Free Water Surface
Wetlands for
Waste water Treatment:
A Technology Assessment
U.S. Environment Protection Agency,
Office of Wastewater Management
U.S. Bureau of Reclamation
City of Phoenix, Arizona
With funding from the
Environmental Technology Initiative Program
Prepared by
Environmental Resources Engineering Department
Humboldt State University
Arcata, California
CHZM-Hili
Gainesville, Florida
March 1999
Wetland Management Services
Chelsea, Michigan
-------
FINAL DRAFT
Free Water Surface
Wetlands for
Wastewater Treatment:
A Technology Assessment
Prepared for
U.S. Environmental Protection Agency,
Office of Wastewater Management
U.S. Bureau of Reclamation
City of Phoenix, Arizona
with funding from the
Environmental Technology Initiative Program
Prepared by
Environmental Resources Engineering Department
Humboldt State University
Arcata, CA 95521
CH2M-HHI
3011 Southwest Williston Rd.
Gainesville, FL 32608-3928
Wetland Management Services
Chelsea, Ml
March 1999
-------
Table of Contents
Table of Contents ii
List of Tables . vi
List of Rgures vii
List of Equations x
List of Acronyms and Symbols xi
Acknowledgments xii
Section 1 Introduction to Free Water Surface Treatment Wetlands 1 -1
Background 1-1
Introduction to the Technology 1 -2
Types of Treatment Wetlands 1-2
Other Benefits of Treatment Wetlands 1-4
Historical Development of the Technology 1-6
Application of the Technology ..1-9
Summary of Technology Issues 1-11
Organization of this Report 1-12
Section 2 Methods for Technology Assessment 2-1
Data Sources •.„ 2-1
Technology Workshop and Peer Review 2-5
Data Quality and Validation 2-7
Section 3 Wetland Processes 3-1
Wetland Hydrology 3-1
Water Balance 3-1
Input Wastewater Flowrate 3-3
Precipitation ~ 3-4
Evapotranspiration 3-4
Output Wastewater Row 3-4
Percolation 3-4
Meteorological Effects on Wetland Water Budget. 3-5
Wetland Hydraulics 3-6
Wetland Hydraulic Definitions 3-6
Water Depth 3-6
Surface Area 3-7
Volume 3-7
Wetland Porosity or Void Fraction 3-7
Hydraulic Detention Time „ 3-8
Hydraulic Loading Rate 3-8
Water Conveyance 3-9
-------
TABLE OF CONTENTS
Aspect Ratio 3-9
Internal Flow Patterns Effects/Physical Facilities 3-70
Water Balance Effects on Wetland Hydraulics and Water Quality.... 3-11
Thermal Effects in Wetlands 3-11
Wetland Biogeochemistry 3-13
Total Suspended Solids 3-75
Processes 3-16
Settleable Solids Reduction-Anaerobic Decomposition » 3-17
Biochemical Oxygen Demand. 3-18
Nitrogen 3-79
Phosphorus 3-27
Chemical Oxygen Demand 3-23
Dissolved Oxygen 3-24
Hydrogen Ion 3-26
Constituent Characteristics 3-28
Aquatic Vegetation 3-29
Types of Wetland Vegetation 3-25
Vegetation Patterns. 3-30
Role of Aquatic Plants in Controlling Treatment Processes 3-32
Section 4 Performance Expectations. 4-1
Approach to Performance Evaluation 4-1
Data Base Evaluation (NADB and TADB) 4-7
Methodology 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 4-70
Temporal TSS Performance 4-72
TSS Permit Compliance 4-73
Nitrogen Performance 4-13
Organic Nitrogen Performance 4-74
Ammonia Nitrogen Performance 4-75
Total Kjeldahl Nitrogen Performance 4-77
Nitrate and TIN Performance 4-79
Total Nitrogen Performance 4-20
Nitrogen Permit Compliance 4-22
Ammonia Nitrogen 4-22
Total Nitrogen 4-22
Total Phosphorus Performance 4-23
Database Assessment 4-23
Temporal Phosphorus Performance.... 4-23
Total Phosphorus Permit Compliance 4-24
Fecal Coliform Performance 4-26
Database Assessment 4-26
Temporal Fecal Coliform Performance 4-28
-------
TABLE OF CONTENTS
Fecal Colifom Permit Compliance 4-28
Metals 4-28
Other Performance Considerations 4-30
Wetland Background Concentrations . 4-30
Natural Variability 4-32
Section 5 System Planning and Design Considerations 5-1
Planning Considerations 5-1
Role of Wetlands in the Watershed. 5-1
Additional Benefits/Habitat Considerations 5-4
Effluent Quality Considerations 5-5
Wetland Treatment System Objectives 5-5
Permitting.... 5-6
Public Access 5-8
Hydrological Considerations 5-8
Precipitation and Evapotranspiration 5-9
Groundwater •. 5-9
Ice and Snow 5-9
Engineering Considerations 5-10
Pre-Treatment Requirements. 5-70
Soils, Slope and Subsurface Geology. 5-70
Percolation and Use of Liners 5-70
Inlet/Outlet Types and Placement 5-7 7
Wildlife/HabitatConsideration 5-77
Environmental Impact 5-12
Land Use. 5-72
Insect Vectors 5-72
Odors 5-72
Wildlife and Ecological Attractive Nuisances. 5-73
Wetland Sizing 5-13
Approaches to Sizing 5-73
Assessment of Predictive Equations 5-74
Areal Loading Rate Method 5-76
Design Approach to Sizing 5-78
Section 6 Lessons Learned and Recommendations 6-1
Information Management 6-1
Planning 6-1
Multiple Benefits and Public Access 6-7
Environmental Education and Interpretation Centers. 6-2
Open Water/Emergent Vegetation Ratio 6-3
Site Topography and Soils 6-4
Hydrology ; 6-4
Wetland Hydraulics 6-5
Inlet/Outiet Structures 6-5
Flow Measuring Devices 6-6
Internal Drainage 6-6
IV
-------
TABLE OF CONTENTS
Internal Flow Pattern 6-7
Engineering 6-7
Bern Construction and Specifications 6-7
Wetland Configuration and Shape 6-8
Sediment Storage Zone at Inlet 6-9
Wetland Planting 6-9
Impermeable Barrier and Liner Materials 6-72
Operation and Maintenance 6-12
Management ofFWS Constructed Wetlands 6-12
Potential Nuisance Conditions 6-73
Vegetation Management Implications 6-14
Mosquito Control . 6-75
Process Control 6-75
Monitoring Requirements 6-76
Database Maintenance and Analysis 6-18
Considerations for Minimizing Variability in Effluent Quality 6-19
Research Studies.. 6-19
Critical Research Issues 6-20
Appendix A - References
-------
TABLE OF CONTENTS
List of Tables
TABLE 1-1 Additional benefits of NADB wetland wastewater treatment systems sorted by
treatment objective 1-6
TABLE 1-2 Timeline of selected events in wetland treatment technology (adapted from Kadlec and
Knight 1996) 1 -7
TABLE 1-3 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. 2-3
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 2-4
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 3-30
TABLE 3-3 Submerged surface area of wetland vegetation, normalized for a depth of 0.5 m 3-33
TABLE 4-1 Water quality constituent data availability for the FWS constructed wetland systems
included in this assessment, as 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 Metal removal data from free water surface treatment wetlands 4-29
TABLE 4-4 Long-term average annual outflow concentrations for lightly loaded FWS wetlands in
the NADB 4-31
TABLE 4-5 Expected range of background concentrations for constituents of interest 4-31
TABLE 5-1 Equations used to compute the performance of FWS constructed wetlands 5-15
TABLE 5-2 Range of areal loading rates for FWS constructed wetlands 5-17
TABLE 6-1 Percent of dominant plant species areal coverage of the Enhancement Wetlands of the
Arcata Marsh and Wildlife Sanctuary 6-11
TABLE 6-2 Minimum monitoring requirements for a FWS constructed wetland 6-17
VI
-------
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) 1-3
FIGURE 1-2 Ecosystem and communities of a FWS (USEPA 1993b) 1-5
FIGURE 1-3 Percentage of all communities utilizing FWS constructed wetlands based upon
community size (n = 135) 1-10
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) 3-3
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 3-6
FIGURE 3-4 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) 3-12
FIGURE 3-5 Conceptual partitioning of treatment processes through a FWS wetland 3-16
FIGURE 3-6 Wetland TSS removal processes 3-17
FIGURE 3-7 Simplified portrayal of wetland carbon processing. Incoming BOD5 is reduced by
deposition of particulate forms and by microbial processing in floating, epiphytic, and
benthal litter layers. Decomposition processes create a return flux 3-19
FIGURE 3-8 Nitrogen transformation processes in wetlands (Gearheart 1998, unpublished data) 3-20
FIGURE 3-9 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) 3-22
FIGURE 3-10 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 3-22
RGURE 3-11 BOD and COD effluent concentration before and during tap water loading to Arcata
Pilot Project wetland 3-23
FIGURE 3-12 Vertical distribution of DO in a submergent plant zone of the Arcata Enhancement
Marsh 3-25
FIGURE 3-13 Vertical distribution of DO in an emergent plant zone of the Arcata Enhancement
Marsh 3-25
FIGURE 3-14 Hydrogen ion (pH) buffering in system 3 at Listowel (Herskowitz 1986) 3-27
FIGURE 3-15 Metal sulfide burial processes in a wetland (Meyers 1998, personal communication) 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) 3-29
FIGURE 3-17 Coverage of plants during the startup period of the Arcata Pilot Project wetland 3-31
VII
-------
TABLE OF CONTENTS
FIGURE 3-18 Stem, leaf and litter cumulative surface area for Typha spp. in Houghton Lake
discharge zone wetland (Kadlec, 1997) 3-35
FIGURE 3-19 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) 3-35
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, MS 4-6
FIGURE 4-5 Influent and effluent monthly BOD for Lakeland, FL 4-7
FIGURE 4-6 Influent and effluent monthly BOD cumulative probability for Fort Deposit, AL 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 4-9
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
17days 4-17
FIGURE 4-18 Ammonia concentration transect through Arcata Pilot Project Wetland 4-17
FIGURE 4-19 Total Kjeldahl nitrogen loading versus effluent ammonia concentrations for the
TADB 4-18
FIGURE 4-20 Cumulative probability distribution of monthly influent and effluent TKN from Central
Slough, SC 4-18
FIGURE 4-21 Nitrate nitrogen loading versus effluent nitrate concentrations for the TADB 4-19
FIGURE 4-22 Cumulative probability distribution of monthly influent and effluent nitrate
concentrations for Orange County, FL 4-20
FIGURE 4-23 Monthly influent and effluent of total inorganic nitrogen (TIN) for the Arcata
Enhancement Wetland 4-20
FIGURE 4-24 Total nitrogen loading versus effluent total nitrogen concentrations for TADB
wetland systems , 4-21
FIGURE 4-25 Range of monthly inlet and outlet TN concentrations for cells 1 through 12 at the
Iron Bridge FWS wetland near Orlando, Florida 4-21
VIII
-------
TABLE OF CONTENTS
FIGURE 4-26 Phosphorus pulsing, as illustrated in a pilot cell in Arcata, California. Marsh 1
received tap water (no phosphorus load), while Marsh 3 received oxidation pond
effluent (Gearheart 1993) 4-24
FIGURE 4-27 Total phosphorus loading versus effluent phosphorus concentrations for the TADB
FWS systems 4-25
FIGURE 4-28 Cumulative probability distribution of monthly influent and effluent total phosphorus
concentrations for Central Slough, SC 4-25
FIGURE 4-29 Influent FC versus effluent FC for the TADB systems 4-26
FIGURE 4-30 Cumulative probability distribution of influent and effluent fecal coliform from Arcata
Pilot Project Cell 8, CA (Gearheart et al. 1986) 4-27
FIGURE 4-31 Cumulative probability distribution fecal coliform from Arcata Enhancement Wetland,
CA (Gearheart 1998, unpublished data) 4-28
FIGURE 4-32 Variation in effluent BOD at the Arcata Enhancement Marsh 4-32
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 5-3
FIGURE 5-2 Annual average BOD concentration vs. annual average area! BOD loading rate for
NADB systems 5-17
FIGURE 5-3 Tracer response curve for Sacramento Cell 7 (Nolle and Associates 1997) 5-19
-------
TABLE OF CONTENTS
List of Equations
(3-1) ~ = Q> -Qo + QC -Qb + Q,m +(P-ET-I)*A ... 3-2
at
(3-2) t = — 3-8
(3-3) Q avg = ^y^ 3'8
(3-4) q = — 3-9
(4-1) Ce = 3.42 + 0.262 Cj 4.3
(5-1) £l=. kappC 5-14
(5-2) C, = Coexp'"1"' 5-16
(5-3) kT = kajef™) 5-16
-------
TABLE OF CONTENTS
List of Acronyms and Symbols
ADEM Alabama Department of Environmental Management
ADEQ Arizona Department of Environmental Quality
ASCE American Society of Civil Engineers
BOD Biochemical oxygen demand
CBOD Carbonecous biochemical oxygen demand
CFU/100 rnL Colony-forming units per one hundred milliliters
CT Crites, Tchobanoglous Model
d Day
DO Dissolved oxygen
EFF Concentration reduction efficiency
ET Evapotranspiration
FAC Florida Administrative Code
FC Fecal coliform
FWS Free water surface
ha Hectare
HRT Hydraulic residence time
IAWQ International Association on Water Quality
kg Kilogram
kg/ha-d Kilogram per hectare per day
L Liter
m Meter
mg/L Milligram per liter
fig/L Microgram per liter
mL Milliliter
NADB North American Wetland Database
NAWCC North American Wetlands Conservation Council
NHs-N Ammonia nitrogen
NOs-N Nitrate nitrogen
NOD Nitrogenous oxygen demand
PFR Plug flow reactor
ppb Part per billion
ppm Part per million
RCM Reed, Crites, Middlebrooks Model
RED Mass reduction efficiency
SCDHEC South Carolina Department of Health and Environmental Control
SFWMD South Florida Water Management District
SRCSD Sacramento Regional County Sanitation District
TADB Technology Assessment Database
TIN Total inorganic nitrogen
TKN Total Kjeldahl nitrogen
TN Total nitrogen
TP Total phosphorus
TSS Total suspended solids
TVA Tennessee Valley Authority
USEPA U. S. Environmental Protection Agency
WEF Water Environment Federation
WPCF Water Pollution Control Federation
yr Year
xt
-------
TABLE OF CONTENTS
Acknowledgments
This report was prepared by the Environmental Resources Engineering Department of
Humboldt State University, Arcata, California under a contract with the City of Phoenix,
Arizona, CH2M-HU1, and Wetland Management Services. Material developed by
Wetland Management Services and CH2M-HU1 was utilized in the development of this
report. Special acknowledgment goes to George Tchobanoglous for reviewing and
editing the final draft report.
The following individuals reviewed the draft document and contributed constructively
to the final report.
Robert Bastian - USEPA Washington, D.C.
Robert Kadlec - Wetland Management Services, Chelsea MI
Robert Knight - Gainesville, FL
JimKriessl - 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 provided an initial point of entry into selecting
sites to be brought up to date and for systems which met the data quality criteria
established for the report prepared by Robert Knight, Robert Kadlec, and Sherwood
Reed under contract to the U.S. Environmental Protection Agency.
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. Considerable data were
obtained from owners, consultants and researchers working on FWS constructed
wetlands. Listed below are the wetland systems added to tine database and individuals
who provided data for this report.
Gustine, CA Mac Walker - 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
XII
-------
TABLE OF CONTENTS
Minot, ND Don Hammer, Norris, TN
Lakeland, FL Robert Knight - CH2M-HH1, 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 BUI Raddey, MGCRWA, MI
Manila, CA Wiley Buck, Manila Community Serv. Dist., Manila, CA
Beaumont, TX Bill Benner, Beaumont, TX
The principal authors of the final report were Robert Gearheart, Brad Finney, Margaret
Lang, Jeffrey Anderson, and Sophie Lagace.
XIII
-------
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. 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 a brief 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
FWS treatment wetlands has been accumulating since that time. 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 serves that purpose by describing the current understanding of FWS wetland
processes, and the performance of FWS treatment wetlands. Additionally, areas of
inadequate understanding of this technology are identified in this report. The findings of
this technology assessment have been incorporated into an update of the U.S.
Environmental Protection Agency's (EPA's) FWS constructed wetland design manual (EPA
1988a) and the Water Environment Federation (WEF) Manual of Practice on Natural
Systems (WEF, 1999), currently in preparation.
Three draft technical assessment documents have been prepared. A technical review team
comprised of researchers, USEPA representatives, consultants, BOR representatives, COE
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.
1-1
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Introduction to the Technology
Wastewater polishing systems utilizing wetland plants have proven to be very reliable.
