Professional Documents
Culture Documents
Creek watershed located in Virginia. Two year simulations were completed for five separate
wetland designs, with reductions in percentage of BOD5 (4%-45%), TSS (85%-100%), total
nitrogen (42%-56%), and total phosphorous (38%-57%) comparable to levels reported by
previous research.
iii
Acknowledgements
I would first like to thank my advisor Professor Saied Mostaghimi, who gave me
countless advice and information on how to do proper and professional thesis work. To my
committee members Professor Theo Dillaha and Professor Ray Reneau, your advice and tutelage
were sage and wise. To our Department head, Professor John Perumpral, I would like to give
thanks for helping me adjust to Virginia and making me feel at home. Big thanks to Kevin
Brannan and Shreeram Inmandar, who knew that when I knocked on their door they were going
to be interrupted for an hour. To Theresa Wynn I give thanks for all the help on my model when
you were dog-tired and the advice about life and other important things.
My parents, Priscilla and Collin Wong, were very encouraging and I am glad that they
made me learn how to cook and clean, because Ive seen plenty of very helpless people in
college. My grandmothers, Lin Kim Lennie Lee and Susie Lum have always been supportive
and understanding. To my brothers, Daryl, William, and Alex, I thank you for giving me the
motivation to study because I wanted to get better grades than you. To my aunts and cousins
who have sent me cookies through my college years, my roommates and I thank you. Of course,
even though I am about to graduate that tradition may continue.
I would also like to acknowledge every one in my family and all of my friends. Now that
I have my Masters in Biological Systems Engineering, I hope that you can finally remember
what the title of my degree is.
iv
Table of Contents
I.
INTRODUCTION ............................................................................................................................................. 1
GOAL AND OBJECTIVES ................................................................................................................................. 2
A.
II.
2.
3.
4.
5.
6.
7.
D.
MODEL OVERVIEW:............................................................................................................................... 58
FWS vs. SSF Modeling ........................................................................................................................... 60
MODEL COMPONENTS: ........................................................................................................................ 62
Wetland main program: .......................................................................................................................... 62
Base submodel: ....................................................................................................................................... 63
Hydrology submodel: .............................................................................................................................. 64
Vegetation Submodel: ............................................................................................................................. 69
Nitrogen/Carbon/DO/Bacteria relations:............................................................................................... 72
Carbon submodel: ................................................................................................................................... 73
7.
8.
9.
Nitrogen submodel:................................................................................................................................. 80
Dissolved oxygen submodel: ................................................................................................................... 87
Bacteria submodel: ................................................................................................................................. 91
a. Autotrophic Dynamics...............................................................................................................................................91
b. Heterotrophic bacteria ..............................................................................................................................................93
1.
2.
3.
V.
VI.
VII.
APPENDICES........................................................................................................................................... 166
A.
B.
C.
D.
E.
F.
VIII.
VITA.............................................................................................................................................................248
vi
List of Tables
TABLE 1: NUTRIENT REMOVAL RATES FOR NATURAL WETLAND SITES RECEIVING WASTEWATER INPUTS .9
TABLE 2: GENERAL HYDROPERIOD TOLERANCE RANGES FOR SELECTED WETLAND PLANT
COMMUNITIES..14
TABLE 3: WETLAND DESIGN PARAMETERS ...........................................................15
TABLE 4: A PARTIAL LIST OF PREVIOUS WETLAND MODELS ..................................................30
TABLE 5: MEASURED INFLOW VALUES TO WETLAND CELL 2 IN BENTON, KENTUCKY USED FOR
VALIDATION AND CALIBRATION OF SET-WET MODEL.....
....107
TABLE 6:INPUT PARAMETER VALUES AND SOURCES FOR CALIBRATION AND VALIDATION PERIODS.109
TABLE 7: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED
VALUES FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR
THE CALIBRATED, PREDICTED VALUES..123
TABLE 8: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED
VALUES FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR
THE VALIDATED, PREDICTED VALUES130
TABLE 9: P-VALUES AND RESULTS OF THE WILCOXON SIGNED RANK TEST PROCEDURE FOR
DIFFERENCES BETWEEN THE MEASURED AND VALIDATED, PREDICTED VALUES OF WETLAND
EFFLUENT IN BENTON, KENTUCKY ....133
TABLE 10: LINEAR REGRESSION DATA FOR OBSERVED (Y-AXIS) AND PREDICTED (X-AXIS) WETLAND
EFFLUENT...134
TABLE 11: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 50% CHANGE IN BASE VALUES ...137
TABLE 12: INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL SIMULATION
RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF NOMINI CREEK
WATERSHED146
TABLE 13: INFLUENT, EFFLUENT, AND % REDUCTION OF NUTRIENTS FOR VARIOUS NUTRIENTS FOR
2 YEAR PERIODS OF WETLAND SIMULATIONS FOR QN2 SUBWATERSHED DATA.....150
TABLE 14: RANGE OF POLLUTANT REMOVAL EFFICIENCIES REPORTED FOR CONSTRUCTED WETLAND
SYSTEMS152
TABLE F.1: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 10% CHANGE IN BASE VALUES ...244
TABLE F.2: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 25% CHANGE IN BASE VALUES.....246
vii
List of Figures
FIGURE 1: BREAKDOWN OF NPS POLLUTION EMANATION FOR RIVERS IN VIRGINIA ....................................4
FIGURE 2: CROSS SECTION OF A FWS WETLAND.........................................................................................10
FIGURE 3: CROSS SECTION OF A TYPICAL SUBSURFACE FLOW WETLAND. .................................................11
FIGURE 4: NITROGEN TRANSFORMATIONS IN WETLANDS. .........................................................................17
FIGURE 5: PHOSPHORUS TRANSFORMATIONS IN WETLANDS......................................................................24
FIGURE 6: WETLAND DESCRIPTION FOR SET-WET MODEL WETLANDS.......................................................59
