) (Dill 1989, Steiner and Freeman 1989). However, fish may not
be able to control mosquito populations in portions of FWS treatment wetlands
that are colonized by dense populations of floating vegetation mats (Walton et
al. 1990). This condition can be avoided by designing the FWS constructed
wetland with open water areas and by paying attention to vegetation densities
in emergent areas. Other animals, such as frogs, birds, and bats, may also
contribute to controlling mosquito populations.
Although biological methods have, and continue to, show promise for
controlling mosquitoes in treatment wetlands, mechanical means are also
available which may complement these efforts. Sprinklers have been
successfully utilized to control adult mosquito populations in constructed
wetlands (Epibare et al. 1993). The spray from overhead sprinklers disrupts the
water surface and affects the ovipositioning. This technique was very effective
6-17
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
in reducing mosquito larva production in a FWS wetland; however, additional
capital investment in the spray equipment and operation and maintenance of the
pump and sprinkler system is required,
Bacterially derived larvicides are another available mosquito control option. As
with any agent, whether it is a fish, another invertebrate, or a larvicide, the
effectiveness depends upon getting that agent in proximity to the target
organism, in this case mosquito larvae. This may require combining vegetation
removal with an efficient means of broadcasting, or otherwise delivering the
larvicide such that adequate basin coverage is achieved. The two most common
mosquito larvicides available are derived from Bacillus thuringiensis (Bti) and B.
sphaericus strains. When adequate basin coverage is achieved, both agents have
been used effectively to control mosquito populations. Bti was applied to the
Sacramento Regional demonstration wetland cells when mosquito larva reached
0.1 larva/dip. Bti was applied to an entire half-cell at a rate of approximately 2
kg (liquid) per hectare. Repeated Bti application and vegetation harvesting
around the edge appeared to be effective in avoiding high larval densities.
Vegetation harvesting and application of 11.2 kg/ha (10 Ibs./acre) of granular B.
sphaericus on roughly 3-week intervals during the summer of 1999 at the Tres
Rios demonstration wetlands also resulted in low mosquito larval counts and
appeared to be effective at lowering adult populations.
Process Control
FWS constructed wetlands have minimal need for active process control. The
only two operational controls for FWS wetlands are hydraulic loading and outlet
weir level control (if designed to allow varying hydro periods). Further,
hydraulic loadings can only be varied if alternative hydraulic pathways exist.
Under certain conditions, increasing the outlet weir level for a given period of
time will result in no discharge. This would allow for short-term periods of no
discharge to a receiving system. This increase in water level will increase the
HRT while maintaining the areal loading at a constant value. The maximum
depth that can be tolerated by emergent plants in the FWS wetland limits the
degree of water level increase. Generally this maximum depth is 1.0 to 1.5
meters while a more normal range for emergent plants is 0.4 to 0.75 meters.
Monitoring Requirements
The most critical monitoring issue during the wetland startup period is
vegetation growth and coverage. A wetland that does not develop sufficient
emergent/submergent vegetation becomes a shallow oxidation pond, producing
algae, BOD, and solids. The planting strategy, combined with hydroperiod
control as the plants grow, determines the effectiveness of vegetation growth
during the startup period. Other monitoring factors include control of aquatic
birds, mammals, and invasive vegetation during the startup period.
Once the wetland vegetation has been established, the system can be brought on
line and wastewater introduced. After the startup period is over, routine-
6-18
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
monitoring requirements will often be necessary. Most important in the
operation of a FWS constructed wetland is monitoring hydraulic and organic
loadings to, and discharge(s) from, the wetland system (including the
monitoring of individual wetland cells). Such tasks require measuring influent
and effluent flowrates, and recording of water depths in each wetland cell. This
information has not been collected routinely from many existing FWS
constructed wetland systems and that has slowed the broad-based acceptance of
this technology because data such as these can be used to assess inlet/outlet
distribution and performance. It is surprising that many wetland treatment
systems were not designed to gather this type of data even though this
information can be used to develop seasonal strategies, based upon hydraulic
and organic loadings, hydraulic detention times, and areal loadings.
Influent and effluent water quality constituents should also be measured on a
weekly or, at minimum, on a monthly basis. Parameters such as BOD, TSS, pH,
nutrients, temperature, specific conductance, and dissolved oxygen should be
monitored as these parameters can be used to assess wetland performance, and
determine constituent loadings. Table 6-2 lists suggested monitoring tasks for a
FWS constructed wetland; these data are important for understanding the
system performance and would improve the state-of-the-art for future design
efforts.
6-19
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
TABLI 6-2
Suggested monitoring requirements for a FWS constructed wetland.
Monitoring requirement
Hydraulic monitoring
Water depth
Inlet flowrate
Outlet flowrate
Water Quality monitoring
Dissolved oxygen
Temperature
Conductivity
pH
BOD
TSS
Nutrients (e.g. TN, NH4, NO3, TP)
Wetland biota monitoring
Vegetation coverage/distribution
Wildlife (nuisance animals)
Vectors (mosquitoes, etc)
Rsh
Birds1
Aquatic insect larva1
Civil Issues
Berm and liner (if used) condition
Inlet/outlet condition
Access road condition
Solids/peat buildup
Public Use1
Trail/sign conditions
Number of people
Location of monitoring
Each cell
Inlet of each cell
Outlet of last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Inlet each cell, outlet last cell
Each cell
Each cell
Each cell
Each cell
Each cell
Each cell
All berms
All inlet/outlet structures
All roads
Each cell
All trails
Access points
Frequency of
Large system
Weekly
Daily
Daily
Weekly
Weekly
Weekly
Weekly
Weekly
Weekly
Weekly
Bi-annually
Bi-annually
Weekly during
season
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Annually
Annually
Annually
monitoring
Small system
Weekly
Weekly
Weekly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Annually
Annually
Weekly during
season
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Annually
Annually
Annually
1. If required as part of management plan
6-20
-------
SECTION 6 AND
Considerations for Minimizing Variability in
Effluent Quality
Many items, most of which have been discussed throughout this Technology
Assessment, combine to influence the variability of effluent quality from FWS
constructed wetlands. The following are important design and operational
considerations, which can influence and help control variability of effluent water
quality, that need to be considered throughout the FWS constructed wetland
planning and design process.