Wetland aquatic plants, through their canopy, biomass, and rhizosphere, 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 forms, provide storage
for metals, cycle phosphorus, and attenuate organisms of public health significance. The
biogeochemical cycling of macro and micro nutrients 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 Gosselink
1993). The ability of wetland ecosystems to improve water quality naturally has been
recognized for more than 30 years. 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).
Types of Treatment Wetlands
Three general types of shallow vegetated ecosystems are used for water quality treatment
(1) free water surface wetlands, (2) subsurface flow wetlands, and (3) floating aquatic plant
treatment systems (Figure 1-1). All three of these vegetated treatment systems are operating
in the U.S. for water quality improvement. Early performance information for all three
system types has been published in a previous design manual (EPA 1988a). An update of
the 1988 manual is due to be published by EPA in 1999. A subsurface flow technology
assessment has already been completed (EPA 1993a). This present technology assessment
report focuses only on the FWS treatment wetland technology.
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
1-2
-------
SECTION 11NTRODUCTION TO REE WATER SURFACE TREATMENT WETLANDS
wetlands are designed to mimic the hydrologic regime of natural wetlands. Current
application of FWS treatment wetland technology is almost exclusively through the
construction of new FWS wetlands specifically for treatment and designed to enhance
possible ancillary benefits.
RGURE1-1
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 Pips
Outlet Weir
Low PermeabllRy Soil
Free Water Surface (Surface Flow)
Distribution Rpe
Outlet Weir
/ / s s / s s
Lined Basin
Free Water Surface with Open Water Zone
Distribution Rpe
Outlet Weir
S/SSSSSSS //'/"/'/'/•
Lined Basin
Floating Aquatic Plant System
This technology assessment includes performance data from both natural and constructed
free water surface wetlands. These 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 than constructed wetlands and to include a well-developed organic soil
1-3
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
component. Natural wetlands generally have variable water depths and stagnant water
zones outside the primary flow path. Natural wetlands can also have highly variable
inflows.
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 TSS) are trapped and tend to settle due to lowered flow velocities and
sheltering from wind. The particulates contain BOD components, fixed forms of total
nitrogen (TN) and total phosphorus (TP), and trace levels of metals and organics. These
particulate pollutants enter the biogeochemical element 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.
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 zones within FWS wetlands are nearly identical to similar zones
within ponds. A surface autotrophic zone dominated by planktonic 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. However, the shallow, emergent macrophyte zones present in FWS
wetlands operate quite differently than any zone within a facultative lagoon. Emergent
wetland plants tend to cool and shade the water surface reducing algae growth and limiting
water reaeration processes that create dissolved oxygen. 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. Applications of floating plant systems (water hyacinths) have been documented
by several studies peBusk et al. 1989, WPCF1989).
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). 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. As shown in Table 1-1, over 40% of the NADB secondary and 36% of the
NADB tertiary treatment applications identified one or more additional benefits with the
system objectives. Some ecological benefits can be claimed for nearly all FWS constructed
wetland systems regardless of their stated objectives. Benefits are often claimed for FWS
1-4
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
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.
FIGURE 1-2
Ecosystem and communities of a FWS (USEPA 1993b).
1-5
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
TABLE 1-1
Additional benefits of NADB wetland wastewater treatment systems sorted by treatment objective.
Primary
Secondary
Advanced Sec
Tertiary
Ponds
Other
None
Unknown
TOTAL
6
45
11
4
45
4
7
13
135
1
4
1
0
0
0
0
1
7
2
5
1
1
1
0
0
1
11
0
1
0
0
0
0
0
0
1
1
3
1
1
0
1
0
1
8
1
4
1
0
1
0
0
0
7
1
1
0
0
0
0
0
0
2
6
18
4
2
2
1
0
3
36
' ft
* /$
//if
V& Illustrative
Brookhaven; NY; Hay River, NWT
Benton, KY; Iron Bridge, FL
Cannon Beach, OR; Tres Rios, AZ
Orange County, FL; Reedy Creek, FL
Arcata, CA; Houghton Lake, MI
Armstrong Slough, FL; Des Plaines, IL
Historical Development of the Technology
Free water surface treatment wetland technology has been under development, with
varying success, for nearly 30 years in the United States (Table 1-2). In early laboratory
studies in Germany, the effects of emergent plants on removal of organic compounds in
industrial wastewater were examined (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 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 in Europe were subsurface flow systems designed
to treat primary treated 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). Subsurface flow wetlands using gravel
substrates have also been used extensively in the United States (Reed 1992). Subsurface
flow wetland technology is not generally applicable to lagoon-pretreated wastewater
because of dogging problems.
1-6
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
TABLE 1-2
Timeline of selected events in wetland treatment technology (adapted from Kadlec and Knight 1996).
Date
Location
Description
Selected Research Efforts
Plon, Germany
1952-late
1970s
1967-1972
1971-1975
1972-1977
1973-1974
1973-1975
1973-1976
1973-1977
1974-1975
1976-1982
1979-1982
1979-1982
1981-1984
1993
Morehead City, NC
Woods Hole, MA
Houghton Lake, Ml
Dulac, LA
Seymour, Wl
Brookhaven, NY
Gainesville, FL
Brillion, Wl
1975-1977 Trenton, NJ
1976-1979 Eagle Lake, IA
Southeast Florida
1979-1982 Humboldt, SK
1980-1984 Listowel, Ontario
Arcata, CA
Humboldt, SK
NSTL Station
Walt Disney World, FL
Florida
San Diego, CA
Santee, CA
Hemet, 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
Batch treatment of raw municipal sewage in lagoons and wetland
trenches by Lakshman and coworkers
Wolverton's work on hyacinths
Pilot-scale wetland work on a variety of wetland plants
Work on various aquatic plant in Florida by Ramesh Reddy, et al.
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
1-7
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
Date
1994
1995
Selected
1972
1973
1974
1975
1977
1978
1979
1979
1984
1986
1987
1987
Location
Tres Rios, AZ
Sacramento, CA
Full-Scale Projects
Bellaire, Ml
Mt. View, CA
Othfresen, West
Germany
Mandan, ND
Lake Buena Vista, FL
Houghton Lake, Ml
Drummond, Wl
Show Low, AZ
Incline Village, NV
Arcata, CA
Orlando and Lakeland,
FL
Myrtle Beach, SC
1987-1988 Benton, Hardin, and
Pembroke, KY
1988
1988
1990
1991
1991
1993
1993
1993
1995
1997
Hayward, CA
Orange County, FL
W. Jackson County, MS
Columbus, MS
Minot, ND
Everglades, FL
Beaumont, TX
Ouray, CO
Hidden Valley
(Riverside), CA
Cheney, WA
Description
Metals removal, effluent polishing, groundwater recharge
Metals removal
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 (m'/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
Hybrid treatment system combining constructed and natural wetland
units
Wildlife refuge linkage
First full-scale constructed wetland for advanced treatment of pulp
and paper mill 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
1-6
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
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. 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, FL, for agricultural drainage and
the 1200 ha Orlando, FL, wetland used to polish municipal effluent.
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 1-3, about 28% of the NADB treatment systems utilize
natural wetlands, 69% of the wetlands are constructed, and 3% are hybrid systems. About
65% of the natural wetland systems are receiving conventional secondary treated
wastewater. More than 45% of the constructed wetland systems are treating pond effluent
and 22% 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:
• NPDES secondary standards
• Total nitrogen
• Ammonia nitrogen
• Total phosphorus
• TMDL requirements
• Advanced secondary (BOD and TSS = 10 mg/L)
• Water reuse - groundwater discharge
Free water surface 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% of the FWS
systems have been built in communities 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, TX, Orlando, FL, HayWard, CA, and Riverside, CA).
Demonstration projects operated by Phoenix, AZ, Albuquerque, NM, and the Sacramento,
1-9
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
CA, Regional Wastewater Facility, are examples of locations for potential future large
community applications.
The largest number of FWS treatment wetlands are located in the states of South Dakota
and Florida (Figure 1-4). Most of the applications in these states utilize natural wetlands.
California has the next largest number of projects. The majority of these applications are to
meet effluent polishing and water reuse objectives.
TABLE 1-3
Percentage distribution of NADB FWS treatment systems by wetland type and level of pretreatmenL
Primary
Secondary
Advanced Secondary
Tertiary
Ponds
Other
None
Unknown
Total Number
6
45
11
4
45
4
7
13
135
33
53
18
50
2
25
43
15
37
67
44
82
25
96
75
57
69
93
0
0
0
25
2
0
0
0
2
0
2
0
0
0
0
0
0
1
0
0
0
0
0
0
0
15
2
RGURE1-3
Percentage of all communities utilizing FWS constructed wetlands based upon community size (n = 135).
1,000 10,000 100,000
Population Class Upper Limit
1,000,000
1-10
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETUNDS
RGURE1-4
Distribution of FWS constructed wetlands utilized for treating wastewater by State - not including pilot projects or
demonstration projects.
Summary of Technology Issues
The scope of this technology assessment is to present information necessary to determine
whether FWS wetlands are appropriate for achieving specific water quality and treatment
goals. The technical tasks of primary importance to applying this technology include the
following:
• Estimate accurately the influent flows and pollutant loads to the FWS treatment wetland
• Estimate wetland performance and the area and volume required to meet the limiting
water quality treatment goal{s)
• Design controls of the wetland hydrology and hydraulics to attain levels of performance
comparable to the performance of the operating systems used to derive empirical rate
constants
• Create and maintain the physical, chemical, and biological wetland system components
necessary to achieve expected pollutant processing rates
1-11
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
The first of these tasks, the need to predict design loading, is a standard procedure for
conventional wastewater treatment technologies and is not covered in this report. 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 (Arizona
Department of Environmental Quality [ADEQ] 1995, Kadlec and Knight 1996, Reed et al.
1995, EPA 1988a and 1988b, and Water Pollution Control Federation [WPCF, now WEF]
1989). Both the EPA and WEF manuals are due to be reissued in 1999. 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.
Organization of this Report
The goal of this report is to summarize nearly 30 years of FWS treatment wetland
information. Many of the volumes documenting the development of FWS treatment
wetland technology are briefly described herein and are cited in the Reference Section.
The methods used to prepare this technology assessment report are discussed in Section 2.
Data sources are described and information concerning data quality and validation are
presented. A FWS treatment wetland technology assessment workshop convened in Mesa,
Arizona, from February 2 to 4,1996, to guide development of this report is also described.
Key components of the physical, chemical, and biological processes occurring in FWS
treatment wetlands are summarized in Section 3. These fundamentals are essential for
presenting and interpreting FWS wetland performance data. The subject areas covered in
this section include: wetland hydrology, wetland hydraulics/ wetland treatment processes,
wetland vegetation and vegetation patterns, and wetland thermal effects.
The fundamentals necessary to evaluate and summarize FWS treatment wetland
performance are presented and discussed in Section 4. Normal 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 permit limitations.
System planning and design considerations are presented in Section 5. 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 are examined. Environmental impact and permit issues
1-12
-------
SECTION 11NTRODUCTION TO FREE WATER SURFACE TREATMENT WETLANDS
associated with constructed wetlands are also summarized in this section. Section 5
presents information concerning wetland system planning, design and sizing. 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. Chapter 5 includes construction considerations,
operation and maintenance considerations, and monitoring and management requirements.
Specific recommendations regarding the use and further development of a database for
FWS constructed wetlands are presented in Section 6. A list of critical operational research
issues is presented. Results from projects that address these critical issues would enhance
the understanding and application of FWS constructed wetlands to treat domestic
wastewater.
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 dear
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 will be 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, ADEQ1995,
Florida Administrative Code [FAC] 1989, South Carolina Department of Health and
Environmental Control [SCDHEC] 1992). The 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.
2-1
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-1
Listing of major treatment wetland conferences.
Date
Location
Description
May 1976
February 1978
November 1978
July 1979
June 1981
June 1982
July 1986
June 1988
August 1988
September 1989
September 1990
September 1990
June 1991
October 1991
July 1992
September 1992
December 1992
April 1994
July 1995
September 1995
May 1996
September 1996
Ann Arbor, Ml
Tallahassee, PL
Lake Buena Vista, FL
Higgins Lake, Ml
September 1979 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
November 1994 Guangzhou, China
Lafayette, IN
Fayetteville, AR
Tampa, FL
Fort Worth, TX
Vienna, Austria
Freshwater Wetland and Sewage Effluent Disposal
(Tiltonetal. 1976)
Environmental Quality Through Wetlands Utilization
(Drew 1978)
Wetland Functions and Values (Greeson et al. 1978)
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. 1985)
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 (Rsk 1989)
Constructed Wetlands in Water Pollution Control
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
(Moshiri 1993)
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 (International Association on
Water Quality [IAWQ] 1994)
Constructed Wetlands for Animal Waste Management
(DuBowy and Reaves 1994)
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 Hubert, in
preparation)
2-2
-------
SECTIONS 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. MOD 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.
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
2-3
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
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 Journal of the
Water Pollution Control Federation)
Water Environment Technology Water Research
Water Resources Journal Wetlands
Wetlands Journal International Association for Water Quality
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, WPCF1989). 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, NADB1993).
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, nine treat industrial wastewater, and six treat stormwater.
Many of these systems do not have detailed operational and performance data i.e., influent
and effluent flow data, multiple cell water quality measurements, vegetation type and
coverage, etc. 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:
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,
2-4
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
Informational/Data Category or Criteria
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 40 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 additional data from NADB sites were incorporated where available.
Source information is given whenever necessary for data or information used in this report,
and a listing of data used from the Table 2-6 FWS treatment wetlands is given in Appendix
A.
Technology Workshop and Peer Review
An initial draft 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. This final report, prepared by Robert A. Gearheart and George
Tchobanoglous with extensive input from numerous reviewers, reflects the data presented
and discussed, and insights offered by panelists at the workshop.
2-5
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
TABLE 2-6
FWS Wetlands used for performance evaluation (Technology Assessment Sites).
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) 1A
Gustine (89-90)18
Gustine (89-90) 1C
Gustine (89-90)10
Gustine (89-90) 2A
Gustine (89-90) 2B
Gustine (89-90) 60
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 Cut
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
NO
OR
FL
CO
KY
FL
FL
FL
CA
SC
AZ
SC
MS
Pretreatment Seasonal Origin
Pond
Pond
Pond
Pond
Pond
Pond
Secondary
Secondary
Primary
AdvSec
Secondary
Secondary
Adv Primary
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Pond
Secondary
Secondary
Pond
Pond
AdvSec
Pond
Tertiary
Pond
Secondary
Secondary
Tertiary
Tertiary
Secondary
Secondary
AdvSec
Secondary
Pond
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 Flow
(ha) (m3/day)
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
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
TABLE 2-7
Panelists for the Mesa, Arizona, workshop held February 2 through 4,1996.
Andrews, Tom L
Crites, Ron
DeBusk, Thomas A.
Dortch, Mark
Gearheart, Robert A.
Hammer, Donald A.
Kadlec, Robert H.
Knight, Robert L.
Mitsch, William J.
Moore, James
Payne, Victor W.E.
Reed, Sherwood C.
Reddy, Ramesh
Schueler, Thomas R.
Schwartz, Larry
Stiles, Eric
Tchobanoglous, George
Southwest Wetlands Group
Brown and Caldwell (formerly with Nolte and Associates)
Azurea, Inc.
U.S. Army Corps of Engineers
Humboldt State University
Hammer Resources, Inc.
Wetlands Management Services
CH2M HILL
Ohio State University
Oregon State University
Payne Engineering
Environmental Engineering Consultants
University of Florida
Center for Watershed Protection
Camp Dresser & McKee
Bureau of Reclamation
University of California, Davis
Data Quality and Validation
Data related to wetland design, operation, and performance have variable 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, benn 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.
The NADB contains data with 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,
2-7
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
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 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.
Additionally, the majority of the systems in the database are lightly loaded systems with
relatively low influent BOD and TSS concentrations. In many of these systems, the effluent
BOD is greater than the influent BOD. Figure 2-1 shows that as of 1993,50% of the systems
(and over 70% of the observations) documented in the NADB had average organic loads of
less than 5 kg BOD/ha-d. Approximately 28% of the systems measured had organic loads
less than 1 kg BOD/ha-d. Only 21% of the systems documented had loadings within the
normally 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). The vast majority 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. Over 44% of the influent BOD measurements for
FWS wetlands in the NADB were less than 10 mg/L (32% 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 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 conclusions stated in this report is proportional to the availability of
corroborating evidence and is indicated, when appropriate, throughout the text.
In summary the NADB was viewed as an overview inventory of wetland technology but
not sufficient in itself to provide loading data and discharge data to be statistically
analyzed. Individual sites and entries do provide data that can be used to predict
performance and to extrapolate performance to other sites.
2-8
-------
SECTION 2 METHODS FOR TECHNOLOGY ASSESSMENT
RGURE2-1
Influent BOD loading rates for FWS Wetland Systems in the NADB.