FIGURE 7: RELATIONSHIP OF SET-WET MAIN CODE TO SET-WET SUBMODELS ...........................................63
FIGURE 8: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE
SUBMODEL FOR FWS WETLANDS IN THE SET-WET MODEL ..................................................................66
FIGURE 9: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE
SUBMODEL FOR SSF WETLANDS IN THE SET-WET MODEL ...................................................................67
FIGURE 10: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE VEGETATION CYCLE
SUBMODEL OF THE SET-WET MODEL...................................................................................................71
FIGURE 11: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................74
FIGURE 12: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE
SUBMODEL FOR SSF WETLANDS OF THE SET-WET MODEL...................................................................75
FIGURE 13: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................81
FIGURE 14: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGENCYCLE
SUBMODEL FOR SSF WETLANDS OF THE SET-WET MODEL...................................................................83
FIGURE 15: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL..88
FIGURE 16: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR SSF .............................................................................................................................89
FIGURE 17: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER .........................................................................92
FIGURE 18: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE ..92
FIGURE 19: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER .........................................................................94
FIGURE 20: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE...94
FIGURE 21: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE SEDIMENT CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................97
FIGURE 22: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE PHOSPHOROUS
CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL100
FIGURE 23: FLOW CHART FOR CALLING ORDER OF SET-WET MODEL FROM MAIN CODE THROUGH
SUBROUTINES ....................................................................................................................................103
FIGURE 24A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR
HYDROLOGIC OUTFLOW FROM THE WETLAND ..................................................................................114
FIGURE 24B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR
HYDROLOGIC OUTFLOW FROM THE WETLAND ..................................................................................114
FIGURE 25A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE
WETLAND...115
FIGURE 25B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE
WETLAND...115
viii
FIGURE 26A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................116
FIGURE 26B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................116
FIGURE 27A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................117
FIGURE 27B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................117
FIGURE 28A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................118
FIGURE 28B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................118
FIGURE 29A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................119
FIGURE 29B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................119
FIGURE 30A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................120
FIGURE 30B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................120
FIGURE 31A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................121
FIGURE 31B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................121
FIGURE 32A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................122
FIGURE 32B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................122
FIGURE 33: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND ........................................................................................................125
FIGURE 34: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................126
FIGURE 35: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................126
FIGURE 36: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................127
FIGURE 37: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................127
FIGURE 38: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................128
IGURE
F
39: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................128
FIGURE 40: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................129
FIGURE 41: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................129
FIGURE 42: SIMULATED AND OBSERVED VALUES FOR DISSOLVED PHOSPHOROUS CONCENTRATIONS,
PLOTTED WITH THE DETERMINED LINEAR REGRESSION, AND IDEAL 1:1 LINE. ................................135
FIGURE 43: LOCATION OF THE NOMINI CREEK WATERSHED IN VIRGINIA WITH RESPECT TO
RICHMOND, VA AND THE CHESAPEAKE BAY. ..................................................................................142
FIGURE 44: NOMINI CREEK WATERSHED (QN1) WITH SUBWATERSHED (QN2; SHADED) .......................142
ix
FIGURE 45: LINEAR REGRESSION OF RECORDED TOTAL BOD5 AND HYDROLOGIC INFLOW TO QN2
SUBWATERSHED OF NOMINI CREEK WATERSHED FOR MARCH 26, 1992 TO MARCH 25, 1994.........144
FIGURE E.1.: SIMULATED AND OBSERVED VALUES FOR OUTFLOW, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ........................................240
FIGURE E.2.: SIMULATED AND OBSERVED VALUES FOR AMMONIUM CONCENTRATIONS, PLOTTED
BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL,
AND IDEAL 1:1 LINE. .........................................................................................................................240
FIGURE E.3.: SIMULATED AND OBSERVED VALUES FOR NITRATE CONCENTRATION, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ...........241
FIGURE E.4.: SIMULATED AND OBSERVED VALUES FOR ORGANIC NITROGEN CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................241
FIGURE E.5.: SIMULATED AND OBSERVED VALUES FOR DISSOLVED OXYGEN CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................242
FIGURE E.6.: SIMULATED AND OBSERVED VALUES FOR BOD5 CONCENTRATIONS, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ...........242
FIGURE E.7.: SIMULATED AND OBSERVED VALUES FOR TOTAL SUSPENDED SOLID CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................243
FIGURE E.8.: SIMULATED AND OBSERVED VALUES FOR TOTAL PHOSPHOROUS CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................243
Introduction
Nonpoint Source (NPS) pollution accounts for more than 50% of the nations total water
quality problems (Novotny and Olem, 1981) and over 65% of the total pollutant load to inland
surface waters (USEPA, 1993). Therefore, developing practices for controlling NPS pollution is
of major importance to the health of humans and wildlife. Various types of best management
practices (BMPs) have been developed to address this acute problem, one of which is the use of
wetlands.