1. Ability to buffer weekly fluctuations in effluent flow by use of
multiple cells.
2. Ability to store water individually in each wetland cell to allow for
longer hydraulic detention times for BOD and TN removal, and
quiescent conditions for settling processes.
3. Minimize the amount of emergent vegetation necessary to reach
treatment goals. The aquatic vegetation contributes to background
BOD, ammonia, and dissolved phosphorus levels in the wetland. The
lower the influent BOD, TSS, and TN, the greater potential
contribution this background source has to the variation in the
effluent value.
4. If wildlife habitat is one of the goals of the project, it is important to
have 3 to 7 days detention time of emergent vegetation at the final
wetland outlet. This emergent vegetation zone of the wetland has
minimal habitat value for migratory and resident birds (source
control), and provides a final clarification/vegetative filter zone.
5. Design outlet collection zones, and inlet/outlet structures to
minimize open water areas, which can attract wildlife and promote
phytoplankton and periphyton production.
6. Minimize the velocity fields at the inlet and outlet zones of the
wetland.
7. Design for solids removal at the inlet region of the FWS constructed
wetland.
Research Studies
Treatment wetland research studies should be designed to answer specific,
design-related questions. The size of wetland research cells, their source of feed
water, water depth controls, and sampling can all affect the ability to scale up
the conclusions to a full-size treatment wetland. Extremely small FWS wetlands
may have edge effects that result in behavior that is unrealistic compared to full-
scale wetlands. A pilot system may receive water in batch loads or in a nearly
continuous mode, neither of which is typical of most full-scale treatment
wetlands. Inlet constituent concentrations may be more constant in some pilot
studies than can be expected with a full-scale system. In small pilot wetland
6-21
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
cells (mesocosm-scale), sampling including plant harvesting can alter
performance significantly.
Additional, long-term, well-funded research studies would be very valuable for
advancing the FWS wetland technology. The Listowel, Ontario, database
represents a major contribution for the development of design criteria and
operational performance estimates for a cold-climate, cattail-dominated,
constructed treatment wetland. Research studies have been performed on the
City of Arcata, California, research wetland cells since 1980, with two major data
reports 1983 and 1986, and several papers summarizing research activities and
findings. The 4-year time frame of this research project and the excellent
monitoring and data reports were essential for maximizing the research benefits
of this project. It is recommended that regional sponsors be solicited to
contribute additional data, following the example established by the Arcata and
Listowel projects. Several facilities are currently available to provide a cost-
effective basis for additional pilot research. These facilities include the
Everglades Nutrient Removal test cells, the Champion pilot wetlands in
Pensacola, Florida, the Tres Rios demonstration wetlands in Phoenix, Arizona,
the Arcata, California, pilot wetland cells, the Orange County Water District,
California, demonstration wetland, and the Eastern Municipal Water District
pilot wetland cells in Hemet, California.
A common goal of all new and continuing wetland research studies should be
the achievement of a high level of quality assurance for all data collected. Water
flow and field parameters should be measured using calibrated instruments, and
analytical tests should follow accepted testing methods with adequate quality
control. All data should be validated prior to analysis and publication.
Following good scientific research practices will help to improve our
understanding of the transformation processes and will reduce the level of
uncertainty in predicting treatment performance of free water surface wetlands
in the future.
As part of an ongoing research effort, an interactive communication link for
operators, owners, designees, and regulators of FWS constructed wetland should
be established. The success of FWS constructed wetlands are not only
dependent on good designs but dependent on good operators and management.
A forum for discussion of operational issues with treatment wetlands wil ensure
the continued success of this important wastewater treatment technology.
Critical Research Issues
A number of critical research issues requiring additional information and data
analysis have been identified in this technology assessment report. These issues
deal with the relationships between design variables and system performance.
Some of the more pressing issues requiring resolution include the following:
• Specific studies on effluent water quality as a function of characteristics
of the inlet and outlet weirs, including number, location, type, spacing,
overflow rate, and outlet capacity
6-22
-------
6 AND
The effect of aspect ratio on internal flow patterns and wetland treatment
performance
Determination of volatile solids destruction and subsequent reuse of
soluble biodegradable by-products in the water column
In depth monitoring of selected full-scale projects to acquire an
understanding of systems receiving medium to high organic loads, and
of systems in cold climates
A qualitative description of the factors affecting the high variability of
removal rate coefficients
Detailed data sets for calibration of the sequential performance equations
used to describe the nitrogen transformations in FWS treatment wetlands
The relationship between the settled/decomposing solids and the
removal of BOD and ammonia in treatment wetlands
The spatial distribution of solids removal and nitrogen transformation
processes to identify wetland configurations and conditions that optimize
performance
The appropriate use of volumetric removal rate coefficients for treatment
wetland data analysis given their dependence on loading rates and water
depth
The importance of dissolved oxygen concentrations in control of wetland
performance for BOD and TN removal
The effect of open/deep water zones on internal flow patterns and
treatment performance
The importance of design criteria such as plant selection and open/deep
water zones on wildlife populations in treatment wetlands
The effects of different plant communities on treatment performance for
all major constituents of concern
The role of substrate surfaces in support of epiphytes and their role in
conversion and transformation processes
Additional information on factors affecting metal removal in treatment
wetlands including mass balances over extended operational time
periods
The normal range of quantitative fates and effects of potentially toxic
metals and organics in treatment wetland biota
Normal populations of levels of mosquitoes in treatment wetlands and
an understanding of the physical, chemical, and biological factors
affecting these populations
6-23
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
• Studies directed at the use of Integrated Pest Management (IPM) for
managing mosquito populations in FWS constructed wetlands
Other important issues are more difficult to study in a single research effort, but
instead need the collective input from wetland designers and operators of full-
scale treatment wetlands. These issues include:
• The optimum design and management of wetlands for multiple uses
such as treatment, habitat, and recreation
• The role of dissolved organic carbon generated from the decomposition
of detritus in treatment wetlands
• The effect of managing the hydroperiod over a weekly and monthly
period on the performance of treatment wetlands
• The role of the full range aquatic microorganisms and aquatic insect larva
as they interact with the partkulate material (public health significant
organisms, plant litter, TSS, etc.) in a FWS constructed wetland
Finally, much effort remains to be done with State and Federal agencies in terms
of defining the role and functions of FWS constructed wetlands in the various
wetland policies, and in the development of appropriate discharge standards.