2.5
10 15 20 25 50 100 150 200
BOD Loading (kg/ha-day)
2-9
-------
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 to 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 ample time for settling and for
the wastewater to react with 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
in FWS constructed wetlands. These increased nutrient loadings generally result in higher
levels of biological production in FWS constructed wetlands receiving wastewater than in
natural wetlands.
Important wetland processes, as they relate to FWS constructed wetlands, are summarized
briefly in this chapter. Topics discussed include wetland hydrology, hydraulics,
biogeochemistry, aquatic vegetation, and thermal effects. The intent of this section is to
provide the reader with a brief introduction to wetland processes. For a more detailed
discussion of wetland processes, the reader may refer to many of 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). 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 turn, 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 the hydrologic balance between 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 waterbome substances and the wetland
ecosystem. A thorough understanding of the dynamic nature of the wetland water balance,
3-1
-------
SECTIONS WETLAND PROCESSES
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. In
contrast, FWS constructed wetlands are isolated from stream inflow, and their primary
sources of water are wastewater inflow, precipitation and runoff; water losses are via
infiltration entry into the outlet, evapotranspiration, and possibly percolation (if the
wetland bottom and sides are unlined and/or permeable). The steady inflow associated
with FWS constructed 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 driven system. Dryout 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 also 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 as:
^=Qi-Q0+Qc-Qb+Qim+(P-ET-irA (3-1)
where:
dV/dt = rate of change in water volume (V) in wetlan with time (t) (L3/t),
Qi = input Wastewater flow rate (L3/t), '
= output wastewater flow rate (L3/t),
= catchment runoff rate (L3/t),
= bank loss rate (L3/t),
Qsm = snowmelt rate (L3/t),
P = precipitation rate (L/t),
3-2
-------
SECTION 3 WETLAND PROCESSES
ET = evapotranspiration rate, meters per day (L/t),
I = exfiltration to groundwater (L/t),
A = wetland top surface area, square meters (L2),
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., Q, Q^ Q^, are generally ignored). In addition, groundwater
exfiltration (I) can be neglected if the wetland is lined with some type of impermeable
barrier.
FIGURE 3-1
Components of overall wetland water mass balance (Kadlec 1993).
Evapotranspkation=ET
Volumetric
Inflow=Qi
Berm Runoff=Qc
Precipitation=P
Berm Runoff=QC
S Surface Area=A
Depth=h~|~
Bankloss=Qb
Groundwater Exfliltration=I
Volumetric
Outflow=Q0
Input Wastewater Flowrate
The daily influent wastewater flow (Qj) 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
3-3
-------
SECTIONS WETLAND PROCESSES
and high infiltration and inflow rates into collection systems, the latter being a condition
which 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 wetiand 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 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.
Output Wastewater Flow
The output wastewater flow 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 to
offset the effects of high seasonal precipitation or evapotranspiration.
Percolation
In a FWS constructed wetland, percolation is the loss of water that occurs into the bottom
soils or berms. The effect of percolation is to reduce the water remaining in a wetland and
change the potential for each constituent transformation. The hydraulic load reduction is
further changed 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.
3-4
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-2
Total annual losses (+) and gains (•) from evapotranspiration and precipitation in cm (ET-P) (Rach, 1973).
-13
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.
The effects of precipitation and ET on monthly outflows from the Arcata, CA, FWS
constructed wetland system are shown in Figure 3-3. The solid line is the wastewater
outflow neglecting the effects of precipitation and ET (Q, = QJ. 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 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).
3-5
-------
SECTION 3 WETLAND PROCESSES
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
500,000
450,000
o
6
e»5
s
o
C
o
hi
01
!
I
400,000 •)>
350,000 \ •
300,000
Qout = Qin
Qout = Qin + (P-ET)A
CO CO CO CO CO CO CO CO CO CO CO CO ^* ^*' ^* ^ ^* ^ ^J* ^* ^* ^* ^
ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON
GJ-» •• ; >, c - w> <
Date
Wetland Hydraulics
Wetland hydraulics is the term applied to the movement of water through the wetland. An
improper hydraulic design can cause problems with water conveyance, water quality,
odors, and vector nuisances. 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
operated over a wide range of depths varying between 0.5 and 1.5 m (20 in. to 5 ft).
Depending on the bottom topography and slope of the water surface, 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
3-6
-------
SECTIONS WETLAND PROCESSES
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. Litterfall below the water surface and detrital buildup on the bottom
begin to reduce the depth therefore reducing the effective hydraulic volume.
Surface Area
The surface area (A) of a FWS constructed wetland is the area of the wetland water surface.
For many FWS constructed wetlands, an accurate surface area can be obtained from
construction drawings or as-built surveys. If construction or as-built information are not
available, a survey of the wetland water surface perimeter is necessary to determine surface
area. The area may also be estimated from an aerial photo. For most situations, the surface
area or the wetland footprint at the surface is a good estimate of the wetland bottom area.
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 peat 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 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 not in good agreement For example, Reed
et al. (1995) give wetland porosity values ranging from 0.65 to 0.75 for vegetated wetlands,
with lower numbers for dense, mature wetlands. However, Kadlec and Knight (1996)
report that average wetland porosity values are usually greater than 0.95, and e = 1.0 can be
used as a good approximation.
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 is
recommended 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, headless is
equally important. The friction coefficient that controls headloss through the wetland
depends on the vegetation density. Highly vegetated areas will have a greater headloss
than open areas, and this increased headloss may cause a significant backwater effect and
can lead to the development of preferential flow paths. If this potential backwater is not
3-7
-------
SECTION 3 WETLAND PROCESSES
accounted for in the FWS wetland design, inlet flooding may occur as the wetland
vegetation density increases or 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:
t=7f (3-2)
where: t = hydraulic detention time (d),
V = volume of wetland basin (m3),
e = wetland porosity, and
Q = flowrate (m3/d).
The flowrate (Q) value used in the hydraulic detention time calculation is generally one of
two values: input wastewater flowrate (Q,), or average flowrate (Q.^). 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
percolation, and assumes Ql = Q0. The input wastewater flowrate (Q,) should only be used
for preliminary calculations, or when no measurement or estimate (i.e. water balance) of the
output wastewater flowrate (QJ exists.
A more realistic measure of detention time can be computed using the average flowrate
(Q«g)m Equation 3-2 to account for the effects of water gains and losses (precipitation,
evapotranspiration and infiltration) that occur in a wetland. The average flowrate can be
estimated by:
n -1.0
(3-3)
The 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. 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:
3-8
-------
SECTIONS WETLAND PROCESSES
q- (3-4)
where: q = inlet hydraulic loading rate (m/d),
Q = flowrate (m3/d), and
A = wetland surface area (m2).
When the input wastewater flowrate (Qi) is used in Equation 3-4, the resulting calculation is
for the inlet hydraulic detention time, which neglects the effects of precipitation,
evapotranspiration and infiltration. Like hydraulic detention time, the average flowrate
(Q^) 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 handle all potential flows without
creating significant backwater problems, such as flooding the inlet structures or
overtopping of berms. In some cases, FWS wetlands were constructed without sufficient
consideration of headloss from submerged wetland vegetation, peat, and litter, resulting in
systems that are constrained hydraulically and cannot carry a range of flows without
overtopping the berms.
Assessment of the headloss from inlet to outlet can usually be done using Manning's
equation. When a more detailed headloss 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
3-9
-------
SECTION 3 WETLAND PROCESSES
with respect to headless. 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 headless and internal flow through the wetland. The weir overflow rate, location
of inlets and outlets, and elevation of berms is as important as 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, peat and other obstructions
(e.g., islands); forcing the water velocity to increase and decrease and continually change
direction. Water iri 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 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, as a result of 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 (both spatial and vertical) 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 is an important factor in
determining the hydraulic response of a FWS constructed wetland to wastewater inputs
and process withdrawals. There have been no studies conducted to test these factors
experimentally. 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. Several strategies exist in terms of the collection
volume at the terminus region of the wetland (Kadlec and Knight 1996). In one approach, a
deeper zone is created in the outlet zone with the outlet weir control structure placed away
from the bank into the collection volumes. Other approaches have been to collect the
influent in shallower depths through vegetated fenced to minimize fish and amphibian
entrapment in the effluent. Square non-adjustable weir structures have been experimented
3-10
-------
SECTIONS WETLAND PROCESSES
with to increase weir overflow rates 225 to 500 L/m -min (Gearheart, 1998, Unpublished
data).
Water Balance Effects on Wetland Hydraulics and Water Quality
The variability inherent in wastewater flowrates and the stochastic nature of meteorological
events controls wetland hydraulics, which in turn affects wetland water quality. The
impacts to wetland hydraulics can best be described by noting the increases and decreases
to the wetland hydraulic detention time caused by water gains and losses in the wetlands
water balance. Likewise, the wetland hydraulic detention time can also be used to explain
water balance impacts to wetland water quality.
Precipitation to a wetland increases inflow, which impacts 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. In systems receiving low influent
constituent concentrations, concentration reduction is likely to be poorer with precipitation
additions; in heavily loaded systems concentration reductions may be higher.
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.
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 temperature effects, however, varies with the constituent. In FWS
constructed wetlands, BOD removal does not always appear to exhibit a 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 which 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 operating water temperatures, and thus the corresponding range in
temperature dependent pollutant removal rates.
3-11
-------
SECTIONS WETUNDPROCESSES
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 temperature, T (Figure 34).
FIGURE 3-4
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).
30
25 •
20-
S> 15 -
|
g 10 -
o>
D.
| 5H
0-
-5-
-10
Listowel Air
Listowel Effluent
Orlando Easterly Air
Orlando Easterly Effluent
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
Month
If a frozen season is present, insulating layers of snow and ice can change the application of
the energy balance considerably. There is no longer a large radiation input to the water, and
energy gains are now solely from deep soil storage. Losses are by heat conduction through
the snow and ice to the cold air above and to ice formation. Incoming sensible heat is
typically dissipated because losses are generally greater than gains. Evaporation from the
water layer is prevented by the ice cap. As a consequence, gains and losses do not balance
as in summer, and temperature decline will typically proceed throughout the flow path.
The amount of ice formation is determined by climatological 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
3-12
-------
SECTION 3 WETLAND PROCESSES
experienced ice thickness on the order of 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.
Wetland Biogeochemistry
Free water surface treatment wetlands support a variety of sequential and often
complimentary treatment processes. The predominant physical, chemical, and biological
processes 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. The specific processes controlling
biological oxygen demand, total suspended solids, nitrogen, dissolved organic phosphorus,
chemical oxygen demand, dissolved oxygen, 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 3-5. Wetland treatment processes are generally associated with vertically
and horizontally differentiated zones within the wetland volume. These zones are linked
both hydrodynamically 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 further 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 released to the water column by
photosynthesis is greatest. Reduction of the ammonia nitrogen (nitrification) in a wetland
occurs where carbonaceous BOD has been generally satisfied, and sufficient dissolved
oxygen is present in the water column. The zone where net removal of total inorganic
nitrogen occurs follows the open water zone where aerobic nitrification occurs.
Denitrification has been shown to be a significant process in FWS constructed wetlands. The
combination of anoxic conditions, a physical substrate for the bacteria films, and an internal
carbon supply provide ideal conditions for nitrate conversion to nitrogen gas. The
dissolved organic carbon produced as a by-product of detrital decomposition supplies the
carbon for this microbial process. Because denitrifiers 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 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 or locations. A
3-13
-------
SECTIONS WETLAND PROCESSES
more detailed discussion of the role of unique features of FWS constructed wetlands and
the processes controlling specific constituents of interest follows.
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).
Mechanism Water Quality Constituent*
BOD TSS N P DO Bacteria Heavy
Virus Metals
Description
Physical
Absorption S P/S
Adsorption/ I S P
desorption
Emulsification S
Evaporation I
Filtration I S
Impaction
Flocculation P P P
Photochemical
reactions
Sedimentation P P I I I S
Thermal I P S
Volatilization P
Chemical
Adsorption P S
Chelation S
Chemical
reactions
Decomposition . P
Oxidation/ P S
reduction
reactions
Gas transfer to and from water surface
I Interparticle attractive force (van de
Waals force); hydrophylic 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); hydrophylic 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
vims 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.
S On substrate and plate surfaces
P Formation of complexed 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
3-14
-------
SECTION 3 WETLAND PROCESSES
Mechanism Water Quality Constituent*
BOD TSS N P DO Bacteria Heavy
Virus Metals
Description
Precipitation
Biological
Algal synthesis
Assimilation, C C
plant
Bacteria/
Metabolism
Aerobic P/C S
Anaerobic P/C
Plant
P
S S
S P/S I/C 1
1 1 P P
C C
S S C
P Formation of co-precipitates with
insoluble compounds
The synthesis of algal cell tissue using
the nutrients in wastewater.
S 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
S Under proper conditions, significant
adsorption
Predation
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:
constituent), C = contributory effect,
= secondary processes, I = incidental effect (occurring with removal of other
S/P = depends on influent and design conditions, N = negative.
3-15
-------
SECTIONS WETLAND PROCESSES
RGURE3-5
Conceptual partitioning of treatment processes through a FWS wetland.
: Litter Layer: I -"-' '\ > " -I. '*-,'." \ " : I 'X' '\ \ v: - 'X1. ^ > " : -' ~.
, \ I *. ' / ^\^^^. , * f _• x ' _ _i f. .L/ .'/ v X. r I / v ' /_ V X! />/_>_'/ ^ /i ^ <. f ^
Soluble BOD Removal
Solubilization
Ammonification
Denitrification
Nitrification
fjDiscrete
Flocculent Settling
••••••i
Anaerobic Decomposition ivHYHTITaiiil Methanoaenisis
Sedimentation-Detrital Buildup-Peat Development
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. Total suspended solids settle in the inlet zone
as colloidal solids flocculated as they pass through vegetated zones. Anaerobic conditions
produce sulfides that assist in binding metals as acid volatile sulfides. Soluble organic
constituents are reduced to carbon dioxide and low molecular weight organic acids. This
combination of removal processes is generally referred to as filtration by wetland scientists,
although stem and litter densities are not typically high enough to act as a filter mat. As
shown in Figure 3-6, in addition to trapping incoming TSS, a number of wetland processes
produce particulate 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.
The dominant TSS removal process is sedimentation, although all of the above processes
contribute to TSS reduction. In wetlands, velocity-induced resuspension is minimal, but gas
lift and bioturbation can reintroduce solids into the water column. Wetland sediments and
micro-detritus are typically near neutral buoyancy, flocculant, and easily disturbed.
Bioturbation by fish, mammals, and birds can resuspend 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.
3-16
-------
SECTIONS WETLAND PROCESSES
Resuspension 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.
FIGURE 3-6
Wetland TSS removal processes.
Sedimentation Invertebrate . .,.. .. Macrophyte Periphyton
'itterfa.1 Autoflocculato" '
The magnitude of wetland particulate cycling is large, with high internal levels of gross
sedimentation and resuspension, and almost always overshadows TSS influent loadings.
TSS effluent concentrations rarely result from an irreducible fraction of the TSS influent,
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 man 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 loadings that stimulate high TSS production may
eventually lead to a measurable increase in bottom elevation (van Oostrom and Cooper
1990). However, no treatment wetland has yet required maintenance because of solids
accumulation, including some that have been in operation for 20 years or more. In
situations of high incoming non-volatile solids, a settling basin can be designed to intercept
a large 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 benthal 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 densities as. anaerobic processes release aerobically degradable by-products
3-17
-------
SECTIONS WETUNDPROCESSES
to the sediment and organic layer pore water. These aerobically degradable by-products
subsequently diffuse into the overlying water column and contribute to the BOD.
The accumulated organic debris degrades at different rates depending on the source and
composition of the organic 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. For example, the half-life of soluble BOD is approximately 3 days while the
half-life of organic sediment, which is temperature-dependent, is more on the order of 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. The oxygen requirements of benthal 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 wastewaters, some fraction of the influent
carbon compounds are dissolved while the rest enters in the form of particulates.
Particulate settling provides one removal mechanism, and typically occurs in the inlet
region of the wetland (Figure 3-7). 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. The 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, and decomposition of plant litter returns a significant fraction of this
carbon back to the water column. The decomposition of litter and sediments produces a
return flux of BOD5 to the water column. The balance between removal of influent BOD5
and the decomposition processes contributing BOD5 determines the wetland effluent BOD5
concentration.
3-18
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-7
Simplified portrayal of wetland carbon processing. Incoming BODs is reduced by deposition of participate forms
and by microbial processing in floating, epiphytic, and benthal litter layers. Decomposition processes create a
return flux.
Settleable solids (paniculate
BOD and suspended solids)
Plant litter
Settled suspended/flocculated solids and detritus (peat buildif
W Release of soluble BOD from destruction of volatile settleable solids
W Release of soluble BOD from destruction and decomposition of solids and detritus
V Release of soluble BOD from decomposing plant litter
Nitrogen
Nitrogen is a key element in biogeochemical cycles and occurs in a number of different
oxidation states in wastewater and treatment wetlands. Numerous biological and physio-
chemical processes can transform nitrogen between its various oxidation states (Figure 3-8).