Wetlands filter out pollutants and act as sinks for nutrients through physical, chemical
and biochemical processes (Novotny and Olem, 1994). Unfortunately humans have not always
perceived wetlands to be beneficial, and wetlands have been converted to other uses such as
agriculture, mining and development at an alarming rate in the United States. The U.S. Fish and
Wildlife Service estimates that over 50 percent of U.S. wetlands have been destroyed during the
last two centuries (Environmental Law Institute, 1993). Iowa alone has lost 99% of its original
natural marshes, while California has had 91% of its wetlands converted to other uses (Tiner,
1984). Nonetheless, wetlands still comprise over 6% of the entire land based area on the planet
Earth (Novotny and Olem, 1994).
In an effort to restore converted wetlands, many Federal management agencies have
active programs to restore wetlands under their jurisdiction and are encouraging private
landowners and other agencies to do the same (Whitacker and Terrell, 1993). Legislation in
Florida requires any natural wetland removal to be replaced with constructed or restored wetland
sites that are at minimum, two times the amount of lost wetland area.
The use of wetlands to control NPS pollution is a relatively new concept (Raisin and
Mitchell, 1995; Teague et al., 1997). Wetland restoration has taken place in northeastern Illinois
(Hey et al. 1989), and constructed wetlands have been established in Massachusetts (Daukas et
al., 1989) with encouraging results, as significant nutrients and sediments have been retained by
these systems. Research has supported the use of wetlands to treat NPS pollution, but the
question is whether these wetlands are being properly designed to optimize a wetlands ability to
decrease NPS pollution.
The design of wetlands for NPS pollution removal can be optimized with the use of
models that accurately represent wetland systems processes. The ability to optimize wetland
design is beneficial for several reasons. Due to the no net loss policy developed at the
National Wetland Policy Forum in 1987, there should be no removal of wetlands without the
construction of replacement wetlands. There are no laws controlling the quality of these
replacement wetlands however, and many are poorly planned and constructed. To use
replacement wetlands effectively, there is a need to predict how effective these replacements will
be. It is pointless to replace an efficient waste removing wetland with a pond that
accomplishes little. Use of models allows comparisons among various designs, and
consequently improves the effectiveness of replacement wetland with respect to NPS pollution
control efforts.
A.
planning tool for the design of constructed wetlands for effective control and treatment of NPS
pollution. The specific objectives are to:
1) Develop a user-friendly, dynamic, long-term, lumped parameter model for the design of
constructed wetlands to optimize NPS pollution control measures.
2) Evaluate the proposed model by comparing its predictions with field data collected from
representative constructed wetland site(s).
II.
Literature Review
In this section a basic overview of the problems associated with NPS pollution is
presented. It describes various best management practices (BMPs) that are utilized to minimize
NPS pollution, but focuses mainly upon the use of wetlands as a pollution controller. A
description detailing the biological and chemical processes in a wetland is also presented,
followed by a general overview of modeling. The concluding section presents specific
descriptions of models previously developed for constructed wetlands.
A.
NPS Pollution
The definition of nonpoint source pollution is tied to the definition of point source
pollution. Todays statutory definition of point sources of pollution is as follows (Water Quality
Act, Sec.502-14, U.S. Congress, 1987):
The term point source means any discernible, confined, and discrete conveyance, including but
not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling
rock, concentrated animal feeding operation, or vessel to other floating craft from which pollutants
are or may be discharged. This term does not include agricultural stormwater and return flow
from irrigated agriculture.
Nonpoint sources are defined as everything else and can be characterized as follows (Novotny
and Olem, 1994):
Nonpoint discharges enter the receiving water at intermittent intervals in a diffuse manner
and are highly correlated with the occurrence of meteorological events.
Pollution arises from an extended area of land and is in transit overland before it reaches
receiving waters or infiltrates into shallow aquifers.
Unlike point sources where treatment is the most effective method of pollution control,
prevention of NPS pollution focuses on land and runoff management practices.
The extent of NPS pollution is related to certain uncontrollable climatic events (rain, floods,
hurricanes, etc.) as well as geographic and geologic conditions.
There are five major forms of NPS pollution: sediments, nutrients, toxic substances,
pathogens, and oxygen demanding substances. Sediments are soil particles carried by runoff into
streams, bays, lakes, and rivers. Nutrients such as nitrogen (N) and phosphorous (P) are
necessary for plant and animal growth, but their usefulness has a plateau after which all excess is
potentially detrimental to the environment. Toxic substances such as pesticides, formaldehydes,
household chemicals, and motor oil, among others could cause human and wildlife health
problems. Pathogens are disease causing microorganisms that are present in animal and human
waste. Oxygen demanding substances decrease dissolved oxygen (DO) concentrations in aquatic
environments through degradation of organic materials.
There are approximately 45 nonpoint sources of pollution identified in the
Commonwealth of Virginia (DCR, 1996). Rivers receive a vast majority of its NPS pollution
impact from farms (64%), urban areas (6%), forest land (6%), and construction areas (6%), as
presented in Figure 1. All other sources of NPS pollution account for only 18% of the total NPS
pollution impact. Therefore, to maximize the use of limited resources (money and people), NPS
pollution control efforts should be directed towards highly contributive areas such as farms,
forest land, urban areas, and construction areas.
Other Sources
18%
Urban 6%
Forest Land 6%
Farms 64%
Construction 6%
B.
level compatible with water quality goals are termed BMPs (Novotny and Olem, 1994). By
definition, BMPs must be economically and technically feasible and can be categorized as
structural, vegetative, or management. Selection of BMPs is based on either controlling a known
or suspected type of pollution from reaching a particular source, or to prevent pollution from a
category of land-use activity (such as agricultural row crop farming) (Novotny and Olem, 1994).