6-24
-------
Appendix A - References
ADEM (Alabama Department of Environmental Management). 1988. Natural
Treatment Systems for Upgrading Secondary Municipal Wastezvater Treatment
Facilities. Prepared by ADEM and the Department of Civil Engineering, Auburn
University, Auburn, Alabama.
ADEQ (Arizona Department of Environmental Quality). 1995. Arizona Guidance
Manual for Constructed Wetlands for Water Quality Improvement. Prepared for
ADEQ by R.L. Knight, R. Randall, M. Girts, J.A. Tress, M. Wilhelm, and R.H.
Kadlec.
Allen, G.H. and R.A. Gearheart. 1988. Proceedings of a Conference on Wetlands for
Wastewater Treatment and Resource Enhancement. August 2-4,1988. Humboldt
State University, Arcata, California. 154 pp.
APAI (Alan Plummer Associates, Inc.) 1995. Use of Constructed Wetlands for
Protection of Water Quality in Water Supply Reservoirs. Prepared for Tarrant
County Water Control and Improvement District Number One in conjunction
with the American Water Works Association Research Foundation.
AZMET 1998. The Arizona Meteorological Network. University of Arizona,
Tucson, Arizona.
Bastian, R.K. and S.C. Reed, eds. September 1979. Proceedings of the Seminar on
Aquaculture Systems for Wastewater Treatment. USEPA Publication No. MCD-67.
Bavor, H.J., DJ. Roser, S.A. McKersie, and P. Breen. 1988. Treatment of Secondary
Effluent. Report to Sydney Water Board, Sydney, NSW, Australia.
Benjamin, T.S. 1993. Alternative Wastewater Treatment Methods as Community
Resources: The Arcata Marsh and Beyond. M.S. Thesis, University of California at
Berkeley.
Bothner, Brian. 1995. Assessment ofToxicity in Arcata's Integrated Wastewater
Treatment Facility. Thesis, Humboldt State University, Arcata, California.
Bouey, J. 1996. Personal communication.
Brix, H. 1993a. "Macrophyte-Mediated Oxygen Transfer in Wetlands: Transport
Mechanisms and Rates." In Constructed Wetlands for Water Quality Improvement,
G.A. Moshiri, ed. Boca Raton, Florida: Lewis Publishers.
Brix, H. 1993b. "Wastewater Treatment in Constructed Wetlands: System
Design, Removal Processes, and Treatment Performance." In Moshiri, G.A. (ed)
Constructed Wetlands for Water Quality Improvement, pp 9-22, Lewis Publishers,
Boca Raton, Florida.
A-1
-------
APPENDIX A - REFERENCES
Brix, H. 1994a. "Constructed Wetlands for Municipal Wastewater Treatment in
Europe." In Global Wetlands: Old World and New, W.J. Mitsch, ed. Amsterdam:
Elsevier.
Brix, H, 1994b. "Humedales Artificiales." Lectures on wetland treatment
presented in Zaragoza, Spain, 19-30 September, 1994.
Brown, D.S. and J.D. Waterman, 1994, "Wetlands Treatment Database." In
Proceedings WEFTEC1994, 67th Annual Conference and Exposition, Vol. 8, Part 1.
Byron Shire Council. Unpublished data.
Campbell, Kenneth, ed. 1995. Versatility of Wetlands in the Agricultural Landscape.
Proceedings of an international conference sponsored by the American Water
Resources Association and American Society of Agricultural Engineers,
September 17 to 20,1995, Tampa, Florida.
Choate, K.D., G.R. Steiner, and J.T. Watson. 1990. Second Monitoring Report:
Demonstration of Constructed Wetlands for Treatment of Municipal Wastewaters,
March 1988 to October 1989. TVA/WR/WQ-90/11. Chattanooga, Tennessee: TVA
Water Quality Department.
CH2M HILL. 1992. Carolina Bay Natural Land Treatment Program, Prepared for the
Grand Strand Water & Sewer Authority.
Cooper, P.P., ed. 1990. European Design and Operations Guidelines for Reed Bed
Treatment Systems. Prepared by EC/EWPCA Emergent Hydrophyte Treatment
Systems Expert Contact Group. Swindon, UK: Water Research Centre.
Cooper, P.P. and B.C. Findlater, eds. 1990. Constructed Wetlands in Water Pollution
Control, Proceedings of the International Conference on the Use of Constructed
Wetlands in Water Pollution Control, Cambridge, UK, September 24-28,1990.
Oxford, UK: Pergamon Press.
Cowardin, L.M., V. Carter, F.C. Golet, and E.T. LaRoe. 1979. Classification of
Wetlands and Deep Water Habitats of the United States. U.S. Fish and Wildlife
Service. FWS/OBS-79/31.
Crites, R.W., G.D. Dombeck, R.C. Watson, and C.R. Williams. 1997. "Removal
of Metals and Ammonia in Constructed Wetlands." In Water Environment
Research, Vol. 69, No. 2.
Crites, R.W., G.D. Dombeck, and C.R. Williams. 1996. "Two Birds With One
Wetland: Constructed Wetlands for Effluent Ammonia Removal and Reuse
Benefits," in Proceedings of the WEFTEC '96, Dallas, Texas.