The dominant nitrogen species in 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 [NJ 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,
Nti,*) and a smaller percentage as un-ionized ammonia (NH3). The distribution of total
ammonia between NrV and NH3 depends on water temperature and pH. Un-ionized
ammonia is volatile and may be lost directly to the atmosphere.
3-19
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-8
Nitrogen transformation processes in wetlands (Gearheart 1998, unpublished data).
Atmosphere
c
J3
O
O
Blue-green
algae and
Azotobacter
Phytoplankton,
bacteria,
and aquatic
macrophyte
Azotobacter
N
Y Aerobic/ Organic
nitrogen
fixation
Denitrification
Ctostridium
pasteurianum
L_
Anaerobic
nitrogen
fixation..
!
CO
Anaerobic
decomposition
Decomposition
resistant N
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.
Nitrite nitrogen is converted to nitrate nitrogen under aerobic conditions. Free dissolved
oxygen is utilized in this process. Nitrate nitrogen is readily transformed to di-nitrogen gas
in treatment wetlands by the anaerobic process, denitrification. Denitrification occurs most
readily in wetland sediments and in the water column below fully vegetated growth where
3-20
-------
SECTIONS WETUNDPROCESSES
dissolved oxygen concentrations are low and available organic carbon is high. Organic
carbon is consumed in this microbial process and alkalinity is produced. To complete the
cycle, atmospheric di-nitrogen gas can be microbially fixed in open zones as organic N and
reintroduced into the wetland through nitrogen fixation. However, this transformation is
not normally a significant contribution of organic N to FWS treatment wetlands. Because of
the complex transformations affecting nitrogen species in wetlands, a sequential series of
reactions must be considered to describe adequately treatment performance, even on the
most elementary level.
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 wetland soils and biota, which provide 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 3-9) 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 plants
senesce and decompose.
In FWS constructed wetlands, soil sorption can initially provide phosphorus removal, but
this partly reversible storage eventually becomes buried with organic solids from the
influent TSS and the accumulated detrital materials. For some antecedent soil conditions,
there may be an initial release of P. This new source of P acts to fertilize the wetland, and
some P is utilized in establishing a larger standing crop of vegetation.
Sustainable P removal processes involve accretion and burial of phosphorus in wetland
sediments. Uptake by small organisms, including bacteria, algae, and duckweed, acts as a
rapid-action, partly reversible removal mechanism (Figure 3-10). 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 trapping of
particulate 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.
U.S. EPA Headquarters Library
Mail code 3201
1200 Pennsylvania Avenue NW
Washington DC 20460
3-21
-------
SECTION 3 WETLAND PROCESSES
RGURE 3-9
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 Cell 3 at 0.15 kg/ha-d (Geartieart 1993).
10
9-
8-
i 7-
£ 6-
| 5-
& 4-
o
£ 3-
2-
1 -
Influent
Cells
Cell 3
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)
RGURE 3-10
Conceptual cycling of phosphorus forms in FWS constructed wetlands. SRP: Soluble reactive phosphorus; POP:
paniculate organic phosphorus; TSS-POP: form of POP in terms of a fraction of the total suspended solids.
Emergent plant
Plant filter
3-22
-------
SECTIONS WETLAND PROCESSES
Chemical Oxygen Demand
The chemical oxygen demand (COD) measures the concentration of oxidizable organic
carbon 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 assumed equal to the BOD5 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 through anaerobic decomposition, or under aerobic conditions in periods of
longer than 5 days.
In the Arcata pilot cells, the effluent COD of 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). From the
consistent COD effluent concentrations from the pilot wetland cells even with a ten fold
range in hydraulic/organic loading, it can be concluded that the effluent concentrations are
more closely associated with the amount and type of aquatic plants within the wetland.
The influent COD/BOD ratio averaged 3.7 for the FWS constructed wetland influent
(oxidation pond effluent) and the effluent COD/BOD ratio varied from 3.1 at the beginning
of the study to 28 at the end of the study. Physical and micrbbial processes remove COD
while other processes produce COD in FWS constructed wetlands.
A study was performed at Arcata, California where a pilot cell was loaded at 50 kg/ha-d
for 15 months after which 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 3-11. The COD/BOD ratio was 3.9 during the BOD
loading period.
FIGURE 3-11
BOD and COD effluent concentration before and during tap water loading to Arcata Pilot Project wetland.
~ 100
3
Tap water used as influent second year
-BOD
-COD
1 1 1 1
Fall Winter Spring Summer Fall Winter Spring Summer
Quarter (1980-1982)
3-23
-------
SECTIONS WETLAND PROCESSES
After the addition of freshwater to the system, 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 contributes 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. The sediment
oxygen demand is the result of decomposing detritus generated by carbon fixation in the
wetland, as well as the decomposition of precipitated organic solids that entered with the
wastewater. The 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.
Plant roots also require oxygen, which is normally transported downward through
passages (aerenchyma) in stems and roots. Some surplus of oxygen may be released from
small roots into their immediate environs, but it is quickly consumed in the reduction of
local oxygen demand (Brix 1994a). Wetland soils are typically anoxic or anaerobic (Reddy
and D'Angelo 1994).
Wetland open-area surface waters are aerated by oxygen transfer from the atmosphere,
through 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 unshaded, 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). However, in vegetated regions of the
wetland, shading prevents high algae concentrations and DO levels are typically low near
the surface and anaerobic conditions persist throughout most of the water column. The
effect of vegetation on DO level in the Arcata Enhancement Marsh is shown in Figure 3-12
and Figure 3-13, with the DO in the non-vegetated zones (Figure 3-12) significantly higher
than that in the vegetated zone (Figure 3-13).
Photosynthesis stops at night, and respiration, which consumes oxygen, dominates. The
result is a strong diurnal variation in water column DO for lightly loaded, algae-rich, open
water wetlands.
3-24
-------
SECTIONS WETLAND PROCESSES
FIGURE 3-12
Vertical distribution of DO in a submergent plant zone of the Arcata Enhancement Marsh.
Gearheart-5
Geareheart-3
Gearheart-1
middle
bottom
FIGURE 3-13
Vertical distribution of DO in an emergent plant zone of the Arcata Enhancement Marsh.
Gearheart-6
Gearheart-4
Gearheart-2
middle
bottom
3-25
-------
SECTIONS WETLAND PROCESSES
Hydrogen Ion
Natural wetlands exhibit pH values ranging from slightly basic in alkaline fens (pH = 7 to
8) 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. Open water
zones within wetlands can develop high levels of algal and submergent plant activity,
which in turn create a high pH environment. Data on 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, creating high pH during the day, followed by a nighttime
sag with low pH as respiration replaces photosynthesis.
The 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 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 al., 1983). Listowel constructed
treatment wetland 3 received lagoon water, which periodically exhibited high pH due to
algal activity in the lagoon (Figure 3-14). 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. In the Arcata's Pilot project, it was found that wetland cells
receiving oxidation pond effluent consistently produced a pH between 6.5 and 7.0
(alkalinity of 90 mg/L), regardless of the organic loading or the pH variation in the influent.
This neutral effluent pH was attributed to the organic acids produced in the decomposition
process and to the ammonia and sulfide production and disassociation mat offered some
buffer capacity
3-26
-------
SECTIONS WETLAND PROCESSES
RGURE3-14
Hydrogen ion (pH) buffering in system 3 at Listowel (Herskowitz 1986).
8.5
8-
7.5-
7-
6.5
{DWCMincO'f-^'h.
1-i-CMCMCMCOCOCO
CO (O
Months of Operation
The metals removed from the water column by settling are bound to particles and may
eventually be buried in the anoxic sediments where sulfides are predominant. As shown in
Figure 3-15, in the sediments below open water zones, most metals of concern are bound to
acid volatile sulfides which minimize their biological mobilization. If sediments are
disturbed or resuspended and moved into oxic regions of the wetland, sequestered metals
may revert to dissolved forms and be released. Metals are also incorporated into the
biomass via the primary production processes 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. Metal actions in
sediments below vegetated zones behave similarly, except mat the aerobic zone is
extremely shallow and gaseous discharge is generally impeded at the air/water interface by
duckweed.
3-27
-------
SECTION 3 WETLAND PROCESSES
FIGURE 3-15
Metal sulfide burial processes in a wetland (Meyers 1998, personal communication).
Atmosphere
CO2
I
o.
g
TJ
D)
Air-Water Interface 1
X
CO 2 + Htf
Organic Particles ^ CH2O + O2
2CH2O + NO'3 + 2H*-+. C02
SOZ' + 2CH20 + H* -*• HST-
t
f/2^ * •* HST •< » I
Water-Sediment Interface
Me2+->.
^^ _ 2+ _2*.
Diffti^on^" A^^ + ^ ""^ AfcS
^^^TT ^^ ^^^^^ ^^TT . ^^f
&\^fl^\S ^^^^^ ^j'ftjt i ^^W
4
1 J
-*- CH2O + 02
-+• C02 + H2O
1
+ #20 + W»J ^
h 2//20 + 2C02
"Permanent"
Metal-Sulfide Burial
'2
•3
o
A
,s
'«
^
k
§
£
_o
A
s
§
c
p
Constituent Characteristics
The characteristics of the 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. This is
best illustrated by determining the solids characteristic through a system. Large particulate
organic solids predominate in raw wastewater. These are measured as either flotable solids,
settleable solids or suspended solids, and dissolved solids. Smaller organic colloidal solids
are usually not removed in primary treatment, for example; wetland solubilization can play
a role in the separation, paticulization, and solubilization of these 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 which are
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 CHJ. upon anaerobic decomposition of these settled solids. Colloidal solids
are also released in the decomposition process which includes the heterotrophic bacteria
responsible for the decomposition as well as organism and/or viruses of public health
3-28
-------
SECTION 3 WETLAND PROCESSES
significance. The latter two particle types are adsorbed or impacted 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 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.
RGURE3-16
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).
D Soluble
El Supracolloidal
• Settleable
Ox. Pond Wetland Ox. Pond Wetland Ox. Pond Wetland
BOD BOD TSS TSS COD 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
3-29
-------
SECTIONS WETLAND PROCESSES
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 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
Roating
Species name
Typha spp.
Scirpus spp.
Juncus spp.
Carexspp.
Phragmftes spp.
Potamogeton spp.
Vallisneriaspp.
Ruppia spp.
Nupharspp.
Elodea spp.
Lemna spp.
Eichhomia 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 fern
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
Rooded
Flooded
Flooded
Vegetation Patterns
When constructed or natural FWS wetlands are used for treating wastewater, they become
one of the more highly productive types of wetlands, primarily due to the high level of
nutrients found in the wastewater. Nutrient cycling proceeds at a rapid rate, creating high
density standing crops of vegetation, especially emergent species. The amount of biomass
is both species and climate dependent, as is the stem density. Cattail has a relatively large
3-30
-------
SECTIONS WETLANDPROCESSES
basal diameter, and can occur at about 30 to 50 stems per square meter in treatment
wetlands. In contrast, bulrush have smaller diameter stems, and may occur at hundreds of
stems per square meter. Stem density is additionally constrained by the growth
requirements of the plant in question.
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 aboveground vegetation structure during different seasons. In
northern climates, the end-of-season standing live crop converts to standing dead, and
subsequently to litter and peat. In warmer climates, such phases are shorter and less
pronounced, but there are dormant periods at all latitudes.
Because FWS constructed wetlands need to be planted, they do not initially possess all
vegetative compartments; typically a startup period of many months to a few years is
required for the vegetation compartment to develop fully. The percent coverage of various
plant species during the first two years following planting of the Arcata Pilot Project
wetland is presented in Figure 3-17. The grass and duckweed that were predominant
during the first year were relatively uncommon in the second year as the cattail and
hardstem bulrush grew taller and either shaded or filled in the open water areas. During
this startup period, wetland treatment processes may not be functioning at their full
potential or may be performing at an unsustainably high level of nutrient uptake,
nitrification, etc.
FIGURE 3-17
Coverage of plants during the startup period of the Arcata Pilot Project wetland.
lAlgal Mat
§>
2
O
i
£
S
I
i
6 c -Cl !•: i >. c
SOJ
Study Period
3-31
-------
SECTION 3 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 control the
pollutant removal processes and act as sources and sinks of certain dissolved and
participate 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 algae growth, which can add carbon back to the system via
photosynthesis. The shading of the water surface also moderates the water temperature of
a wetland. A distinguishing characteristic of FWS constructed wetlands is that the water
temperature profile is buffered from the 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 instream
temperature standards.
Well-developed stands of vegetation also reduce the natural reaeration process by
controlling 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 BOD, 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. Low
dissolved oxygen concentrations are mitigated somewhat by the contribution of oxygen to
the water column by common wetland plants.
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 thrive in the unshaded regions of FWS constructed
wetlands. These plants contribute dissolved oxygen directly to the water column while
affording a physical substrate for periphytic 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. Significant solids handling problems can exist with dredged or
harvested aquatic plants. Storage of these materials can result in odors.
The wetland vegetation is also a source of dissolved and particulate material that combines
with the influent wastewater to produce a mixture of biodegradable compounds. A wide
range of heterotrophic and autotrophic organisms degrade these compounds, similar to the
production of BOD via algal growth and degradation in an oxidation pond.
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 provided by plant
3-32
-------
SECTIONS WETLAND PROCESSES
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/mz-d has been measured in wastewater treatment
wetlands at 60% 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, Ml
Pembroke, KY
Dominant Vegetation
Scirpus acutis
Typha latifolia
Scirpus cyperinus
Typha latifolia
Carexspp.
Typha angustifolia
Typha latifolia.
Scirpus validus
Typha angustifolia
Submerged Area
(m2/™2)
7.6
2.6
1.8
1.0
2.4
2.7
2.1
1.2
1.5
Depth
(m)
0.6
0.6
0.25
0.25
Unknown
0.3
0.3
0.2
0.2
Normalized
Submerged Area
(m2/™2)
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, Geameart et al. 1999 (publication in
progress).
Depending on the dominant plant type, plant surface area may 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, and effluent quality may be largely independent of
water depth in treatment wetlands (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 to this finding,
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 0.3 m (Figure 3-18). 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 3-19. For the
Arcata marsh, 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
3-33
-------
SECTIONS WETLAND PROCESSES
surface area for attached growth does increase significantly with depth. As a result, those
water quality constituent removal mechanisms that are dependent on attached growth
surface area are also a function of water depth.
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 headloss and flocculation
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
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 while minimizing incoming radiation
addition. The fact that a sedimentation process coupled with an anaerobic digester and
fixed film reactor is possible in a shallow aquatic system is due to the ecosystem created by
aquatic macrophytes. Without the aquatic macrophytes, the same physical conditions
would result in an oxidation pond producing a large amount of total suspended solids
(algae) in the effluent.
3-34
-------
SECTIONS WETUNDPROCESSES
RGURE3-18
Stem, leaf and litter cumulative surface area for Typha spp. in Houghton Lake discharge zone wetland (Kadlec,
1997).
3.0
a
£
o
o
2.5 -
O Typha latifolia
D Typha angustifolia
o
0.05 0.1 0,15 0.2
Water Depth (m)
0.25
0.3
0.35
HGURE3-19
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).
•5
2
E
E.
O
10.0
9.0-
8.0-
7.0-
6.0-
5.0-
4.0-
3.0-
2.0
1.0
0.0
0,
• Scirpus acutis
• Typha latifolia
H 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 h
0 0.2 0.4 0.6
Stem Depth (m)
0.8
3-35
-------
SECTION 4
Performance Expectations
Approach to Performance Evaluation
The performance and permit compliance of operating FWS constructed wetland treatment
systems reveals the range of effluent quality and the variability of performance possible
with these types of systems. Evident in this analysis is the great variety in the range of
treatment capacities, and thus the loadings, to which FWS treatment wetlands are subjected.
Given the wide variety of settings, design criteria, configuration, and I/O placement,
considerable variation in performance should be expected. This section describes and
compares the performance of operating FWS constructed wetlands for which sufficient data
are available to assess their performance and for which information is available to identify
the factors controlling performance.
In addition to the performance assessment, an analysis of permit compliance for those FWS
constructed wetland 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 of each system during the period
evaluated. For the 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 the 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 in this section 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.
Data Base Evaluation (NADB and TADB)
FWS treatment wetlands were selected for performance analysis based on the data quality,
frequency and availability of criteria listed in Table 2-5. Most of these FWS treatment
wetlands performance data are included in the NADB [Knight et al. 1993]. If additional
information is available for a particular site, it has been included for this technical
assessment. Additionally, data from new systems and recent positions, currently operative
large scale pilot/prefeasibility studies, have been included because they incorporate high
quality data and meet or exceed the criteria in Section 2, Table 2-5.
The NADB carries with serious limitations for evaluation performance. An attempt was
made to select the highest quality data for evaluating FWS wetland performance.