Various BMPs exist, but selection of a BMP is dependent upon the particular pollutants
and the forms in which they are being transported. The following process can be used when
selecting which particular BMP to implement (USDA, Soil Conservation Service, 1988):
1) Identify the water quality problem (e.g., eutrophication in a lake).
2) Identify the pollutants contributing to the problem and their probable sources.
3) Determine how each pollutant is delivered to the water source (e.g., runoff from a feedlot).
4) Set a reasonable water quality goal for the resources and determine the level of treatment
needed to meet that goal.
5) Evaluate feasible BMPs for water quality effectiveness, effect on groundwater, economic
feasibility, and suitability of the practice to the site.
Various structural BMPs such as terraces and sediment basins have been developed.
Structural BMPs help control NPS pollution with changes to the landscape that either capture
and contain, or slow pollutant movement. A terrace is an earthen embankment, channel or a
combination of ridges and channels constructed across a slope to intercept runoff (Novotny and
Olem, 1994). Terraces decrease the effective slope of the land, which decreases runoff velocity.
A decreased runoff velocity allows soil particles and adsorbed pollutants to settle out, thus
preventing transport from the field to the receiving water source. Terraces can remove up to 95%
of sediment, up to 90% of sediments associated adsorbed nutrients, and between 30% to 70% of
dissolved nutrients (Novotny and Olem, 1994). Sediment basins, sediment control basins, and
detention-retention ponds are earthen embankments that are generally designed as large pools
that control water outflow. These structures retard water flow, allowing heavier particulates to
settle out. Sediment basins can remove 40%-87% of the incoming sediment, up to 30% of the
adsorbed N and 40% of the total P (Novotny and Olem, 1994). Detention-retention ponds are
generally more effective than sediment basins due to the uptake of nutrients by associated
vegetation.
Vegetative BMPs include cropping practices, and vegetative filter strips. Cropping
practices such as conservation tillage and cover crops stress maintenance of vegetative cover
during critical times (heavy rains and strong winds) of NPS pollution generation (Novotny and
Olem, 1994). Conservation tillage is any tillage practice that leaves at least 30% of the soil
surface covered with crop residue after planting. Cover crops are close growing legumes,
grasses, or small grain crops that cover the soil during critical erosion periods for the area. Both
practices reduce NPS pollution by reducing erosion through decreased soil detachment, which
also decreases adsorbed pesticide and nutrient movement. Cover crops also store nutrients that
would otherwise be lost during fallow periods. Conservation tillage has been found to be highly
effective in sediment reduction (30-90%), but has very little effect on controlling soluble
nutrients and pesticides (Novotny and Olem, 1994). Cover crops have been found to be 40-60%
effective in reducing sediment, and 30-50% in removing total P (Novotny and Olem, 1994).
Vegetative filter strips utilize strips of closely growing vegetation, such as bunch grasses, sod, or
small grain crops with the primary objective of water quality protection. They are generally
placed between the source of pollution and the receiving water body. Vegetative filter strips are
designed to slow water velocity from sheet runoff and allow sediment and adsorbed pollutants to
deposit. They are effective in removing sediment and sediment-bound N (about 35-90%) but
much less effective in removing P, fine sediment, and soluble nutrients (Novotny and Olem,
1994).
Management BMPs focus on the use of potential pollutants and include integrated pest
management (IPM) and nutrient management. The combination of practices to control crop
pests (insects, diseases, weeds) while minimizing pollution is termed IPM. It works primarily by
decreasing the amount of pesticide or crop-protection chemical available for runoff by choosing
resistant crop varieties, modified planting dates, and selection of the least toxic, least mobile and
least persistent chemicals (Novotny and Olem, 1994). By decreasing the available chemical
amounts, pollution potential is reduced. The effectiveness of IPM is still being debated, with
some estimates being extremely high and others low. Nutrient management works with the same
Wetlands
Wetlands provide many important ecological functions. Wetlands provide flood storage
and conveyance; stream flow modification; erosion reduction and sediment control; groundwater
recharge/discharge; wildlife habitat; recreation and enjoyment; and pollution control (Novotny
and Olem, 1994). In many aspects, wetlands are excellent BMPs because they provide so many
benefits to the environment and can also be appreciated by wildlife and humans alike. For the
purpose of this study however, the focus will be on wetlands abilities towards pollution control.
Mitsch and Gosselink (1993) described wetlands as the kidneys of the landscape.
Wetlands filter out pollutants and act as sinks for nutrients by purifying the water through
physical (sedimentation, filtration), physical-chemical (adsorption on plants, soil, and organic
substrates), and biochemical processes (biochemical degradation, nitrification, denitrification,
decomposition, and plant uptake) (Novotny and Olem, 1994). The mild slopes of wetlands serve
to slow the velocity of water, which consequently allows sediment and absorbed nutrients to
settle; enhances bacterial die-off due to longer retention times; allows wetland vegetation to
uptake nutrients; and provides a carbon source for microbial action (Novotny and Olem, 1994).
A precise definition which satisfactorily describes all wetland types is not possible due to
the varying types of wetlands (Mitsch and Gosselink, 1993); however, the most comprehensive
definition for wetlands was advanced by the U.S. Fish and Wildlife Service (Cowardin et al.,
1979):
Wetlands are lands transitional between terrestrial and aquatic systems where the water table is usually
at or near the surface or the land is covered by shallow water. Wetlands must have one or more of the
following attributes: (1) at least periodically, the land supports predominately hydrophytes; (2) the
substrate is predominately undrained hydric soils; or (3) the substrate is nonsoil (organic matter) with
water or covered by shallow water at some time during the growing season each year.