A-2
-------
APPENDIX A - REFERENCES
Crites, R,W. and G. Tchobanoglous. 1998. Small and Decentralized Wastewater
Management Systems, WCB — McGraw-Hill, New York.
Crites, R.W., R.C. Watson, and C,R. Williams. 1995. "Removal of Metals in
Constructed Wetlands." In Proceedings of WEFTEC '95, Miami, Florida. Water
Environment Federation: Alexandria, Virginia.
Crumpton, W.G., T.M. Isenhart and S.W. Fisher. 1993. "The Fate of Non-Point
Source Nitrate Loads in Freshwater Wetlands: Results from Experimental
Wetland Mesocosms." In Constructed Wetlands for Water Quality Improvement,
G.A. Moshiri, ed. Boca Raton, Florida: Lewis Publishers.
DeBusk, T. A., K. R. Reddy, T. D. Hayes, and B. R. Schwegier. 1989.
"Performance of a Pilot-Scale Water Hyacinth-Based Secondary Treatment
System." In Journal WPCF, 61:1218-1224.
Dill, C.H. 1989. "Wastewater wetlands: user friendly mosquito habitats." In
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and
Agricultural, D.A. Hammer, ed. Chelsea, Michigan: Lewis Publishers.
Drew, M.A., ed. 1978. Environmental Quality Through Wetlands Utilization. A
Symposium on Freshwater Wetlands. Sponsored by the Coordinating Council on
the Restoration of the Kissimmee River Valley and Taylor Creek-Nubbin Slough
Basin. February 28-March 2,1978, Tallahassee, Florida.
DuBowy, P. and R. Reaves, eds. 1994. Constructed Wetlands for Animal Waste
Management. Proceedings of workshop sponsored by the Conservation
Technology Information Center, the U.S. Department of Agriculture Soil
Conservation Service, U.S. Environmental Protection Agency Region V, and
Purdue University Agricultural Research Program, April 4 to 6,1994, Lafayette,
Indiana.
Edwards, M.E. 1993. Unpublished data from the Tennessee Valley Authority.
Epibare, R. E. Heidig, and D. W. Gibson. 1993. "Prevention of Mosquito
Production At An Aquaculture Wastewater Reclamation plant in San Diego,
California, using an innovative sprinkler system." In Bulletin of the Society for
Vector Ecology 18 (l):40-44.
Ewel, K.C. and H.T. Odum, eds. 1984. Cypress Swamps. Gainesville, Florida:
University of Florida Press.
Fisk, D.W., ed, 1989. Wetlands: Concerns and Successes, Bethesda, Maryland:
AWRA.
Flach, K.W. 1973. "Land Resources." In Recycling Municipal Sludges and Effluents
on Land. Champaign, University of Illinois.
A-3
-------
APPENDIX A - REFERENCES
Florida Administrative Code (FAQ 1989. Florida Department of Environmental
Protection. Tallahassee, Florida,
Gearheart, R.A. 1992. "Use of Constructed Wetlands to Treat Domestic
Wastewater, City of Areata, California." In Water, Science, and Technology.
26:1625-1637.
Gearheart, R. A. 1993. "Phosphorus Removal in Constructed Wetlands."
Presented at the 66th WEF Conference and Exposition, Anaheim, California.
Oct. 3-7.
Gearheart, 1998, Unpublished data from Arcata Treatment Marshes, California.
Gearheart, R.A. and B. A. Finney. 1996. "Criteria for Design of Free Surface
Constructed Wetlands Based Upon a Coupled Ecological and Water Quality
Model." Presented at the Fifth International Conference on Wetland Systems for
Water Pollution Control, Vienna, Austria.
Gearheart, R.A., R. M. Gersberg, and Mike Ives. 1989. "Pathogen Removal in
Constructed Wetlands," presented at International Conference on Wetlands for
Wastewater Treatment, Chattanooga, Tennessee June 16-17,1988. Proceedings
published by Ann Arbor Press.
Gearheart, R.A., F. Klopp, and G. Allen. 1989. "Constructed Free Surface
Wetlands to Treat and Receive Wastewater Pilot Project to Full Scale." In D.A.
Hammer (ed) Constructed Wetlands for Wastewater Treatment, pp. 121-137, Lewis
Publisher, Inc., Chelsea, Michigan.
Gearheart, R.A., R. Norman, and L. Finger. 1993, Review of Wetland
Evapotranspiration Literature. Bullhead Lake, Arizona: Bureau of Reclamation
Gearheart, R.A., S. Wilbur, J. Williams, D. Hull, B. Finney, and S. Sundberg.
1983. City of Arcata Marsh Pilot Project: effluent quality results-system design and
management. Final report. Project No. C-06-2270, State Water Resources Control
Board, Sacramento, California. 127 pp.
Gearheart, R.A., Williams, J., Holbrook, H., and Ives, M. 1986. City of Arcata
Marsh Pilot Project Wetland Bacteria Speciation and Harvesting Effects on Effluent
Quality, Environmental Resources Engineering Department, Humboldt State
University, Arcata, California.
Gersberg, R. M., S.R. Lyons, B. V. Elkins, and C. R. Goldman. 1984. "The
Removal of Heavy Metals by Artificial Wetlands." In Proceedings Water Reuse
Symposium III. Vol. 2, American Water Works Association Research Foundation,
p. 639-648.
Gersberg, R. M., B.V. Elkins, R. Lyons, and C. R. Goldman. 1985. "Role of
Aquatic Plants in Wastewater Treatment by Artificial Wetlands," In Wafer
Research, 20:363-367.
A-4
-------
APPENDIX A - REFERENCES
Gersberg, R.M., R.A. Gearheart, and M. Ives. 1989, "Pathogen Removal in
Constructed Wetlands." In Constructed Wetlands for Wastewater Treatment:
Municipal, Industrial, and Agricultural, D.A. Hammer, ed. Chelsea, Michigan:
Lewis Publishers.