A major limitation is the absence of hydraulic description of the systems influent and
effluent flows. Another major constraint is the lack of intra-cellular and individual cell
influent and effluent water quality data. These two constraints limit the utility of using the
4-1
-------
SECTION 4 PERFORMANCE EXPECTATIONS
NADB for the purpose of performance assessment. Free water surface constructed
wetlands tend to function as a sequence of coupled processes: discrete settling, flocculant
settling, benthal 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. Using the NADB, in its
current form, to assess wetland technology is similar to comparing a wide range of
wastewater treatment system types operating under different influent flow and constituent
characteristics to determine a relationship between flow, wastewater plant area, and
effluent quality. It is also important to know the level of treatment and which unit
processes preceded the wetland and what the regulatory agency expects in terms of
discharge requirements and permit limitations.
Methodology
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). The
Technology Assessment Database (TADB) includes selected systems from the NADB and
additional systems for which operational data are available, that have been completed since
1993. The specific data used from each system in the technology assessment are reported in
Table 4-1. For some systems, data on all water quality constituents were available, while for
other systems only some 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 the data for the systems
listed in Table 4-1, determining the mean influent and effluent concentrations and their
range of values. The mean and range of loadings for each water quality constituent are
given in Table 4-2. This first level of assessment is useful only in the context of
summarizing the range of operating conditions of FWS constructed wetlands and their
range of response in terms of effluent concentration. At this level of analysis, only the wide
range of application 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
at this level of analysis. Each of the factors listed above can significantly affect the effluent
quality of a FWS constructed wetland.
In the second level of performance data analysis, the performance of those systems with the
most extensive monthly influent/effluent data for the constituents of interest are compared.
This level of analysis is displayed in terms of cumulative probability over the period of data
collection. 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 loading 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,
as identified in Table 2-6.
Water Quality Parameter
Wetland System BOD TSS NH.-N TKN NO,-N TN OrgN TP DP FC
Arcata Pilot I 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) 1A
Gustine (89-90)18
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
Ustowel 4
Manila
Minot
Mt. Angel
Orange County
Ouray
Pembroke FWS 2
Poinciana Boot
Reedy Creek WTS1
Reedy Creek OFWTS
Sacramento
Sea Pines Boggy Cut
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 systems analyzed in this assessment (listed in Table 4-1).
Constituent
Biological Oxygen Demand (BOD)
Total Suspended Solids (TSS)
Ammonia (NH4-N)
Total Kjeldahl Nitrogen (TKN)
Nitrate (NO3-N)
Total Nitrogen (TN)
Organic Nitrogen (OrgN)
Total Phosphorus (TP)
Dissolved Phosphorus (DP)
Fecal Coliform (PC) (col/1 OOmL)
Influent (kg/ha-d)
Win Mean Max
0.04
0.07
0.02
0.04
0.05
0.12
0.02
0.01
0.01
31
22
3.5
5.8
0.9
3.0
1.8
1.2
0.6
183
92
16
20
3.5
9.9
5.7
4.4
1.3
Influent (mg/L)
Min Mean Max
1.7
1.0
0.63
1.3
0.31
2.1
0.74
0.27
0.23
1.7
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)
Mln 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
systems in Table 4-1 is 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 increasing loading and increased 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 in Figure 4-1. As
shown in Figure 4-1, considerable effluent variation exists for a given BOD loading. For
example, at a BOD loading of 25 kg/ha-d, the effluent concentrations vary from 9 to 35
mg/L. Considerable variation in effluent quality at the lower BOD loading rates is evident
in Figure 4-1. For example, the effluent BOD varied from 1 to 8 mg/L within the 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
Temporal BOD Performance
A summary of BOD loading versus effluent BOD concentration for the treatment and
enhancement wetlands at Arcata, CA, is given in Figures 4-2 and 4-3, respectively. The
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.
RGURE4-1
Average BOD loading rate versus effluent BOD concentration for TADB sites.
80
I
o
o
m
0)
i
70-
60-
50-
40-
30-
20-
10
0
k+ * *
t**+ «
50 100
BOD Load (kg/ha-d)
150
200
FIGURE 4-2
Monthly influent and effluent BOD values for Arcata's treatment wetland.
120
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, MS are shown in
Figure 4-4. This particular system shows effluent concentrations between 2 and 20 mg/L
4-5
-------
SECTION 4 PERFORMANCE EXPECTATIONS
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 4-3
Monthly influent and effluent BOD values for Arcata's enhancement wetland.
Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
Date
FIGURE 4-4
Influent and effluent monthly BOD cumulative probability values for West Jackson County, MS.
.o
2
§
99.00
95.00
90.00
70.00
50.00
30.00
3
3 10.00
5.00
1.00
W. Jackson County
Influent
Effluent
10 20 30
BOD (mg/L)
40
50
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
4-6
-------
SECTION 4 PERFORMANCE EXPECTATIONS
suspended solids and dissolved BOD from aquatic plant and epiphyte primary production
and decomposition increase the effluent BOD above the influent BOD.
FIGURE 4-5
Influent and effluent monthly BOD for Lakeland, PL
Q
O
GO
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. The effluent BOD is
consistently low, between 2 and 15 mg/L, while the influent varies from 18 to 100 mg/L.
FIGURE 4-6
Influent and effluent monthly BOD cumulative probability for Fort Deposit, AL
Cumulative Probability
-» u 01 -4 coco a
-t UlO OOO OO1 C
3 OO OOO OO C
3 OO OOO OO C
J y"*"*
~ / /
~j ^ Fort Deposit
J / Influent
, , , 1 , , , 1 , . . 1 . . , 1 , , , 1 , , ,
20 40 60
BOD (mg/L)
80
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. The general linear trend between BOD loading and effluent quality
is perhaps more evident than the trend shown for all the TADB systems depicted in Figure
4-1. As might be expected, better relationships between loading and effluent concentrations
were found on a site-by-site basis than observed when lumping data from all the sites
together. The same situation occurred 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 seven 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 also be seen in Figure 4-8, in which the accumulated BOD mass in and out of the
treatment wetland is plotted.
For example, the effluent BOD 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)
The equation fit the 3 years of experimental data for cells with hydraulic residence time
from 6 to 12 days, with an R2 of 0.91.
FIGURE 4-7
Monthly BOD loading rate versus BOD effluent concentration for Arcata Treatment Marsh.
§
o
0
Effluent B
80 -
60 -
40-
20-
o -
A \
*••
*^t
*#
*4*
.
*/
^^
V »
A AA A ^
^ "* * '
•
100 200 300 400 500
BOD Load (kg/ha-d)
600
700
4-8
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-8
Cumulative monthly mass influent and effluent BOD for the Arcata Treatment Wetland.
I
a
o
CO
I
I
o
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
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 BODS limits were more restrictive than winter limits
for five of these systems. In general, FWS constructed wetlands have been very effective at
meeting BODS effluent limits, even with limits as low as 5 mg/L.Only four of the 12 FWS
constructed wetlands had less than 100 percent compliance with BODS 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 for this system was about 40 percent of design flow during that period. The Fort
Deposit, Alabama, constructed treatment wetland exceeded the summer BODS 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 flow. The Norwalk, Iowa, system exceeded its BODS 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 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, resuspension, 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 the data from sites like Orange
County, it can be concluded that wetlands generally will not reduce TSS concentrations
below 3 mg/L, and in cases where the influent TSS is less than 10 mg/L, little if any
additional TSS removal should be expected.
The 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 are 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 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%
for the last six years. An operational change in January of 1991 increased the BOD removal
4-10
-------
SECTION 4 PERFORMANCE EXPECTATIONS
rate, and TSS removal has continued to date. An increase in hydroperiod (0.25 to 0.5
meters) coupled with no alteration in the weir setting over the year has stabilized the
effluent TSS and BOD levels. The effluent TSS concentration does not track the influent
levels with the operational strategies used for the last six years.
FIGURE 4-9
Monthy TSS loading versus effluent TSS concentration for TADB wetland systems.
45
S 30-
CO 05
OT ^&
t 20
5-2
0
50 100 150
Solids Load (kg TSS/ha*d)
200
RGURE4-10
Cumulative probability distribution of monthly influent and effluent TSS concentration for Fort Deposit wetland.
i* 95.00
2 90.00
CO
g 70.00
» 50.00
| 30.00
1 10.00
5 5.00
1.00
(
-
1 r"'~
r /
T- x
i , i
Fort Deposit
Influent
._... fiTffliinrrt
I
) 50 100 150 200
TSS (mg/L)
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-
§100-
& 80-
w 60-
40-
20-
0
+
+
Average
Detention Time
RGURE4-12
Weekly Influent and effluent TSS concentration for Arcata Enhancement Wetland.
60.00
Influent
Effluent
0.00
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 six years (Figure 4-13). 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
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 resuspension. Of the thirteen systems in the NADB, eight had
100 percent compliance with TSS effluent limits.
FIGURE 4-13
Cumulative yearly mass influent and effluent TSS for Arcata Treatment Wetland.
1.6E+06
_ 1.4E+06
1»
£- 1.2E+06 -\
-------
SECTION 4 PERFORMANCE EXPECTATIONS
to their respective loadings. However, the other 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, spatial, denitrification support (alkalinity/carbon, 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.
0.0
£
s
*
2
D.
1
"3
E
O
93.UU
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 nn
—
j
/
/
- /
- /
f
-I ,
,
•
xx
^~~~'~~'
f
f"
^.s
,'
/
j
W. Jackson County
Influent
bniuent
I.I.I,
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 due to organic nitrogen
(org N) in the TSS. Ammonia nitrogen shows considerable variability for a given loading.
At loadings between 2.0 and 3.0 kg/ha-d the effluent ammonia concentrations ranged from
0 to 20 mg/L. The data in the lightly loaded region generally showed low effluent
ammonia levels.
FIGURE 4-15
Ammonia nitrogen loading versus effluent ammonia concentrations for TADB systems.
25
i20^
? 10 H
o>
S 5-
6 8 10 12
NH4 Loading (kg/ha* d)
14
16
18
4-15
-------
SECTION 4 PERFORMANCE EXPECTATIONS
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 four year period,
the 8 cell system 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 are solubilized. This increase mirrors die contribution of dissolved
organic carbon as settled solids decompose in the inlet zone of the wetland. The
contribution of ammonia from decomposition under two different influent ammonia
conditions is shown in Figure 4-18. In the winter, Arcata oxidation pond effluent (influent
to the marsh) has high ammonia levels, typically 12 to 15 mg/L, and limited ammonia is
oxidized within the marsh. During warmer periods, spring and summer, the oxidation
pond contributes little to no ammonia to the wetlands, but decomposition adds 5-6 mg/L
(Gearheart 1989).
FIGURE 4-16
Cumulative probability distribution of monthly influent and effluent ammonia nitrogen from Beaumont, Texas.
^^
ff
2
£
Q.
0)
3§
"5
E
0
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 00
J ' '
— / /
— f i""'
f J
J "" "" ~*
_/ /
I I
(— r*
_ f
Beaumont
Influent
bniuent
I* *
i 1 i 1 i
0.0 5.0 10.0 15.0 20.0
NH4 (mg/L)
4-16
-------
SECTION 4 PERFORMANCE EXPECTATIONS
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
FIGURE 4-18
Ammonia concentration transect through Arcata Pilot Project Wetland
Contact Time (hr)
Total Kjeldahl Nitrogen Performance
Total Kjeldahl nitrogen (TKN) loading versus effluent levels for TADB systems shows
general trends of increased loading producing increased effluent concentrations (Figure 4-
19). 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
4-17
-------
SECTION 4 PERFORMANCE EXPECTATIONS
than 1 mg/L, indicating that in treatment wetlands, the background level of TKN is
attributed to the organic nitrogen. The cumulative probability distribution of the influent
and effluent TKN concentration for the Central Slough wetland is shown in Figure 4-20.
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 4-19
Total Kjeldahl nitrogen loading versus effluent ammonia concentrations for the TADB.
10 15
TKN Load (kg/ha*d)
20
25
FIGURE 4-20
Cumulative probability distribution of monthly influent and effluent TKN from Central Slough, SC.
99.00
95.00
i 90.00
I
o 70.00
e 50.00
1 30.00
3
o 10.00
5.00
1.00
f
Central
Influent
Effluent
I
10 15 20
TKN (mg/L)
25
30
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Nitrate and TIN Performance
As discussed in Section 3, nitrates are also transient nitrogen species in FWS wetlands. The
extent of nitrate removal or production depends on the presence and distribution of oxious
(nitrification produces nitrate from ammonia) and anoxieus (denitrification in which nitrate
is converted to nitrogen gas) regions within a FWS wetland. As shown in Figure 4-21,
essentially no relationship exists between nitrate loading and effluent quality in the NADB
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 4-21
Nitrate nitrogen loading versus effluent nitrate concentrations for the TADB.
~ 3.5 -
*& 3.0 -
E
^ 2.5-
i 2.0-
i«-
* 0.5-
On -
* *
^ *
*
• +
1&.& AA A. *
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.
NO3 Loading (kg/ha*d)
The performance for Orange County, FL, as shown in Figure 4-22 is typical of a lightly
loaded system. The Orange County system has nitrate effluent concentrations less than 0.1
mg/L with mean influent nitrate concentrations of 0.80 mg/L. Iron Bridge operates under
similar conditions with comparable performance, 95% of the effluent nitrate concentrations
are less than 0.1 mg/L with a mean influent concentration of 1.1 mg/L.
The Arcata Enhancement Wetland receives a high loading of total inorganic nitrogen TIN
(sum of nitrite, nitrate and ammonia nitrogen) and shows a TIN reduction from a mean of
26 mg/L in the influent to a mean of 4 mg/L in the effluent (Figure 4-23). The performance
of this system was very consistent. The org N is approximately 15 percent of the total
nitrogen for this system. The majority (95 percent) of the TN is in the form of ammonia and
nitrate nitrogen.
4-19
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-22
Cumulative probability distribution of monthly influent and effluent nitrate concentrations for Orange County, FL
£
A
•g
o.
o
1
1
99.UU
95.00
90.00
70.00
50.00
30.00
10.00
5.00
1 on
•
4 , --—
j ^'
[ /"""
— y Orange County
/" Influent
/ citiuent
I . I , I , I ,
0.0 0.5 1.0 1.5
N(>3 (mg/L)
2.0
2.5
RGURE4-23
Monthly influent and effluent of total inorganic nitrogen (TIN) for the Arcata Enhancement Wetland.
45
§40-
30-
25
20-
15
ID-
S'
Influent
Effluent
/\
/
V / \
^ ,' ^
• / •
t=- OH 1 1 1 1 1 1 1 ^H h
Jan-94 Feb-94 Mar-94 Apr-94 May-94 Jun-94 Jul-94 Aug-94 Sep-94 Oct-94
Date
Total Nitrogen Performance
Total nitrogen, the sum of the organic and inorganic nitrogen, removal from FWS
constructed wetlands shows a correlation between increased loading and increased effluent
concentrations (Figure 4-24). However, within the range of 0.1-6 3 kg/ha-d considerable
variation exists in the effluent concentrations.
4-20
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-24
Total nitrogen loading versus effluent total nitrogen concentrations for TADB wetland systems.
12
E 8
£ 6-|
*tf
| 4-
uj 2 -
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 4-25. Individual
maximum monthly outlet concentrations are more than two times higher than the long-
term average.
FIGURE 4-25
Range of monthly inlet and outlet TN concentrations for cells 1 through 12 at the Iron Bridge FWS wetland near
Orlando, Florida.
8
7
^ 6
4
.§ 5
o
I "
Z
2 3
2
1
Aug-87 Feb-88 Sep-88 Apr-89 Oct-89 May-90 Nov-90 Jun-91 Dec-91 Jul-92 Jan-93
4-21
-------
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. The 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 (full nitrification) 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. The 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.
4-22
-------
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 particulate phosphorus performance from the organic particulate phosphorus
performance.
The relationship between the total P loading and effluent concentration for the TADB data
set is shown in Figure 4-27. Over a range of loading from 0.5 to 4.5 kg/ha-d, total
phosphorus effluent concentration increases with loading. At the 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 4-28). Iron Bridge was the only data set in the TADB
that included dissolved phosphorus. At this site, the mean influent and effluent dissolved
phosphorus values were 0.35 and 0.1 mg/L, respectively.
Temporal Phosphorus Performance
Phosphorus removal in FWS constructed wetlands follows a seasonal pattern in most
temperate climate conditions. The form of the phosphorus, the type and density of the
aquatic plants, the phosphorus loading rate, and the climate determine the amount of
phosphorus removed in FWS constructed wetlands. The aquatic plants serve as seasonal
reservoirs for phosphorus as they take up SRP (soluble reactive phosphorus) during the
growing season. There is a finite amount of SRP that can be incorporated in the aquatic
plants and plankton in the water column. In those temperate climates where senesring of
the aquatic plants occur in the fall, the majority of biologically incorporated phosphorus is
released back to the water column upon decomposition of the particulate organic
phosphorus (POP) and detrital plant material.