As seen by this definition, the hydrology, soil type, and vegetation play significant roles
in determining the functionality and effectiveness of wetlands in retaining pollutants. This
significance will be explored more thoroughly in the section dealing with the design of
constructed wetlands.
1.
Classification
There are various ways to classify wetlands but a consistent method has not been
developed to describe them. The easiest way to differentiate wetlands are to divide wetlands
between natural and constructed types, but beyond this simplistic categorization, a clear cut
classification scheme for wetlands does not exist. The confusion in terminology stems from the
vast diversity of wetland types that exist throughout the world and the lack of direct equivalent
translations between various languages (Mitsch and Gosselink, 1993).
The U.S. Fish and Wildlife Service (Shaw and Fredine, 1956) developed the first
classification scheme in 1956. In this classification, twenty types of wetlands were described
under the following four categories; 1) inland fresh areas, 2) inland saline areas, 3) coastal
freshwater areas, and 4) coastal saline areas. Presently, the classification scheme used in the
United States, as part of the National Wetlands Inventory (Cowardin et al., 1979) is very formal
and all encompassing, but very difficult to use. The classification system is based on a
taxonomic separation scheme, in which all wetland and deep-water habitats are divided into five
systems (marine, estuarine, riverine, lacustrine, and palustrine), and further subdivided into
various subsystems and classes. Mitsch and Gosselink (1993) divide wetland types into two
initial systems (coastal and inland) and then further subdivide these systems into seven separate
categories that encompass most, but not all wetland types.
a.
Natural Wetlands
Natural wetlands originate in geological settings due to water movement and
accumulation. The major geological settings in which wetlands form are areas of 1) slope
discontinuity, 2) topographic depression, 3) stratigraphic features which inhibit infiltration, and
4) permafrost (Widener, 1995). Wetlands that are formed in lowland areas tend to be underlain
by glacial outwash, clay and silt, or alluvial outwash comprised of sand or a mixture of sand and
TABLE 1: NUTRIENT REMOVAL RATES FOR NATURAL WETLAND SITES RECEIVING WASTEWATER INPUTS
Type of
Wetland
Northern Peatland
Bog
Location
Loading
Nutrient Removal
(Population
(percent)
/Hectare) Substrate Total N Total P
Wisconsin
30
98
78
Nontidal freshwater
marsh
Cattail marsh
Lacustrine marsh
Deepwater marsh
Lacustrine marsh
Riverine swamp
Wisconsin
Ontario
Florida
Hungary
South Carolina
17
n/a
99
n/a
n/a
O
n/a
O
n/a
O
80
38
n/a
95
n/a
88
24
97
n/a
50
Louisiana
New Jersey
n/a
198
O
I
51
40
53
0
Chesapeake bay
Georgia
Massachusetts
n/a
Sludge
Sludge
O/I
O/I
O/I
0
50
85
1.5
n/a
n/a
gravel, while wetlands formed in upland areas tend to be underlain by bedrock and glacial till
(Baker, 1973). Mitsch and Gosselink (1986) compiled data on the performance of natural
wetlands for removal of nutrients. As indicated in Table 1, retention of nutrients varies greatly
among different areas. This variability complicates modeling of wetland processes as further
explained in the modeling section.
b.
Constructed Wetlands
Constructed wetlands are man-made systems designed to imitate the functions of natural
wetland systems. There are two fundamental types of constructed wetlands, the free water
surface (FWS) system, and the subsurface flow system (SSF) (Novotny and Olem, 1994). The
FWS system usually consists of basins or channels with a natural or subsurface barrier of clay or
impervious geotechnical lining to prevent seepage (U.S. EPA, 1988). The basins are then
filled with soils to support the accompanying planted vegetation (Figure 2). The water level in a
FWS wetland is above the soil substrate with water flow occurring primarily above ground. A
SSF system consists of a trench or bed underlain with an impermeable layer of clay. The trench
is back filled with media that usually consists of crushed stone, rock fill, gravel, and different
soils. Water flows through the medium and is purified through filtration; absorption by
microorganisms; and adsorption onto soils, organic matters, and plant roots (U.S. EPA, 1988)
(Figure 3). Hence, the performance of the wetland depends on the detention time of incoming
pollutants, the loading rates, the biotic condition within the system, and oxygen availability.
2.
conditions, the potential chemical and biological elements necessary for a properly functioning
wetland exist. Hydrologic conditions can directly modify or change physical and chemical
properties, such as soil salinity, pH, sediment properties, substrate anoxia, and nutrient
availability (Mitsch and Gosselink, 1993). Hydrology is less forgiving than other biological
components, and if improperly accounted for, can cause a constructed wetland to fail.
10
Ultimately, the hydrologic conditions determine success of a wetland system, for it determines
the depth, residence time, and hydroperiod. The hydraulic residence time is the average length
of time a volume of water is detained in a wetland before exiting the system (Novotny and Olem,
1994), and can be estimated as:
HRT =
p *V
Q
(1)
where HRT is the hydraulic residence time for a FWS system (T); p is the porosity ((ratio of
water volume)/(total volume); 0.9-1.0 for FWS); V is the active volume of the wetland (L3); and
Q is the average flow rate (L3/T).
The hydroperiod is the seasonal pattern of water level in a wetland or the water depth
above or below wetland surface level over time (Mitsch and Gosselink, 1993). The hydroperiod
is the dominant factor controlling the plant community composition of wetlands (Duever, 1988).