Godfrey, P.J., E.R. Kaynor, S. Pelczarski, and J. Benforado, eds. 1985. Ecological
Considerations in Wetlands Treatment of Municipal Wastewaters. New York: Van
Nostrand Reinhold Company.
Greeson, P.E., J.R. Clark, and J.E. Clark, eds. 1978. Wet land Function and Values:
The State of Our Understanding. Minneapolis, Minnesota: American Water
Resources Association.
Hall, B.R. and G.E. Freeman. 1994. "Study of Hydraulic Roughness in Wetland
Vegetation Takes New Look at Manning's n." In The Wetlands Research Program
Bulletin 4(l):l-4. USACOE, Waterways Experiment Station, Vicksburg,
Mississippi.
Hammer, D.A. and R.L. Knight. 1994. "Designing Constructed Wetlands for
Nitrogen Removal." In Water Science and Technology (29) 15-27,
Hammer, D.A. 1992. Creating Freshwater Wetlands, Lewis Publishers, Boca
Raton, Florida.
Hammer, D.E. and R.H. Kadlec. 1986. "A Model for Wetland Surface Water
Dynamics." In Water Resources Research. 22(13): 1951-1958.
Hammer, D.A., ed. 1989. Constructed Wetlands for Wastewater Treatment:
Municipal, Industrial, and Agricultural. Chelsea, Michigan: Lewis Publishers.
Hendry, G.R., J. Clinton, K, Blumer, and K. Lewin 1979. Lowland Recharge Project
Operations, Physical, Chemical, and Biological Changer 1975 —1978. Final Report to
the Town of Brookhaven. Brookhaven National Laboratory, Brookhaven, New
York.
Herskowitz, J. 1986. Listowel Artificial Marsh Project Report. Prepared for Research
Advisory Committee, Ontario Ministry of the Environment, Water Resources
Branch, Toronto. Project No. 128RR, October 1986.
Herskowitz, J.,S. Black, and W. Lewandowski. 1987. "Listowel Artificial Marsh
Treatment Project." In K. R. Reddy and W. H. Smith (ed) Aquatic Plant for Water
Treatment and Resource Recovery,. Magnolia Publishing, Inc., Orlando, Florida, p
248-254.
Hovorka, R.B. 1961. An Asymmetric Residence-Time Distribution Model for Flow
Systems, Dissertation, Case Institute of Technology.
IAWQ (International Association on Water Quality). 1994. Proceedings of the
Fourth International Conference on Wetland Systems for Water Pollution Control,
A-5
-------
APPENDIX A - REFERENCES
November 6 to 10,1994, Guangzhou, China. Guangzhou, P.R. China: Center for
International Development and Research, South China Institute for
Environmental Sciences.
Ives, M. 1988. "Viral Dynamics in Artificial Wetlands." In Proceedings of a
Conference on Wetlands for Wastewater Treatment and Resource Enhancement, G.H.
Allen and R.A. Gearheart, eds. Arcata, California: Humboldt State University.
Kadlec, R.H. 1994. Personal communication.
Kadlec, R.H., R.K. Knight, J. Vymazal, H. Brix, R. Cooper and R. Haberl. 2000.
Constructed Wetlands for Pollution Control—Process, Performance, Design, and
Operation. IWA Publishing, London, U.K.
Kadlec, R.H., W. Bastiaens, and D. T. Urban, 1993. "Hydrological Design of Free
Water Surface Treatment Wetlands." In Constructed Wetlands for Water Quality
Improvement, G.A. Moshiri, ed. Boca Raton, Florida: Lewis Publishers.
Kadlec, R.H. and R.L. Knight. 1996. Treatment Wetlands. Boca Raton, Florida:
Lewis-CRC Press.
Kadlec, R.H. and D.L. Tilton. 1979. "The Use of Freshwater Wetlands as a
Tertiary Wastewater Treatment Alternative." CRC Critical Reviews in
Environmental Control.
Kadlecik, Laura. 1996. Organic Content of Wetland Soils, Arcata Enhancement
Marsh, Special Project ERE Department Wetland Workshop. Arcata, California.
Kays, W.B. 1986. Construction of Linings for Reservoirs, Tanks, and Pollution
Control Facilities. 2nd ed., Wiley-Interscience, New York, New York.
Knight, R.L. Unpublished data from Champion pilot wetlands.
Knight, R.L. 1997. "Wildlife Habitat and Public Use Benefits of Treatment
Wetlands." In Water Science and Technology 35(5): 35-43.
Knight, R.L. and K.A. Ferda. 1989. "Performance of the Boggy Gut Wetland
Treatment System, Hilton Head, South Carolina." In Proceedings of the
Symposium on Wetlands: Concerns and Successes, D. Fisk, ed. Bethesda, Maryland:
American Water Resources Association.
Knight, R.L., J. Pries, R. Borer, R. Clarke, and V. Payne. 1996. Constructed
Wetlands for Livestock Wastewater Management. Literature Review, Database, and
Research Synthesis. Prepared for the Gulf of Mexico Program.
Knight, R.L., R.W. Ruble, R.H. Kadlec, and S.C. Reed. 1993. "Wetlands for
Wastewater Treatment Performance Database." In Constructed Wetlands for
Water Quality Improvement, G.A. Moshiri, ed. Boca Raton, Florida: Lewis
Publishers.
A-6
-------
APPENDIX A - REFERENCES
Knight, R.L., R.W. Ruble, R.H. Kadlec, and S.C. Reed. September 1993. Database:
North American Wetlands for Water Quality Treatment. Phase II Report. Prepared for
U.S. Environmental Protection Agency,
Kusler, J.A. and G. Brooks 1988. Proceedings of the National Wetland Symposium:
Wetland Hydrology. Berne, New York: Association of State Wetland Managers.
Lagrace, Sophie, C. Miller, and R.A. Gearheart. 2000. "Emergent Macrophyte
Density in a Constructed Wetland System for Wastewater Treatment." In
Proceedings of the Constructed Wetland Conference, Humboldt State University 2000.