Figure 4-26 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 is being
released as the plants stop growing and begin to senesce, in late July. For example, by early
August the effluent from Marsh 3 is 1-2 mg/L higher than the influent to the marsh cell. A
cell which 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 is released from the plant material and the detrital
layer.
4-23
-------
SECTION 4 PERFORMANCE EXPECTATIONS
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 microbial communities dominate. The standing crop in this
particular wetland was approximately 15,000 kg/ha-yr above-ground material.
FIGURE 4-26
Phosphorus pulsing, as illustrated in a pilot cell in Arcata, California. Marsh 1 received tap water (no phosphorus
load), while Marsh 3 received oxidation pond effluent (Gearheart 1993).
Influent
Marshl
-*- Marsh 3
12/1/81
3/2/82
6/2/82
9/1/82
Date
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 the limited
data, it appears that FWS constructed wetlands can comply with very stringent TP effluent
limits.
4-24
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-27
Total phosphorus loading versus effluent phosphorus concentrations for the TADB FWS systems.
£4.5-
g> 4.0 -
£3.5-
i 3.0-
| 2.5-
£2.0-
g 1.5-
S 0.5
0.0-
0
* ;
* * *
* *
b •
0 1.0 2.0 3.0 4.0 5.
Total P Loading (kg/ha»d)
FIGURE 4-28
Cumulative probability distribution of monthly influent and effluent total phosphorus concentrations for Central
Slough, SC.
99.00
95.00
90.00
70.00
50.00
i
2
o.
I
« 30.00
. 3
3 10.00
5.00
1.00
Central
Influent
Effluent
012345678
Total P (mg/L)
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 five 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 Coliform Performance
Database Assessment
As shown in Figure 4-29, 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
(Gersperg 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 4-30. Fecal coliform removal was also found to be
correlated with TSS removal in this system.
FIGURE 4-29
Influent FC versus effluent FC for the TADB systems.
o
10,000
» 1,000
|
g 100
0
UJ
10
10 100 1,000 10,000 100,000 1,000,000
Influent FC (number/100 mL)
4-26
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-30
Cumulative probability distribution of influent and effluent fecal coliform from Arcata Pilot Project Cell 8, CA
(Gearheartetal. 1986).
95.00
s
s
a.
§
90.00
70.00
50.00
1 30.00
E
o
10.00
5.00
Influent
Effluent
i i 1
i i 1 1 ml i i i 1 1 nil
10 100 1000 10000
Arcata Pilot Project Cell 8 FC (cfu/100ml)
Estimates of the internal production of background of 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 4-31). 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 over 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.
In studies performed with MS-2 bacteriophage, virus removal appears to follow the
removal of fecal coliforms (Ives 1988).
4-27
-------
SECTION 4 PERFORMANCE EXPECTATIONS
RGURE4-31
Cumulative probability distribution fecal coliform from Arcata Enhancement Wetland, CA (Gearheart 1998,
unpublished data).
.0
I
E
U
1 10 100 1000
Arcata Enhancement Marsh Effleunt FC (MPN/100mL)
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, exiting organisms 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
this review of 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,
4-28
-------
SECTION 4 PERFORMANCE EXPECTATIONS
molybdenum, nickel, selenium, sulfur, and zinc), these same metals may be toxic at higher
concentrations (Gersberg et al. 1984, Crites et al. 1997). 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 soils
(Crites et al. 1997). A summary of published treatment wetland inlet/outlet metal
concentrations from a variety of sites is presented in Table 4-3. 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.
TABLE 4-3
Metal removal data from free water surface treatment wetlands.
Concentration (ug/L)
Metal
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Iron
Lead
Wetland Type
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
Constructed
. Constructed
Constructed
Constructed
Constructed
Natural
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
In
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
Out
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
Mass Removal
(kg/hayr)
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
Reference
Nolte & Associates 1998
Nolte & Associates 1998
Nolte & Associates 1998
Herskowftz 1986
Nolte & Associates 1998
Herskowitz 1986
Crumpton et al. 1993
Nolte & Associates 1998
Herskowftz 1986
Crumpton et al. 1993
Nolte & Associates 1998
Cooper 1990
Herskowitz 1986
Ewel and Odum 1984
Cooper 1990
Crumpton et al. 1993
Ewel and Odum 1984
Nolte & Associates 1998
4-29
-------
SECTION 4 PERFORMANCE EXPECTATIONS
Concentration (ug/L)
Metal
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
Wetland Type
Natural
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
Natural
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
Natural
In
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
Out
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)
-0.03
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
Cooper 1990
Herskowitz 1986
Ewel and Odum 1984
Cooper 1990
Nolle & Associates 1998
Herskowitz 1986
Crumpton et al. 1993
Nolte & Associates 1998
Cooper 1990
Nolte & Associates 1998
Cooper 1990
Nolte & Associates 1998
Herskowitz 1986
Crumpton et al. 1993
Nolte & Associates 1998
Cooper 1990
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 4-32 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 of 3.5 to 4.0 mg/L occur
4-30
-------
SECTION 4 PERFORMANCE EXPECTATIONS
in the fall and 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. The lower values in the summer are correlated with low decomposition rates (low
recent litter production) and decreased bird activity.
Treatment wetland background concentration ranges can be estimated from systems that
are loaded at a low enough rate to result in asymptotic concentration profile along a
gradient of increasing distance from the inflow (several examples exist in the NADB).
Long-term average annual outflow constituent concentrations for this selected group of
FWS treatment wetlands are summarized in Table 4-4. Wetland systems typically have
background concentrations within the ranges listed in Table 4-5.
TABLE 44
Long-term average annual outflow concentrations for lightly loaded FWS wetlands in the NADB.
System
Eastern Service Area, PL
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 4-5
Expected range of background concentrations for constituents of interest
Constituent
5-day biochemical oxygen demand (BODs)
TSS
Organic and total nitrogen
Fecal coliforms (FC)
Ammonium N
Nitrate N
Total Phosphorus
Unit
mg/L
mg/L
mg/L
MPN/100 mL
mg/L
mg/L
mg/L
Concentration Range
1to10
1to6
1 to 3
50 to 500
less than 0.1
less than 0.1
less than 0.1
4-31
-------
SECTION 4 PERFORMANCE EXPECTATIONS
FIGURE 4-32
Variation in effluent BOD at the Arcata Enhancement Marsh.
July-91 July-92
July-96 July-97
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 the treatment wetland, there is always significant spatial and temporal
variability in wetland water pollutant concentrations. The stochastic character of rainfall
and the periodicity and seasonal fluctuation in ET contribute to much of this variability in
the concentrations in wetland effluents.
4-32
-------
SECTIONS
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 designer can also 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 comprising 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 inevitability of the "scatter" in 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. In addition, compliance of
existing FWS constructed wetland with their permit limits for common target pollutants
and related constituents (BOD, TSS, NH4-N, TN, TP, fecal coliform and DO) are
summarized.
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.
Free water surface constructed wetlands used for wastewater treatment can also provide
considerable benefits beyond water quality improvement and these additional objectives
should be integrated into the feasibility and planning process. Ideally, a master plan
establishing restoration goals for the watershed and its receiving waters will exist and the
benefits of a FWS constructed wetland can be incorporated into this plan.
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
5-1
-------
SECTION 5 FEASBUTY CONSIDERATIONS AND STTE REQUIREMENTS
assessments. The process is similar to the evaluation of conventional wastewater unit
treatment processes because FWS constructed wetlands function similarly to conventional
wastewater treatment processes 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
Stepl - Identify the goals and objectives of the project. In this initial step, the role the
wetland will play in maintaining, restoring, or enhancing the beneficial uses in the
receiving system is established.
Step 2 - Characterize the wastewater 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 will 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 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 to estimate 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.
5-2
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
RGURE5-1
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.
Primary Goal and Objectives 1 r*~
„ Literature *«•
, 1 .
' t
\
—^.Alternative ^_
r*. Treatment 48
A
i
KUDIIC Access
Employment 5a
i vegetation „
(Requirement 68
Not Suitable
^h •Onstrucuo'i
Cost 98
Constituents/Properties *% ^
Fate and Effects/Mass Flow
i.,
Discharoe Standards 3 <—
Permit Requirements
f'bt AmenaDieio WeUarids ^ Unknoi
Appro? Treatment Processes 4
, ,
huncoorv va ue 01 weuands
In this Application 5
i - i
Treatment 5b| Ecological 5c |Hydrolog
|
t
"System Manning
Basis for Design 6
^LjoacSrxji Griteria)
f
— ^^^to ^^^^^ i^M ^
• ^ System Hydrology 6q^ ^
^
Lflnf^^jffcit iirBnrtf^n^^^'
f
^^nftfjOostrsinte D |nn&
?
^ in-Kf hTSyslenri ]^ [.jperatinn/
^| UeCyde 9 |^ costs
4 1
^. Testina,2b
" Studies
\
i
uencn sscaie
Pilot Studv
|
i
cat 5d) Biogeochems
I
Soil Requirements 6c
Maim. ^ Monitonng
Qb Costs
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 habitat, recreation, flood control, and water resource, should be included
in the development of a total cost for the system.
5-3
-------
SECTION 5 FEASBIUTY CONSIDERATIONS AND SITE REQUIREMENTS
Step 10 - Develop 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. The soils which 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 is 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 FWS constructed
wetlands takes from 18 to 36 months because it takes time for the plants to reach
operational density. The discharge permits for a wetland must reflect the lag time
necessary to develop the standing vegetation to support the treatment processes.
Step 12 - Full scale operation requires 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 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 be developed and ready to implement.
Step 13 - Daily monitoring of influent flow and effluent flow and monitoring at minimum
monthly average (weekly samples) BOD, TSS, coliform and others (ammonia, nitrates, etc.).
The vegetation coverage should be monitored annually along with the detrital
accumulation (TSS and plant detritus, and hydroponic 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 conditions monitoring for mosquito larvae and
adults might be required during the mosquito breeding season. Other nuisance organisms
such as nutria, beavers, and muskrats need to be monitored monthly to optimize their effect
on carrying capacity of populations in the system. These organisms might have an 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 have had to
turn to the ample literature on wildlife management to find dues to optimizing wildlife use.
There is a significant amount of published and unpublished literature on habitat richness
and wildlife populations in FWS treatment wetlands, but these data have not yet been
assembled and correlated to wetland design criteria to elucidate relationships. 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). 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 published a book on Created and Natural Wetlands for Controlling Nonpoint Source
Pollution which has chapters on habitat considerations (USEPA, 1993). The habitat quality
5-4
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
of two FWS constructed wetlands was evaluated by the EPA's Environmental Research
Laboratory 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 range of effluent values and meet their permit
limitations. The limitation to using FWS constructed wetlands as a wastewater treatment
system is the background concentration of constituents produced by the loading and
internal wetland processes.
The 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 is often synchronous 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 conditions often occur during periods of high biological ammonia uptake in the
wetland resulting in the highest rates of ammonia removal. A similar situation often
applies for phosphorus. Soluble phosphorus releases occur during the non-growing season
(low plankton standing crop), whereas some fraction of the soluble phosphorus is
incorporated into the plant material during the growing season.
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. The 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 processes, 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.
5-5
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Another critical treatment objective consideration is the discharge point of the effluent.
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. Infiltration
wetlands are designed to combine the horizontal processes in the FWS constructed wetland
with the vertical processes through the sediment and soil to meet water quality objectives
for either groundwater infiltration or surface water discharge. Examples of infiltration FWS
constructed wetlands performance can be found in the Hillsboro, OR, data and the Orange
County Water District, FL, wetland demonstration project.
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). As a result,
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
• Conflicts with other uses could not be mitigated adequately such as adjacent land
use activity, availability and cost of land.
Most natural wetlands are waters of die U.S. These wetlands are either adjacent to other
waters of the U.S., or whose use, degradation, or destruction could affect interstate or
foreign commerce, and as such are afforded protection under the programs of the Clean
Water Act. Additionally, 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)
• National Pollutant Discharge Elimination System (NPDES) Permits (Section 402)
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
• Discharge of Dredge/Fill Permits (Section 404).
For each program area, there are existing specific program regulations, guidance and
procedures. However, the use of wetlands for wastewater management has not been
addressed specifically by any program, and dear guidelines do not exist. Minimum criteria
relating to waters of the U.S. that can be applied to wetlands discharge require that:
• Water quality standards be maintained;
• A minimum of secondary treatment is required for discharges from municipal
treatment facilities to natural wetlands considered waters of the U.S.;
• An NPDES permit is required for each discharger; and
• A 404 Permit would be required for the discharge of dredge and fill material into
wetlands.
The regulations for the U.S. Environmental Protection Agency's (EPA) three major
wastewater management programs (Water Quality Standards, NPDES Permit, 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 nature as a transition between fully terrestrial and fully aquatic
systems. As such, wetlands are often hydroldgically 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 usually are not applicable to wetland wastewater management
systems.
Although wetlands that are waters of the U.S. cannot be classified for "waste transport,"
they can be used in wastewater management as long as established uses are protected.
Many wetland functions and values (e.g., storm buffering, water storage), however, are not
covered by existing use classifications. Additional qualitative or quantitative criteria
addressing wetland characteristics (e.g., hydroperiod, water depth, seasonal influences)
may be necessary and appropriate to protect wetland uses.
Section 402 of the Clean Water Act authorizes EPA and delegated states to administer the
NPDES Permit Program. This program requires a permit for the discharge of pollutants
from any point source into waters of the U.S. Therefore, the discharge to wetlands that are
waters of the U.S. or from treatment wetlands into 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, establishing 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.
5-7
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
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 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 treament facility. 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 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 does not allow any public access except for
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 in where intensive volunteer involvement and management efforts exist.
Public access which does not disturb wildlife is generally considered to be 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 the same array of locations and conditions. However, although these
systems are robust enough to operate under a variety of conditions, consideration must be
given to the effects of local conditions on the performance. When possible, these local
condition effects need to be mitigated by design constraints.
5-8
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
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 accumulation in the wetland can dilute
the effluent concentration and reduce the hydraulic residence time (HRT). High
evapotranspiration rates act in the opposite manner, concentrating the effluent water
quality constituents and increasing the HRT.
In the arid regions of the United States, the monthly net loss of water 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 material and other participates being
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. Operationally, it is also important to
prevent the presence of an air gap between the water surface and the bottom of the ice
layer. An air gap may allow a second layer of ice to form on the new water surface,
complicating the system hydraulics. 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.
5-9
-------
SECTION 5 FEASBIUTY CONSIDERATIONS AND SITE REQUIREMENTS
Engineering Considerations
Pre-Treatment Requirements
FWS constructed wetlands have pre-treatment requirements similar to other biological
wastewater treatment processes. Floatable solids and large settleable solids should be
removed from the influent wastewater. Excessive levels of oil and grease should be
avoided. Specific constituents or constituent loadings that may upset biological processes
should receive pre-treatment. The influent delivery system should be designed to distribute
evenly the incoming solids loads across the wetland cross-section to maximize the
treatment volume available to remove settleable and suspended solids.
Also important to a FWS constructed wetland is the incoming metal concentrations. While
a FWS constructed wetland can remove and immobilize many heavy metals, the same limits
that apply to receiving waters should apply to FWS constructed wetlands influent to
prevent metals accumulation. A source reduction program and an industrial waste
pretreatment ordinance are required if significant metals concentrations are present in a
wastewater.
So/7s, 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 of infiltration losses from wastewater ponds and wetlands range from IxlO"9
to 7 x 10"* 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 the suitability of the soil
to wetlands 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 and plant colonization and growth is more rapid.
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, 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. Clay liners are
5-10
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
generally more effective and a more sustainable component of the wetland structure than
the geosynthetic membranes.
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 distribution system should insure 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 (control 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 insure an
even distribution of the influent through the aquatic plants in the wetland.
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. No 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 inlet weir structures. In
general, weir structures are placed every 8 to 25 m along the effluent receiving zone of a
FWS constructed wetland. Similar to the influent collection/distribution zone, some
systems have effluent collection volumes which then flow to a weir collection/control
structure. This type of system tends to 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/HabitatConsideration
A FWS constructed wetland utilized for treating municipal wastewater can also function as
wildlife habitat, and in some cases 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 certain factors
which encourage and support a wide range of wildlife communities. In the case of FWS
constructed wetlands, the amount of openwater 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 in Arcata, California (1986), which was subsequently used to design
enhancement wetlands, it was shown that having 25 to 70 percent of the water surface
dominated by submergent and floating macrophytes allows optimal water quality and
habitat enhancement objectives to be met.
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
5-11
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
plants, while allowing easy access for aquatic fowl. Islands have been vised effectively in
many wetlands to support resident and migrating bird populations.
Environmental Impact
Planning level considerations for the possible use of FWS constructed wetlands are
important in communicating advantages and disadvantages of these types of 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 the community, 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. This beneficial impact 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 a 400 m from their
breeding area. However, under certain circumstances, wind direction and speed can
disperse mosquitoes and black flies distances farther than 400 m. 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
introducing natural adult mosquito predators (dragonflies and damsel flies, bats, swallows,
frogs), larva predators (mosquito fish, guppies, three spine stickleback, aquatic insect larva),
growth inhibitors (methoprene), and parasites (Bti). Chemical adultcides (pesticides) are
not generally required to manage mosquitoes populations.