When hydrologic conditions in a wetland change even slightly, the biota may respond with
massive changes in species richness, composition, and ecosystem productivity.
11
The hydrologic conditions for a wetland are affected by various inputs, outputs and
storage patterns. The general balance between water storage and the outflows and inflows can
best be expressed with the following equation (Kadlec, 1996):
dV
= Qi + Qc + Qsm Qo Qb Q gw + ( P ET ) A
dt
(2)
where A is the wetland surface area (L2); ET is the evapotranspiration rate (L/T); P is the
precipitation rate (L/T); Qb is the bank loss rate (L3/T); Qc is the catchment runoff rate (L3/T);
Qgw is the percolation to groundwater (L3/T); Qi is the input stream flow rate (L3/T); Qo is the
output stream flow rate (L3/T); Qsm is the snowmelt rate (L3/T); t is the time step (T); and V is
the volume of water storage in wetland (L3).
The underlying soil strata play a very important role in wetland development. It
functions both as the medium in which many of the wetland chemical transformations take place
and as the primary storage of available chemicals for wetland vegetation (Mitsch and Gosselink,
1993). The soil is often described as hydric, defined by the U.S. Soil Conservation Service
(1987) as a soil that is saturated, flooded, or ponded long enough during the growing season to
develop anaerobic conditions in the upper part. Wetland soils usually have very high organic
matter content. Highly permeable soils are not suitable for wetlands that are not fed by
groundwater because a high permeability does not allow sufficient water storage for hydric soil
conditions to establish. Permeability must be kept below a certain threshold value, which may
vary according to site-specific and geographic conditions (Novotny and Olem, 1994).
Wetlands plants may be characterized as submersed (i.e., completely submerged),
emergent (i.e., those plants with a root system and stem below the water, but which reaches to
or above the surface), or terrestrial (land based) (Dennison and Berry, 1993). Due to the
anoxia, wide salinity range, and water fluctuations characteristic of an environment that is
neither aquatic nor terrestrial, wetland conditions can be physiologically harsh. The constant
fluctuations in living environment can be taxing to organisms as the changing conditions requires
limited energy supplies to be directed toward growth, and more towards survival practices.
Aquatic organisms can not easily adjust to the periodic drying that occurs in many wetlands and
terrestrial organisms could become stressed by long periods of flooding (Mitsch and Gosselink,
12
1993). To deal with anoxia, wetland plants have developed aerenchyma, or air spaces that run
from the stems to the roots, allowing the diffusion of oxygen from the aerial portions of the
plants to the roots. This adaptation allows plants to generate the required energy needed for
survival (Mitsch and Gosselink, 1993). Other adaptations are used by the species of woody trees
(mangroves, cypress, tupelo, willow and a few others) that have successfully adapted to the
wetland environment. Many woody trees have developed adventitious roots above the anoxic
zone, which allow them to attain the necessary air diffusion requirements for biological
processes. A whole plant strategy adopted by many wetland plants concerns the timing of seed
production and transport. Seed production occurs in the nonflooding season and is accompanied
by either delayed or accelerated flowering (Bloom et al., 1990); the production of buoyant seeds
that float until they lodge on unflooded, higher ground; and seed germination while fruit is still
attached to the trees (Mitsch and Gosselink, 1993). All of these mechanisms increase the
probability of plant survival in a wetland environment. Table 2 lists the general depth and
hydroperiod for selected wetland plant communities.
TABLE 2: GENERAL HYDROPERIOD TOLERANCE RANGES FOR SELECTED WETLAND PLANT COMMUNITIES
Average Water
Depth (m)
Average
Hydroperiod *
0.5 -02
70-100
Submerged
aquatic
0.5-3. 0
80-100
Emergent
marsh
Cattails, pickelrelweed,
bulrush, sedgem maidencane
0.1-1
40-100
Floodplain
0.2-0. 3
10 to 50
Swamp Forest
0.3-1. 0
50-80
0.1-0. 3
50-75
0.1-0. 2
20-50
Wetland Type
Floating Deep
Typical Species
Hyacinths, pennywort
Floating root ed
aquatic
Cypress dome
Wet prairie
* The average % of the year the wetland water surface is above wetland ground level.
Source: Adapted from Novotny and Olem (1994)
13
Constructed wetlands, as compared with natural wetlands, provide a better chance for
management and control of NPS pollution for two reasons; 1) government regulations, and 2)
location. In the Unites States, natural wetlands are considered natural receiving surface-water
bodies like oceans and lakes; hence they are protected from excessive pollution discharges, and
any discharge requires a permit (Novotny and Olem, 1994). There are limits on how much
pollution can be released to a wetland and this consequently reduces its use for water treatment.
Unlike natural wetlands, constructed wetlands do not have these restrictions placed upon them
and can therefore receive higher pollutant loadings for treatment. Consequently, constructed
wetlands are used more often for water quality improvement. In addition, constructed wetlands
can be created wherever the proper hydrologic, chemical and biological requirements can be
established. This allows constructed wetland systems to be more flexible for NPS pollution
treatment for they can be created where water treatment is necessary.
Novotny and Olem (1994) have summarized the basic principles of wetland design:
1. Design the system for minimum maintenance, where the system of plants, animals, microbes,
substrate and water flows are self-maintaining.
2. Design a system that utilizes natural energies, such as gravity flow and the potential energy
of streams.