Lakshman, G. 1982. A Demonstration Project at Humboldt to Provide Tertiary
Treatment to the Municipal Effluent Using Aquatic Plants. Saskatchewan Research
Council, SRC Publication No. E-820-11-B-28.
Martin, C.V. and B.F. Eldridge. 1989. "California's Experience with Mosquitoes
in Aquatic Wastewater Treatment Systems." In Constructed Wetlands for
Wastewater Treatment: Municipal, Industrial, and Agricultural, D.A. Hammer, ed.
Chelsea, Michigan: Lewis Publishers.
McAllister, Lynne S. 1993. Habitat Quality Assessment of Two Wetland Treatment
Systems in the Arid West - A Pilot Study. Corvallis, Oregon: U.S. Environmental
Protection Agency, Environmental Research Laboratory. EPA/600/R-93/117.
Meyers, J. 1998. Personal communication.
Mitsch, W.J., ed. 1994. Global Wetlands: Old World and New. Amsterdam: Elsevier.
Mitsch, W.J. and J.G. Gosselink. 1993. Wetlands. New York: Van Nostrand
Reinhold.
Mitsch, W. and X. Wu. March 1995. Olentangy River Wetland Research Park at the
Ohio State University. Annual Report 1994. Columbus, Ohio: Ohio State University
School of Natural Resources.
Moshiri, G.A., ed. 1993. Constructed Wetlands for Water Quality Improvement. Boca
Raton, Florida: Lewis Publishers.
NADB (North American Treatment Wetland Database). 1993. Electronic
database created by R. Knight, R. Ruble, R. Kadlec, and S. Reed for the U.S.
Environmental Protection Agency. Copies available from Don Brown, EPA, (513)
569-7630.
Nahar, Shaila N., P. Fox, and R. Wass 1998. "Sub-surface Investigation of Water
Quality from a Percolating Constructed Wetlands." In Proceedings ofWEFTEC
'98, the 71" Annual Conference and Exposition of the Water Environment Federation,
Orlando, Florida., pp. 325 - 336.
A-7
-------
APPENDIX A - REFERENCES
NAWCC (North American Wetlands Conservation Council). 1995. Wastewater
Applications of Wetlands in Canada. Prepared by CH2M HILL and Canadian
Wildlife Service. Sustaining Wetlands Issues Paper No. 1994-1.
Nolte and Associates. 1989. Harwich Septage Treatment Pilot Study - Evaluation of
Technology for Solar Aquatic Septage Treatment System. Prepared for Ecological
Engineering Associates, Marion, Massachusetts.
Nolte and Associates. 1997. Sacramento Regional Wastewater Treatment Plant
Demonstration Wetlands Project - 1996 Annual Report. Prepared for the
Sacramento Regional County Sanitation District, Elk Grove, California.
Nolte and Associates. 1998. Constructed Wetlands Demonstration Project. 1997
Annual Report. Prepared for Sacramento Regional County Sanitation District, Elk
Grove, California.
O'Connor, D.J. and W.E. Dobbins, 1958. "Mechanism of Reaeration in Natural
Streams." In ASCE Transactions, Paper No. 2934, pp. 641-684.
Odum, H.T. 1985. Self-Organization ofEstuarine Ecosystems in Marine Ponds
Receiving Treated Sewage. Data From Experimental Pond Studies at Morehead City,
North Carolina, 1968-1972, A Data Report. University of North Carolina Sea Grant
Publication #UNC-SG-85-04.
Ogden, M.C. 1994. Personal Communication.
Olson, R.K., ed. 1992. "The Role of Created and Natural Wetlands in Controlling
Non-Point Source Pollution." Proceedings of a U.S. EPA Workshop, Arlington,
Virginia, 10-11 June, 1991. In Ecological Engineering. 1(1/2) 170 pp.
Phelps, Earle B. 1944. Stream Sanitation. New York, New York: John Wiley and
Sons.
PBSJ (Post, Buckley, Schuh, and Jernigan). 1991. Orlando Easterly Wetlands: 1990
Monitoring Report, to City of Orlando, Florida.
Pilgram, D., Chairman. 1992. Wetland Systems in Water Pollution Control.
Proceedings of the International Association on Water Quality (IAWQ) Specialist
Conference, November 30 to December 3,1992, Sydney, Australia.
Pullin, B.P. and D.A. Hammer. 1991. "Aquatic Plants Improve Wastewater
Treatment." In Water Environmental Technology. March: 36-40.
Pullin, B.P. and D.A. Hammer. 1989. "Comparison of Plant Density and Growth
Forms Related to Removal Efficiencies in Constructed Wetlands Treating
Municipal Wastewaters," Preprint of paper presented at 62nd Annual
Conference WPCA, San Francisco, California, October 1989.
A-8
-------
APPENDIX A-REFERENCES
Reckhow, K. and Song S. Qian. 1994. "Modeling Phosphorus Trapping in
Wetlands Using General Models." In "Water Resources Research, Vol. 30 No. 11,
pp. 3105-3114.
Reddy, K.R. and E.M. D'Angelo. 1994. "Soil Processes Regulating Water Quality
in Wetlands." In Global Wetlands: Old World and New, W. Mitsch, ed.
Amsterdam: Elsevier.
Reddy, K.R. and W.H. Smith, eds. 1987. Aquatic Plants for Water Treatment and
Resource Recovery, Orlando: Magnolia.
Reddy, K.R. 1984. "Nutrient Transformations in Aquatic Macrophyte Filters
Used for Water Purification," in Proceedings of Water Reuse HI, Vol. 2,
American Water Works Association.
Reed, S.C. 1992. "Subsurface Flow Constructed Wetlands for Wastewater
Treatment - Status and Prospects." In Wetland Systems in "Water Pollution Control,
- J. Pilgram, Chairman. Proceedings of the International Association on Water
Quality (IAWQ) Specialist Conference, November 30 to December 3,1992,
Sydney, Australia.