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
5-12
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
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 the present time 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.
A FWS constructed wetland that has been designed to provide habitat will attract wildlife.
One major potential impact is the problem of attracting too large a population of migrating
birds. If the wetland support large bird populations and water quality conditions
conducive to pathogen survival exist, then potential disease problems develop (vibrio,
dostridium). The disease potential is particularly a problem for several wetlands in the San
Francisco Bay Area (Hayward Marsh). For example, Hayward Marsh is the only source of
freshwater on the Bay perimeter and attracts large bird populations. Hayward Marsh has
limited vegetation cover which results in a large extent of open areas for resting, watering
and feeding. The potential for introduction and spread of disease in migratory bird
populations can be minimized by diversifying the types of aquatic and riparian habitats and
by having the flexibility to set 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 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. Domestic animal control will frequently require
capturing and destroying these animals.
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
5-13
-------
SECTION 5 FEASBOJTY CONSIDERATIONS AND SITE REQUIREMENTS
as the water/sediment interface. Instead, all reactions are lumped into one overall reaction
rate parameters 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 four 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, NH^
NOy TP and coliform. The four design relationships are summarized in Table 5-1 along
with two new relationships proposed by Gearheart et. al. (1998). Regression equations have
been used to summarize system performance for a wide variety of constituents and physical
parameters. General loading relationships have been used to predict removals for TSS,
BOD, nitrogen, phosphorus 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.
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 will 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:
£-
where: C = pollutant concentration (m/L3),
t = mean hydraulic detention time (t), and
klpp = apparent first-order rate constant (t"1).
5-14
-------
SECTION s FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
TABLE 5-1
Equations used to compute the performance of FWS constructed wetlands
Formula
Type
Definition of Terms
Reed etal. (1995)
exp(-kvt)
Volumetric
Kadlec and Knight (1996)
Areal
Retardation Model (Crites and
Tchobonoglous, 1998, Gearheart, 1999
pn preparation])
Ce=C0e
Sequential Model (Gearheart, 1999 pn
preparation])
Volumetric
BOD only
Volumetric
Two-rates
BOD only
A = fraction of BOD not removed as settleable soids
near headworks of the system, a variable
depending on water quality (decimal fraction)
Aw = total surface area of the wetland (m2)
a = Delaying constant, temperature-dependant
Co = background BOD concentration contributed by
decaying plants (g/m3)
Ce = effluent BOD concentration (g/m3)
Co = influent BOD concentration (g/m3)
Ci = BOD concentration due to solubization of TSS
and residual total BOD (1 to 65 days)
C* = background BOD concentration (g/m3) curve-
fitting parameter
e = porosity of system (decimal fraction)
kAT = temperature corrected first-order areal reaction
rate constant (m/yr)
kvr = temperature-dependent first-order rate
volumetric reaction rate constant (cT1)
Kvi = volumetric based solids/particulate BOD
removal rate
Kv2 = volumetric based dissolved BOD removal rate
- temperature-dependent
L = length of the system parallel to flow path (m)
q = nominal hydraulic loading rate (m/yr)
Q s average flow in the system (m3)
t = theoretical hydraulic detention time (d)
tr = adjusted nominal hydraulic residence time (d)
V = volume of wetland (m3)
W = width of the system (m)
y = average water depth in the system (m)
5-15
-------
_ SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
This differential equation has the exact solution:
C^Coexp-"1"' (5-2)
where: C0 = initial pollutant concentration at t = 0 (m/L3).
The apparent first order reaction rate constant (k^) may be a function of temperature and
values are generally reported at 20° C. The k^ 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),
kj,, = apparent first order reaction rate constant at 20° C (t'1),
0 = 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 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 can be 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, Kaldlec and Knight report that the areal
reaction rate constant is not temperature dependent. In mis 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. Also, 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. Areal loading rates 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 the
data from the FWS wetland systems listed in Table 2-5. The areal loading rates can also be
used to give a preliminary estimate of the FWS wetland surface area required for a given
5-16
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
constituent loading, and can also be used 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. Knowing the areal loading rate,
effluent concentration can be estimated from or compared to the long term average
performance data of full scale operating systems.
TABLE 5-2
Range of area! loading rates for FWS constructed wetlands
Constituent
Hydraulic loading rate (mm/day)
BOD
TSS
TN
NH4
NO3
TP
Typical Influent
Concentration
(mg/L)
10-100
5-60
5-60
2-20
2-20
2-10
1 -10
Target Effluent
Concentration
(mg/L)
5-20
5-20
1 -10
1-10
0.5 - 3
0.5-3
Loading Rates
(kg/had)
10-50
10-60
2-10
2-10
1 -5
1 -5
FIGURE 5-2
Annual average BOD concentration vs. annual average areal BOO loading rate for NADB systems
100 150
BOD Loading (kg/ha*d)
200
5-17
-------
SECTON 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
Design Approach to Sizing
The approach to design of free 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 disadvantages discussed in Chapter 4. The equation
parameters incorporate many factors and should be applied carefully when the setting and
condition are different than those used to generate the parameters. As discussed in Chapter
4, most of the systems in the database were underloaded and therefore are over-designed in
terms of areal requirement. None of the design formulas used to determine wetland areal
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 on die performance of most of the wetlands in the database. As more
experience is gained from multiple celled and/or intermediate sample point systems, a
more useful database for the removal constant values will be developed. At present, the
approach to design should include using one 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 deviation from plug
flow of an existing FWS constructed wetland can be determined through the use of 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 (Figure 5-3). The 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.
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 system with a different degree of actual to theoretical
detention time can lead to serious over or under-designed systems. For example, using the
tracer data shown in Figure 5-3, Treatment Marsh 1 (TM1) at Arcata has an observed
hydraulic detention time of 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 Arcata TM1 theoretical detention time is used for sizing a new system where the
ratio of theoretical to actual hydraulic detention time is higher (say 3.5:1), the new system
will not meet performance expectations due to the relatively shorter actual detention time.
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). The degree of non-ideality should be similar in wetlands with
5-18
-------
SECTION 5 FEASIBILITY CONSIDERATIONS AND SITE REQUIREMENTS
similar geometry, vegetation patterns and hydraulic loadings. The treatment wetland
literature typically provides only apparent plug flow k values.
FIGURE 5-3
Tracer response curve for Sacramento Cell 7 (Nolte and Associates 1997).
0.90
Observed
Rnite Stage Model
6 8 10
Time (days)
12
14
16
5-19
-------
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 potential nuisance conditions, providing adequate 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. The most
useful information has been generated by well documented moderate to high loaded
systems with cell by cell flow and constituent level data.
Databases 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 efforts on some
of the key technology issues that have been identified in this report.
Planning
Multiple Benefits and Public Access
The general public rather than individual landowners primarily receive benefits produced
by wetland areas. The wetland ecology, multiple benefits, and pubic access (birdwatching,
walking, jogging, and picnicking) aspects of FWS constructed wetlands are one 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
-------
SECTIONS LESSONS LEARNED AND RECOMMENDATIONS
(7) landuse set aside for future public use and treatment. These overlays of multiple 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 facility is normally restricted due to the potential
risk associated with wastewater. Many states have specific regulatory constraints
concerning public access to wastewater treatment facilities. Clearly, a paradigm shift must
occur before full acceptance and consideration of FWS constructed wetlands can take place.
Some communities have successfully convinced regulatory agencies to allow full and/or
limited public access to the wetland component of the wastewater treatment facility. Public
access is provided for or encouraged at a number of treatment wetland sites, including
Arcata, Hayward, and Martinez, CA; Cannon Beach, Oregon; Incline Village, Nevada;
Phoenix, AZ; and Iron Bridge and Everglades National Park, Florida. 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.
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 should 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 public 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, CA (East Bay Park District), and the
Arcata Marsh and Wildlife Sanctuary, CA (Friends of the Arcata Marsh). Many other
wetland systems have incorporated information signs into the trail system surrounding the
wetlands for environmental education. Local education institutions typically 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 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.
6-2
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
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. Many of these systems proved
problematic with very low water column dissolved oxygen levels, that resulted in high odor
production and vector problems, primarily mosquitoes.
Many natural wetlands contain a mix of open water and emergent vegetation areas. These
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 will not only provide the same
functions as for natural wetlands, but will also provide increased BOD reduction and
nitrification of wastewater because of the increase in oxygen levels. It is recommended that
a FWS constructed wetland not be vegetated fully, but should include some open water
areas. Open water areas in a FWS constructed wetland will result in a more complex,
dynamic, and self-sustaining wetland ecosystem, that 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 minimimizing 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 availablility 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 and greater for enhancement wetlands. While higher open water is desirable
treatment wetlands can operate successfully at the suggested lower limits if land
availability and construction cost are a major constraint. Generally,enhncement wetlands
will be designed with large open water areas for waterfowl and other wildlife, and open
water to emergent vegetation is usually not a concern.
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 also be used to control
the types of plant communities that can exist in FWS treatment wetlands. The type of
macrophytes (i.e. emergent, submergent, and floating) can be somewhat controlled 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 Srirpus spp. and Typha spp. If the water column depth is between 0.2 to
0.6 m and 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 is included in Section 3.
6-3
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
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. If open water zones are adjacent to the wetland outlet, the wetland may not be able to
consistently meet stringent standards for BOD, TSS, or nutrients. For this reason, it is
recommended that a large vegetated zone exist at the outlet of a FWS constructed wetland.
Site Topography and Soils
Pre-existing topographic, geological, and soil chemistry conditions can greatly affect
wetland cost and performance. Excessive site topography creates large earthwork volumes
for a given wetland area, significantly increasing wetland construction costs. Surface and
subsurface geologic conditions can also increase costs by requiring removal of rock or by
resulting in the need for liner materials to reduce groundwater exchanges. For the most
part, level land with day 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 substantial engineering, earthwork, construction requirements, 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. The substrate for these plants should be agronomic
in nature (e.g. top soil), well loosened, and at least 150 mm deep. If this type of soil exists at
the site it should be scraped off prior to excavation and saved. After the wetland basin,
berms and other earthen structures are constructed, and the liner is installed (if required),
then 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).
Hydrology
Wetland performance can be affected by hydrological factors. Some of these factors can be
taken into account in the design of the wetland. For example, effluent values can be
concentrated due to high ET rates as a result of overdesigning a wetland (areal requirement)
in an area of high ET rates. Maximum ET rates for wetlands are in the range of 2-3 cm/day
(Gearheart et al. 1993). For hydraulic loadings of 7-10 cm/day, the effluent values could be
concentrated by 20 to 40 percent. High precipitation rates can both dilute and reduce the
hydraulic retention times in wetland systems. For example, daily precipitation rates of 10
cm/day could dilute effluent concentrations by half at a hydraulic loading rate of 10
cm/day. These effects of the hydrological cycle on wetland effluent quality are for the most
related to extreme values of ET and precipitation.
6-4
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Infiltration losses can increase the performance of a FWS wetland by increasing the HRT in
the system. Infiltration can also afford some level of treatment as liquid moves through the
anoxic organic layer on the bottom of the wetland.
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) the 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.
Generally, the perforated pipes are connected to a manifold system by a flexible tee joint,
which allows die pipes to be adjusted up or down. In some cases, a wetland with this type
of inlet/outlet structure will cover the perforated pipes with gravel to provide more
uniform distribution or collection of flows. This type of inlet/outlet structure requires some
level of operation and maintenance to ensure equal flow through the pipe, to clean clogged
orifices, and to maintain a level pipe alignment normal to the direction of flow.
For larger wetland systems, multiple weirs or drop boxes are generally used for inlet and
outlet structures. Weirs or drop boxes are generally 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 hydraulically to insure 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.
6-5
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
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 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. Perforated pipe inlet/outlet structures can
be difficult to operate and maintain when they are submerged.
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 then upper layers. An outlet structure design
which 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. Separate flow measuring
devices should be provided on each inlet for multiple wetland cell configurations. Typical
examples of flow measuring devices include simple 90° V-notch or rectangular weirs, and
more sophisticated Parshall flumes. Depending on the size and layout of the wetland, flow
measuring 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 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. Culverts
connecting internally constructed drainage channels can be used under submerged berms
to allow for drainage through a wetland with varying water level elevations.
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Internal Flow Pattern
Due to the low gradients found in FWS wetlands, water generally moves at very low
velocities, approaching conditions of laminar flow (non turbulent flow). 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 continual flow direction or path.
Instead, water flows through a complex maze of submerged vegetation, litter, peat and
other obstructions (e.g. islands); forcing the flowing water to increase and decrease
velocities and continually change direction. Water in open areas located away from
submerged vegetation and/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 minimal water exchange
occurs. The bottom topography may also form deeper pockets or pools, creating more dead
space zones. The result is an internal flow pattern that is intermediate between the ideal
extremes of plug flow and complete mixing.
All of these processes combined can cause water to flow through a FWS wetland in a
shorter time period than defined by the theoretical hydraulic detention time (flowrate
divided by wetland volume). In some extreme cases, such as a poorly designed wetland,
water can flow at high velocities through a small portion of the total wetland volume,
significantly lowering the hydraulic detention time: a phenomena known as short-
circuiting.
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
costly 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 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 headloss through densely vegetated wetland
cells with high aspect ratios. Secondly, the berms must be high enough to account for
6-7
-------
SECTIONS LESSONS LEARNED AND RECOMMENDATIONS
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 m, which provides 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 failures 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, can be
installed during construction, which will minimise mammal burrowing and/or root
penetration. Also, planting the berm using vegetation with a shallow root system can also
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.
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. It is generally recommended for treatment and
water quality purposes that a FWS constructed wetland should consist of a minimum of 2 to
3 cells in series. The effects of headless and inlet/outlet structures need to be considered for
wetland cells in series.
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,
6-8
-------
SECDONS LESSONS LEARNED AND RECOMMENDATIONS
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, any wetland shape needs to be designed and
configured following the general guidelines of this report, and other wetland design
manuals. Design issues such as hydraulic detention time, short circuiting, headless,
inlet/outlet structures, internal configurations, etc., do significantly affect wetland effluent
quality, and some wetland shapes can compound these problems over other shapes. For
example, a long rectangular shaped wetland with a poorly designed inlet/outlet structure
will probably perform better than a square, oval, or kidney shaped wetland with the same
inlet/outlet structure, by reducing the potential for short circuiting.
Sediment Storage Zone at Inlet
A majority of the incoming settleable total suspended solids are removed by discrete
settling in the inlet region of a FWS constructed wetland. Because a significant portion of
the solids can 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 area (settling zone) be provided in the inlet region of a FWS constructed wetland.
The settling zone should consist of an open water area that exists across the entire width of
the wetland inlet. A recommended guideline is to design a settling zone that provides
approximately 1 to 2 days hydraulic detention time at the average wastewater flowrate.
Most 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. It is likely (and
encouraged) that floating aquatic vegetation will exist in the settling zone. Inlet structure
location and design will directly influence inlet velocities in the settling zone. Velocities in
the outlet zone is a function of the cell geometery, 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.
Two periods exist when wetland planting is most successful: fall and spring. In the fall,
tubers or dumps of aquatic emergent vegetation can be planted. Fall planting allows the
6-9
-------
SECTIONS LESSONS LEARNED AND RECOMMENDATIONS
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 dumps 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 soil dumps of 4 to 10 plants into the wetland on 0.6 to 1 m checker board centers.
These dumps include the native soils, along with multiple tubers, which insures 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. The stems
can be cut off in late summer and fall plantings to facilitate transporting and planting of the
dumps of emergent vegetation. The cost of planting dumps is dependent on the distance to
the source of material. It is possible to have a fully functioning wetland in 1 to 2 years after
planting with emergent dumps.
Other planting techniques indude the use of purchased tuber stock and seeds from
commercial sources. The tuber stock are typically planted in a similar fashion to
transplanting seedlings, and depending on the size of the stock can be planted in spring of
fall. For small tuber stock, spring planting is best. The use of seed is the most risky way to
vegetate wetlands. Seed treatment (add, 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 then planting dumps or tubers, but
the success 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 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 dumps 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 algae growth, sediment resuspension, and wildlife activity
in the more open shallow water units. Permit requirements should be written to take this
start-up period into consideration.
Regional sources are usually able to supply relatively small amounts of plant material.
Planting tubers or dumps on 05 m centers, for example, requires approximately 40,000
plants per hectare. Planting on 1 m centers require 10,000 plants per hectare. Given a
planting budget constraint, it is better to place more plants in the last half of the cell, then in
the first half. It is important to insure 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
6-10
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
potential. Hardstem bulrush, for example, affords much greater specific surface area in the
water column than cattails. This specific surface area is a critical growth location for
attached microflora and microorganisms. Cattails are generally larger in diameter than the
bulrush, and have a much larger stem to tuber transition in the water column. 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. Scirpus spp. wetlands
have about 1/3 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 duckweek and sago
pondweed coverage from spring to fall is evident from the data given in Table 6-1.