3. Consider the landscape for system design. Do not overengineer wetland design with
unnatural basin shape, structures, uniform depths, and regular morphology. Try to mimic
nature.
4. Design the entire system as an ecotone, including the use of buffer strips around the site.
5. Consider the surrounding lands and future land-use changes.
6. Hydrologic conditions are paramount. A detailed surface and groundwater study is
necessary.
7. Give the system time to develop. Wetlands are not created overnight.
8. Soil surveys should be conducted, as highly permeable soils do not support wetland systems.
Table 3 lists wetland design parameters for constructed wetlands and compares them to
natural systems.
14
Constructed
SFS
Natural
2 to 4
1.2 to 17
5 to 10
2.5 to 5
5.8 to 8.3
1 to 2
Minimum Size
requirement
(ha/1000m3/d)
Hydraulic Loading
(cm/day)
Maximum water
50
depth (cm)
Bed depth (cm)
50; depend on
ground surface
native vegetation
n/a
30 to 90
n/a
5 to 10
5 to 10
14
2 to 1
n/a
1 to 4
Minimum hydraulic
residence time (days)
Minimum aspect
ratio
Minimum
pretreatment
Primary; secondary
Primary
Primary; secondary;
is optional
nitrification; TP
reduction
Configuration
Multiple Cells in
parallel and series
Distribution
swale, perforated
pipe
Multiple beds in
parallel
Inlet zone (0. 5m)
of large gravel
multiple discharge
series
swale, perforated
pipe
Maximum Loading,
(kg/ha-day )
BOD5
100 to 110
Suspended Solids
up to 150
TKN
80 to 120
10 to 60
10 to 60
0.3 to 0.4
Phosphorous
Additional
Consideration
Mosquito control
with mosquitofish;
remove vegetation
Allow flooding
capability for
weed control
Natural hydroperiod
should be >50%; no
vegetation harvest
3.
surface waters, and substrates of wetlands is very complex. The basic forms of N in soils and
sediments are ammonium ion (NH4+), nitrate (NO3-), organic phytonitrogen in plants and plant
residues, and protein N in living and dead bacteria (Novotny and Olem, 1994). As a negatively
15
charge ion, NO3- is not subject to adsorption by negatively charged soil particles like the
positively charged NH4+ ion, and is thus more mobile in solution. In flooded soils and sediments,
the organic forms of N predominate, while NH4+ is the predominant inorganic N form (Reddy
and Patrick, 1984). Some researchers refer to N content in an area as either Total Kjeldahl N
(TKN) or as total N (TN). Total Kjeldahl N is a measure of reduced N equal to the sum of
organic N and NH4+-N (Kadlec and Knight, 1996). Total N is a measure of all organic and
inorganic forms and is essentially equal to the sum of TKN, NO3- and NO2-N (Kadlec and
Knight, 1996).
Sources of N that contribute to wetland sites include: a) precipitation on the surface of
flooded soils and sediments; b) N fixation in the water and the sediments; c) inputs from surface
and ground water infiltration/percolation; d) application of fertilizers; e) N release during
decomposition of dead aquatic vegetation and animal community inputs; and f) discharge of
waste water effluents (Reddy and Patrick, 1984).
A number of processes can transport or translocate N compounds from one point in a
wetland to another without molecular transformation. These transfer processes are physical in
nature and include: 1) particulate settling and resuspension, 2) diffusion of dissolved forms,
3) litterfall, 4) plant uptake and translocation, 5) NH3 volatilization, 6) sorption of soluble N on
substrates, 7) seed release, and 8) organism migrations (Kadlec and Knight, 1996).
Important processes that transform the basic forms of N in soils and sediments are
presented in Figure 4. These processes are mineralization (ammonification), nitrification,
denitrification, nitrogen (N2) fixation, and assimilation (plant and bacterial uptake).
Understanding the N transfer and transformation processes is very important to the design
of a wetland system. If these processes are not understood, the design of constructed wetland
systems will be negatively affected. The following sections describe the transformations and
transport processes of the N cycle in further detail.
16
a.
i.
Mineralization (ammonification)
Mineralization is the biological transformation of organic N to NH4+ that occurs during
organic matter degradation (Gambrell and Patrick, 1978). Mineralization occurs through
microbial breakdown of organic tissues containing amino acids, hydrolysis of urea and uric acid,
and through excretion of ammonia directly by plants and animals (Kadlec and Knight, 1996).
Mineralization occurs under both anaerobic and aerobic conditions but proceeds at a slower rate
in anaerobic conditions due to the decreased efficiency of heterotrophic bacteria in these
environments (Reddy and Patrick, 1984).
The mineralization rate is affected by temperature, pH, carbon to nitrogen (C:N) ratio of
the substrate, available nutrients in the soil, and soil properties such as texture and structure
(Reddy and Patrick, 1988). The effect of these factors on mineralization in well-drained soils is
fairly well understood, but less is known about their effects in flooded soils. Reddy et al. (1979)
17
concluded that the rate of mineralization doubles with a temperature increase of 10 C, while the
optimum temperature of mineralization was found to be between 40 to 60 C (Reddy and
Patrick, 1984), a rare field condition. The optimal pH range for the mineralization process is
between 6.5 and 8.5 (Reddy and Patrick, 1984), a condition found under most flooded conditions
because the oxidation of organic material produces CO2, which buffers the system.