Reed, S.C., R.W. Crites, and E.J. Middlebrooks. 1995. Natural Systems for Waste
Management and Treatment. 2nd Ed. New York: McGraw-Hill.
Richardson, B., ed. 1981. Selected Proceedings of the Midwest Conference on Wetland
Values and Management, St. Paul, Minnesota, June 17-19,1981. Navarre,
Minnesota: Freshwater Society.
Sartorius, J., et al. 1999. "Investigations of Nitrogen Transformations in a
Southern California Constructed Wastewater Treatment Wetland." presented
for publication to Ecological Engineering.
SCDHEC (South Carolina Department of Health and Environmental Control).
1992. Grant for Development of a Wetlands Classification and Standards System,
Effluent Criteria, and Criteria for Other Activities in Wetlands. Final Report.
Columbia, South Carolina.
Schwegler, B. R. 1978. Effects of Sewage Effluent on Algal Dynamics of a Northern
Michigan Wetland, M.S. Thesis. The University of Michigan, Ann Arbor, 53 pp.
Seidel, K., 1976. "Macrophytes and Water Purification." In Biological Control of
Water Pollution, J. Tourbier and R.W. Pierson, Jr., eds. Philadelphia: University of
Pennsylvania Press.
SFWMD (South Florida Water Management District). Unpublished data from
Everglades Nutrient Removal Project.
Small, M. and C. Wurm. 1977. Data Report. Meadow/Marsh/Pond System,
Brookhaven National Laboratory. BNL 50675.
A-9
-------
APPENDIX A - REFERENCES
Steele, Kenneth, ed. 1995. Animal Waste and the Land-Water Interface. Boca Raton,
Florida: Lewis Publishers.
Stefan, et al. 1994. "Influence of Macrophytes on Nitrate Removal in Wetlands."
In Ambio Vol. 23, No. 6.
Steiner, G.R. and R.J. Freeman, Jr. 1989. "Configuration and Substrate Design
Considerations for Constructed Wetlands Wastewater Treatment." In
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and
Agricultural, D.A. Hammer, ed. Chelsea, Michigan: Lewis Publishers,
Stephenson, M., G. Turner, P. Pope, J. Colt, A. Knight, and G. Tchobanoglous.
1980. The Use and Potential of Aquatic Species for Wastewater Treatment, Appendix
A, "The Environmental Requirements of Aquatic Plants," Publication No. 65,
California State Water Resources Control Board, Sacramento, California.
Stowell, R.,R. Ludwig, J.Colt, and G. Tchobanoglous. 1980. "Towards the
Rational Design of Aquatic Treatment Systems." Presented at the American
Society of Civil Engineers Spring Convention, Portland, Oregon. April 14-18,
1980. 43 p.
Stowell, R., S. Weber, G. Tchobanoglous, B.A. Wilson, and K.R. Townzen. 1985.
"Mosquito Considerations in the Design of Wetland Systems for the Treatment
of Wastewater," In Ecological Considerations in Wetlands Treatment of Municipal
Wastewaters. PJ. Godfrey, E.R. Kaynor, S. Pelczarski, and J. Benforado, eds. New
York: Van Nostrand Reinhold Company.
Sutherland, J.C. and R.H. Kadlec, eds. 1979. Freshwater Wetlands and Sanitary
Wastewater Disposal. Conference Abstracts. Higgins Lake, Michigan.
Tchobanoglous, G. 1993. "Constructed Wetlands and Aquatic Plant Systems:
Research, Design, Operational, and Monitoring Issues." In G.A. Moshiri (ed)
Constructed Wetlands for Water Quality Improvement, pp. 23-34, Boca Raton,
Florida: Lewis Publishers.
Tchobanoglous, G., R. Gearheart and R. Crites. 1991. "System Operation and
Monitoring," presented at the Natural/Constructed Wetland Treatment Systems
Wolverton, B.C., R.C. McDonald, and W.R. Duffer. 1983 Microorganisms and
Higher Plants for Wastewater Treatment. /. Environ. Qual, 12(2):236-242.
Workshop, sponsored by the U.S. Environmental Protection Agency, Region VTII
and the Colorado Department of Health, Denver, Colorado. September 4-6,
1996.
Tchobanoglous, Gv R. Stowell, R. Ludwig, J. Colt, and A. Knight. 1979. "The
Use of Aquatic Plants and Animals for the Treatment of Wastewater: An
Overview." In Aquaculture Systems for Wastewater Treatment: Seminar Proceedings
A-10
-------
APPENDIX A-REFERENCES
and Engineering Assessment, R.K. Bastian and S.C. Reeds, eds. Washington, D.C.:
U.S. Environmental Protection Agency. EPA 430/9-80-006.
Tilton, D.L., R.H. Kadlec, and CJ. Richardson, eds. 1976. "Freshwater Wetlands
and Sewage Effluent Disposal." In Proceedings ofNSF/RANN Conference. Ann
Arbor, Michigan: The University of Michigan. NTIS PB259305.
Tojimbara, T. 1986. A Quantitative Study ofPeriphyton Colonized on Artificial
Substrates at the City ofArcata's Marsh Pilot Project. M.S. Thesis. Arcata,
California: Humboldt State University.
TVA (Tennessee Valley Authority). 1990. Unpublished data from Benton,
Kentucky.
U.S. Environmental Protection Agency. 1983a. The Effects ofWastewater Treatment
Facilities on Wetlands in the Midwest. EPA 905/3-83-002.
U.S. Environmental Protection Agency. 1983b. Freshwater Wetlands for Wastewater
Management. Region IV Environmental Impact Statement. Phase I Report. EPA
904/9-83-107.
U.S. Environmental Protection Agency. 1984. The Ecological Impacts ofWastewater
on Wetlands: An Annotated Bibliography. EPA 905/3-84-002.