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
Common cattail
Marsh pennywort 5.6
Sago pondweed
Alkali bulrush
Lesser duckweed
Hardstem bulrush
Common spikerush
Upland grass spp. 30.0
36.2 83.8
6.3 5.5
10.0
NV
11.9
40.0"
5.0
6.0
11.8
77.2
0.7
32.5
10.5
27.0
NV*
30.0*
23.0
4.3
NV
0.8
69.6
2.3
* Duckweed coverage was too low because the wind had pushed it into windrows.
" NV = not visible because of duckweed coverage
Wetland plant growth and survival is also dependent on environmental factors other man
hydroperiod. Two of these factors include soil texture and soil chemistry. Many wetland
plants grow rapidly in soils of sandy to loamy texture. Soils with excessive rock or day
material may retard plant growth and actually result in mortality. Excessively acidic or
basic conditions may limit the availability of the nutrient 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.
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 higher levels in open water areas,
6-11
-------
SECTIONS LESSONS LEARNED AND RECOMMENDATIONS
which supports aquatic organisms such as aquatic insect larva, amphibians and fish. All of
these organisms feed on mosquito larva in the water columns.
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 of the applications require some type of barrier to prevent groundwater
contamination. Under ideal conditions, the wetland site will consist of natural soils with
low permeability that restrict infiltration. However, many wetlands have been constructed
or proposed on sites were soils have high permeability. In these cases, some type of liner or
barrier will 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"6 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. Burrowing mammals such as muskrats, nutria's rats, etc., can
damage liners 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 and
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.
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
6-12
-------
SECTION 6 LESSONS LHAHNED AND RECOMMENDATIONS
constructed wetlands are not mutually exclusive. Typically, one management decision or
action influences other management goals.
Listed below is some management considerations that need to be considered 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)
wildlife management
vector control (mosquitoes)
structural integrity
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 minimum, 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
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 habitat, creation of odors,
attraction of dangerous reptiles (snakes and alligators), potential for accidental drowning
and attractive nuisance for 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. There is inadequate data on
any of these issues to help assess their possible effects on implementation of FWS treatment
wetlands.
6-13
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
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 the
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
FJ?A-funded project in progress.
Vegetation Management Implications
Routine harvesting of vegetation is not necessary for FWS constructed wetlands (Reed et al.,
1995). In many cases, the only routine vegetation 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 % in 12 years with no apparent change in performance. This type of harvesting
may only be required if the vegetation significantly affects removal efficiencies and or
restricts water flow.
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. The strip of harvested vegetation should
be replanted and allowed to grow, before the next adjacent strip of vegetation is harvested.
This process should 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 period of time.
6-14
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
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 revegetate 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. Mosquito populations appear to
be controlled effectively in FWS treatment wetlands by small fish, such as the mosquito fish
(Gambusis 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. Other animals,
such as frogs, birds, and especially bats, are also effective in controlling mosquito
populations. When a FWS constructed wetland is designed to mimic a natural wetland,
mosquito populations can be controlled effectively by the natural wetland ecosystem.
Sprinklers have also 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 in reducing
mosquito larva production in a FWS wetland. However, this technique requires additional
capital investment in the spray equipment, and operation and maintenance of the pump
and sprinkler system.
A bacterial insecticide, Bacillus thuringiensis israeliensis (Bti), has 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. Application of Bti on a six week interval during the mosquito breeding season
appears to be best for meeting a less than 0.1 larva/dip threshold.
Process Control
FWS constructed wetlands have minimal need for active process control. The only two
operational control for FWS wetlands are hydraulic loading and outlet weir level control (if
designed to allow varying hydro periods). 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. Water level increase is limited by the maximum
hydroperiod for emergent plants in the FWS wetland. Generally this maximum depth is 1.0
6-15
-------
SECTION 6 LESSOR LEARNED AND RECOMMENDATIONS
to 1.5 meters. Normal hydroperiods for emergent plants is usually 0.4 to 0.75 meters.
Vegetation and detritus removal is necessary for long term performance of a FWS
constructed wetlands. It appears that 15-20 years is a typical reoccurance internal for some
of the vegetation/detritus removal. To-date no detritus or vegetation has had to be
removed from Arcata's treatment marsh after 15 years of service. Based on studies by
Gearheart, 1986 selected partial removal of vegetation/detritus maintains effluent quality,
while reinstated flow through volume.
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 established, the wetland can be brought on line and
wastewater introduced. After the startup period is over, routine monitoring requirements
will be necessary
The most important monitoring task in the operation of a FWS constructed wetland is
monitoring hydraulic and organic loadings, and discharge from the wetland system
(including the monitoring through individual wetland cells). Such monitoring requires
measuring influent and effluent flowrates, and water depths in each wetland cell. This
information has not been collected routinely from many existing FWS constructed wetland
systems. In fact, many of these systems were not designed to gather this type of data. This
information can be used to develop seasonal strategies, based upon hydraulic and organic
loadings, hydraulic detention times, and areal loadings. Such information can also be used
to assess inlet/outlet distribution and performance.
Influent and effluent water quality constituents should also be measured on a weekly or at a
minimum on a monthly basis. Parameters such as BOD, TSS, pH, nutrients, temperature,
specific conductance, and dissolved oxygen should be monitored. These parameters can be
used to assess wetland performance, and determine constituent loadings. The minimum
monitoring requirements for a FWS constructed wetland are summarized in Table 6-2.
6-16
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
TABLE 6-2
Minimum monitoring requirements for a FWS constructed wetland.
Frequency of monitoring
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, NO,, TP)
Wetland biota monitoring
Vegetation coverage/distribution
Wildlife (nuisance animals)
Vectors (mosquitoes, etc)
Fish
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
Allberms
All inlet/outlet structures
All roads
Each cell
All trails
Access points
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
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
6-17
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
Database Maintenance and Analysis
The initial NADB project was initiated 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 tentatively identified during that effort. It is 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
that are included in the NADB is often suspect and 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 serious limitations of the NADB, it can be used for a variety of purposes. One
use 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. Insufficient information exists to optimize design of free water surface
treatment wetlands. 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. Insuring
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 constructed and evaluated in the NADB are lightly
loaded systems. Some of these systems have influent BOD and TSS values close 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.
6-18
-------
SECTIONS USSONS LEARNED AND RECOMMENDATIONS
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.
These items include design issues covered in Section 5, and information discussed in this
section. 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
migratorial and residential birds (source control), and provides a final
clarification/vegetative filter zone.
5. Design final outlet, collection zones, and inlet/outlet structures to minimize
open water areas, which attract wildlife and promote phytoplankton and
periphyton production.
6. Have a maximum weir overflow length to reduce the velocity field at the inlet
and outlet zones of the wetland.
7. Design a solids settling zone across 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 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
6-19
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
for a cold-climate, cattail-dominated, constructed treatment wetland. Research studies have
been performed on the City of Arcata, CA, research wetland cells since 1980, with two
major data reports 1983 and 1986, and several papers summarizing research activities and
findings. The effects of loading rate, operating depth, and plant types on effluent quality
are summarized in the 1983 report (Gearheart et al., 1983). The 4-year time frame of this
research project and the excellent monitoring and data reports were essential for
maximizing the research benefits this project. It is recommended mat regional sponsors be
solicited to contribute additional data, following the example established by the Arcata and
Listowel project. 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, FL, the Tres Rios research wetlands in
Phoenix, AZ, the Albuquerque, NM, wetland cells, the Arcata, CA, pilot wetland cells, the
Orange County Water District, CA, demonstration wetland, and the Eastern Municipal
Water District pilot wetland cells in Hemet, CA.
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 will assure 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
• 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
6-20
-------
SECTION 6 LESSONS LEARNED AND RECOMMENDATIONS
• 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
• Studies directed at the use of Integrated Pest Management (EPM) 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
6-21
-------
SECTIONS LESSONS LEARNED AND RECOMMENDATIONS
• 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 participate 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.
6-22
-------
Appendix A - References
ADEM (Alabama Department of Environmental Management). 1988. Natural Treatment
Systems for Upgrading Secondary Municipal Wastewater Treatment Facilities. Prepared by
ADEM and the Department of Qvil Engineering, Auburn University, Auburn, AL.
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, CA. 154 pp.
Bastian, RJC 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., D.J. 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 Berkely.
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 Improvementt G.A.
Moshiri, ed. Boca Raton, FL: 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, FL.
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 Artifiriales." 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.
A-1
-------
APPENDIX A-REFERENCES
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, TN: TVA Water Quality
Department.
CH2M HILL. 1992. Carolina Bay Natural Land Treatment Program. Prepared for the Grand
Strand Water & Sewer Authority.
Cooper, P.F., ed. 1990. European Design and Operations Guidelines for Reed Bed Treatment
Systems. Prepared by EC/EWPCA Emergent Hydropohyte Treatment Systems Expert
Contact Group. Swindon, UK: Water Research Centre.
Cooper, P.F. 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. US. 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, TX.
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, FL: 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.
A-2
-------
APPENDIX A-REFERENCES
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, MI: 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.
Epibare, R. E. Heidig, and D. W. Gibson. 1993. "Prevention of Mosquito Production At
An Aquaculture Wastewater Reclamation plant in San Diego, Ca. 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, FL: University of
Florida Press.
Fisk, D.W., ed. 1989. Wetlands: Concerns and Successes. Bethesda, MD: AWRA.
Flach,K.W. 1973. "Land Resources." In Recycling Municipal Sludges and Effluents on
Land. Champaign, University of Illinois.
Florida Administrative Code (FAC) 1989. Florida Department of Environmental
Protection. Tallahassee, FL.
Gearheart, RA. 1992. "Use of Constructed Wetlands to Treat Domestic Wastewater, City
of Arcata, 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, Ca. Oct. 3-7.
Gearheart, 1998, Unpublished data from Arcata Treatment Marshes, California.
Gearheart, RA. 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.
A-3
-------
APPENDIX A-REFERENCES
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, MI.
Gearheart, R.A., R. Norman, and L. Finger. 1993. Review of Wetland Evapotranspiration
Literature. Bullhead Lake, AZ: Bureau of Reclamation
Gearheart, R.A., S. Wilbur, J. Williams, D. Hull, B. Finney, and S. Sundberg. 1983. City
ofArcata Marsh Pilot Project: effluent quality results-system design and management. Final
report. Project No. C-06-2270, State Water Resources Control Board, Sacramento,
California. 127pp.
Gearheart, R.A., Williams, J., Holbrook, H., and Ives, M. 1986. City ofArcata 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 Water Research, 20:363-367.
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, ML 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. dark, and J.E. dark, eds. 1978.Wetland Function and Values: The State
of Our Understanding. Minneapolis, MN: 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, MS.
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, FL.
Hammer, D.E. and R.H. Kadlec. 1986. "A Model for Wetland Surface Water Dynamics."
In Water Resources Research. 22(13): 1951-1958.
A-4
-------
APPENDIX A-REFERENCES
Hammer, D.A., ed. 1989. Constructed Wetlands for Wastewater Treatment: Municipal,
Industrial, and Agricultural. Chelsea, MI: Lewis Publishers.
Herskowitz, J. 1986. Listowel Artificial Marsh Project Report. Prepared for Reasearch
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, FL, 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, 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, CA: Humboldt State University.
Kadlec, R.H. 1994. Personal communication.
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, FL: Lewis Publishers.
Kadlec, R.H. and R.L. Knight 1996. Treatment Wetlands. Boca Raton, FL: 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, CA.
Kays, W.B. 1986. Construction of Linings for Reservoirs, Tanks, and Pollution Control
Facilities. 2nd ed., Wiley-Interscience, New York, NY.
Knight, R.L. Unpublished data from Champion pilot wetlands.
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, MD: American Water Resources
Association.
A-5
-------
APPENDIX A-REFERENCES
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, FL: Lewis Publishers.
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.
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, MI: Lewis
Publishers.
McAllister, Lynne S. 1993. Habitat Quality Assessment of Two Wetland Treatment Systems
in the Arid West - A Pilot Study. Corvallis, OR: U.S. Environmental Protetion 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, WJ. 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, OH: Ohio State University School of
Natural Resources.
Moshiri, G.A., ed. 1993. Constructed Wetlands for Water Quality Improvement. Boca Raton,
FL: 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, 1(513) 569-7630.
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, MA.
A-6
-------
APPENDIX A-REFERENCES
Nolle and Associates. 1997. Sacramento Regional Wastewater Treatment Plant
Demonstration Wetlands Project -1996 Annual Report. Prepared for the Sacramento
Regional County Sanitation District, Elk Grove, CA.
Nolte and Associates. 1998. Constructed Wetlands Demonstration Project. 1997 Annual
Report. Prepared for SacramentoRegional County Sanitation District, Elk Grove, CA.
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, VA, 10-11
June, 1991. In Ecological Engineering. 1(1/2) 170 pp.
Phelps, Earle B. 1944. Stream Sanitation. New York, NY: John Wiley and Sons.
PBSJ (Post, Buckley, Schuh, and Jernigan). 1991. Orlando Easterly Wetlands: 1990
Monitoring Report, to City of Orlando, FL.
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 DA. 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, CA, October 1989.
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 EM. 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 IE, Vol. 2, American Water Works
Association.
A-7
-------
APPENDIX A - REFERENCES
Reed, S.C. 1992. "Subsurface How Constructed Wetlands for Wastewater Treatment -
Status and Prospects." In Wetland Systems in Water Pollution Control, D. 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, MN, June 17-19,1981. Navarre, MN: 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, SC.
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 RAV. 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.
Steele, Kenneth, ed. 1995. Animal Waste and the Land-Water Interface. Boca Raton, FL:
Lewis Publishers.
Stefan, etal. 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, MI: 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, CA.
A-8
-------
APPENDIX A- REFERENCES
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, OR 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.
P.J. Godfrey, E.R. Kaynor, S. Pelczarski, and J. Benf orado, 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, FL: Lewis Publishers.
Tchobanoglous, G., R. Gearheart and R.Crites. 1991. "System Operation and
Monitoring," presented at the Natural/Constructed Wetland Treatment Systems
Workshop, sponsored by the U.S. Environmental Protection Agency, Region Vin and
the Colorado Department of Health, Denver, CO. September 4-6.
Tchobanoglous, G., 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
Atjuaculture Systems for Wastewater Treatment: Seminar Proceedings 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 C.J. 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 of Arcata's Marsh Pilot Project. M.S. Thesis. Arcata, CA: Humboldt State
University.
TVA (Tennessee Valley Authority). 1990. Unpublished data from Benton, KY.
U.S. Environmental Protection Agency. 1983a. The Effects of Wastewater 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 of Wastewater 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.
A-9
-------
APPENDIX A - REFERENCES
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, OH.
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-
001.
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. 1999. Manual for Constructed Wetlands Treatment
of Municipal Wastewater. Cincinnati, OH: Technology Transfer Branch (in preparation).
van Oostrom, A.J. and R.N. Cooper. 1990. "Meat Processing Effluent Treatment in
Surface-Flow and Gravel-Bed Constructed Wastewater Wetlands." In Constructed
Wetlands in Water Pollution Control, P.F. 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, NY: 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, LJ». and M.R. Walker. 1990. City qfGustine Marsh Evaluation Study. Report to the
City of Gustine. Davis, CA: Larry Walker Associates, Inc.
Walton, W.E.; E.T. Schreiber, and MS. 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, AZ.
Water Environment Federation (WEF). 1999. Natural Systems for Wastewater Treatment,
2" ed. (in preparation). Alexandria, VA. Washington, DC: WEF.
Water Pollution Control Federation (WPCF)-1989. Natural Systems for Wastewater
Treatment. Manual of Practice FD-16. Alexandria, VA. 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
A-10
-------
APPENDIX A -REFERENCES
Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and Agricultural,
D.A. Hammer, ed. Chelsea, MI: 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.
Western Consortium for Public Health with EOA, Inc. 1996. Total Resource Recovery
Project, final Report, prepared for the City of San Diego, CA. Oakland, CA.
Wieder, RJC, G. Tchobanoglous, and R.W. Tuttle. 1989. "Preliminary Considerations
Regarding Constructed Wetlands for Wastewater Treatment." In Constructed Wetlands
for Wastewater Treatment: Municipal, Industrial, and Agricultural, D.A. Hammer, ed.
Chelsea, MI: Lewis Publishers.
Wile, L, G. Miller, and S. Black. 1985. "Design and Use of Artificial Wetlands." In
Ecological Considerations in Wetlands Treatment of Municipal Wastewaters, P.J. 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, MI.
Williams, C.R., R.D. Jones, and S A. Wright. 1996. "Mosquito Control in a Constructed
Wetlands." In Proceedings ofWEFTEC 96, Dallas, TX.
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-11
------- |