U.S. Environmental Protection Agency. 1985. Freshwater Wetlands for Wastewater
Management Handbook. EPA 904/9-85-135.
U.S. Environmental Protection Agency. 1988a. Design Manual. Constructed
Wetlands and Aquatic Plant Systems for Municipal Wastewater Treatment. Office of
Research and Development. Center for Environmental Research Information.
Cincinnati, Ohio. EPA/625/1-88/022.
U.S. Environmental Protection Agency. 1988b. Report on the Use of Wetlands for
Municipal Wastewater Treatment and Disposal. EPA 430/09-88-005.
U.S. Environmental Protection Agency. 1993a. Subsurface Flow Constructed
Wetlands for Wastewater Treatment: A Technology Assessment. Office of Water. EPA
Report 832-R-93-008.
U.S. Environmental Protection Agency. 1993b. Constructed Wetlands for
Wastewater Treatment and Wildlife Habitat. 17 Case Studies. EPA 832-R-93-005.
U.S. Environmental Protection Agency. 1999a. Constructed Wetlands Treatment of
Municipal Wastewater. Process Design Manual. EPA 625-R-99-010. Cincinnati,
Ohio: Technology Transfer Branch.
U.S. Environmental Protection Agency. 1999b. Treatment Wetland Habitat and
Wildlife Use Assessment Executive Summary. EPA 832-S-99-001.
A-11
-------
APPENDIX A-REFERENCES
U.S. Environmental Protection Agency. 1999c. Draft Guiding Principles for
Constructed Treatment Wetlands: Providing for Water Quality and Wildlife Habitat.,
Prepared by the Interagency Workgroup on Constructed Wetlands available on-
line at
van Oostrom, A.J. and R.N. Cooper. 1990. "Meat Processing Effluent Treatment
in Surface-How and Gravel-Bed Constructed Wastewater Wetlands." In
Constructed Wetlands in Water Pollution Control, P.P. Cooper and B.C. Findlater,
eds. Oxford, UK: Pergamon Press.
Valelia, I., S. Vince, and J.M. Teal. 1976. "Assimilation of Sewage by Wetlands
in Estuarine Processes." In Volume 1; Uses, Stresses, and Adaptations to the Estuary,
M, Wiley, ed. New York, New York: Academic Press, pp. 234-253.
Vela,G.R. 1974. "Effect of Temperature on Cannery Waste Oxidation." In
Journal of WPCF, Vol. 46. No. 1, pp. 198-202.
Walker, L.P. and M.R. Walker. 1990. City ofGustine Marsh Evaluation Study.
Report to the City of Gustine. Davis, California: Larry Walker Associates, Inc.
Walton, W.E., E.T. Schreiber, and M.S. Mulla. 1990. "Distribution of Culex tarsalis
Larvae in a Freshwater Marsh in Orange County, California." In Journal of the
American Mosquito Control Association. 6:539-543.
Wass, R, 1997. Tres Rios Demonstration Constructed Wetland Project - 1996/1997
Operation & Water Quality Report. Prepared for the City of Phoenix, SROG, and
Bureau of Reclamation. Phoenix, Arizona.
Water Environment Federation (WEF). 1999. Natural Systems for Wastewater
Treatment, 2nd ed. (in preparation). Alexandria, Virginia. Washington, DC: WEF.
Water Pollution Control Federation (WPCF). 1989. Natural Systems for Wastewater
Treatment. Manual of Practice FD-16. Alexandria, Virginia. Washington, DC:
WPCF.
Watson, J.T., S.C. Reed, R.H. Kadlec, R.L. Knight, and A.E. Whitehouse. 1989.
"Performance Expectations and Loading Rates for Constructed Wetlands." In
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and
Agricultural, D.A. Hammer, ed. Chelsea, Michigan: Lewis Publishers.
Watson, J.T., F. D. Diodato, and M. Launch. 1987. "Design and Performance of
the Artificial Wetlands Wastewater Treatment Plant at Iselin, Pennsylvania." In
K.R. Reddy and W. H. Smith (eds) Aquatic Plants for Water Treatment and Resource
Recovery.
Weber, A.S. and G. Tchobanoglous. 1986. "Prediction of Nitrification in Water
Hyacinth Treatment Systems." In Journal WPCF. 58(5):376-380.
A-12
-------
APPENDIX A-REFERENCES
Western Consortium for Public Health with EGA, Inc. 1996. Total Resource
Recovery Project, Final Report, prepared for the City of San Diego, California.
Oakland, California.
Wieder, R.K., G. Tchobanoglous, and R.W. Turtle. 1989. "Preliminary
Considerations Regarding Constructed Wetlands for Wastewater Treatment." In
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and
Agricultural, D.A. Hammer, ed. Chelsea, Michigan: Lewis Publishers.
Wile, I., G. Miller, and S. Black. 1985. "Design and Use of Artificial Wetlands."
In Ecological Considerations in Wetlands Treatment of Municipal Wastewaters, PJ.
Godfrey, E.R. Kaynor, and S. Pelczarski, eds. New York: Van Nostrand Reinhold.
Wilhelm, M., S.R. Lawry and D.D. Hardy. 1989. Creation and Management of
Wetlands Using Municipal Wastewater in Northern Arizona: A Status Report,
in D.A. Hammer (ed) Constructed Wetlands for Wastewater Treatment, p. 179-185,
Lewis Publishers, Inc., Chelsea, Michigan.
Williams, C.R., R.D. Jones, and S.A. Wright. 1996. "Mosquito Control in a
Constructed Wetlands." In Proceedings ofWEFTEC 96, Dallas, Texas.
Wilson, B.A., K.R. Townsend, and T.H. Anderson. 1987. "Mosquito and
Mosquitofish Responses to Loading of Water Hyacinth Wastewater Treatment
Ponds." In Aquatic Plants for Water Treatment and Resource Recovery, K.R. Reddy
and W.H. Smith, eds. Orlando: Magnolia.
A-13
------- |