Measured mineralization rates in natural wetlands range from 0.3 to 35 mg N/m2/d
(annual average of 1.5 g/m2/yr)) in a swamp forest in central Minnesota (Zak and Grigal, 1991),
and 4.3 to 5.9 g/m2/yr in a Minnesota bog (Urban and Eisenrich, 1988). Higher rates were
reported in organic soils in Florida by Reddy (1982), with rates of 41 to 125 g/m2/yr.
ii.
Nitrification
After NH4+ ions are formed through the mineralization process, it can take several
(3)
(4)
18
wetland soils, and 3) the oxidized rhizosphere of plants (Mitsch and Gosselink, 1993).
Nitrification can still occur at low levels of about 0.3 mg/L of DO (Reddy and Patrick, 1984).
iii.
Denitrification
As stated before, NO3- is far more mobile in solution than NH4+. If NO3- is not
(5)
Nitrogen Fixation
Nitrogen fixation is the process by which atmospheric N2 gas diffuses into solution and is
reduced to organic N by autotrophic and heterotrophic bacteria, blue-green algae, and higher
plants (Kadlec and Knight, 1996). N fixation is an adaptive process that provides N for
organisms to grow in conditions that are otherwise depleted of N. N fixation is inhibited by high
concentrations of available N; and is generally not observed in N rich ecosystems.
19
In wetlands, N fixation can occur in overlying waters, in the anaerobic or aerobic soils
layers, in the oxidized rhizosphere of the plants and on the leaves and stem surface of plants
(Mitsch and Gosselink, 1993). Observations of N fixation values vary greatly from differing
wetland sites. Dierberg and Brezonik (1984) observed fixation rates ranging from 1.2 to 19.0
kg/ha/yr in a Florida cypress dome receiving municipal wastewater, but fixation was concluded
to be an insignificant contributor to total N loading.
v.
forms into organic compounds that serve as building blocks for cells and tissues (Kadlec and
Knight, 1996). The two most commonly used forms of N are NH4+-N and NO3--N. NH4+ is
more reduced energetically than NO3-, thus it is the more preferred source for assimilation by
plants and bacteria.
Depending upon the loading rate to the wetland, plant N assimilation can involve a
significant fraction of the total N load. Adcock et al. (1994) determined that a SSF treatment
wetland in Australia had 65% of the N load contained in macrophyte biomass due to its low N
loading rate (25 to 40 g/m2/yr). At sites with higher loading rates, the amount of N lost to
assimilation is a smaller overall percentage.
In temperate climates, plant assimilation is a spring-summer phenomenon. Depending on
location, plant species can either be sinks or sources of N. During the spring and summer when
growth is taking place, plants uptake N, but during the winter months when vegetation dies,
uptake ceases and decomposition occurs.
Microorganisms assimilate nutrients for growth, as NH4+ is readily incorporated into
amino acids by many autotrophs and microbial heterotrophs (Kadlec and Knight, 1996). The
amino acids are transformed into proteins, purines, and pyramidines that are used as energy. The
magnitude of the uptake process has not been quantified for treatment wetlands (Kadlec and
Knight, 1996).
20
b.
forms of N are involved including particulate, dissolved, inorganic and organic. Wetfall (rain or
snow) contributes more than dryfall, and rain contributes more than snow (Kadlec and Knight,
1996).
Nitrogen concentrations in rainfall are highly variable and dependent on atmospheric
conditions, air pollution and geographic location. A typical range of TN concentrations
associated with rainfall is 0.5 to 2.0 mg/L, with about 50% of this present as NO3- and NH3-N
(Kadlec and Knight, 1996). Atmospheric sources are usually negligible contributors to the
overall wetland N budget.
ii.
Ammonia Volatilization
Un-ionized NH3 is relatively volatile and can be removed through mass transfer of NH3
from the water surface to the atmosphere (Kadlec and Knight, 1996). Volatilization has limited
importance for wetlands. Volatilization practically ceases if pH is at or below 7 (Novotny and
Olem, 1994). Typically, volatilization is an insignificant factor when discussing the N cycle in
wetlands. However, in wetlands with a high concentration of NH3-N (20mg/L) and a pH greater
than 8, volatilization can play a significant role (Kadlec and Knight, 1996).
21
iii.
Adsorption
Adsorption is the adherence of chemical ions to the surface of a solid. NH4+ can be
removed from solution through a cation exchange adsorption reaction with inorganic sediments
and detritus (Kadlec and Knight, 1996). The adsorbed NH4+ is loosely bound to the substrate
and can be released when water chemistry conditions change. Most forms of N are very soluble
and do not attach to sediment and other particle types; therefore adsorption plays a limited role in
the overall N balance.
iv.
unavailable for additional nutrient cycling due to burial and peat formation. Burial of N can be
important for light N loading conditions, but becomes insignificant for high N loads (Kadlec and
Knight, 1996). For example, Reddy et al. (1991) reported a N burial rate of 14 to 34 g N/m2/yr
for a lightly fertilized zone of wetland, while the N burial rate was 365 g N/m2/yr in a treatment
wetland.
v.
Biomass Decomposition
The N that is assimilated by macrophytes, microflora, and microfauna is partially
released during decomposition. Turnover times for leaf litter can vary from several months to
over 2 years in colder climates, but decomposition rates during warmer months do not vary much
with geographical conditions (Kadlec and Knight, 1996). The decomposition process is typified
by a rapid initial weight loss that is followed by an exponential loss of the remaining weight to
an irreducible residual which contributes to sediment and soil building (Kadlec and Knight,
1996).
4.
22
sequester this element (Kadlec and Knight, 1996). P is rendered relatively unavailable to
microconsumers and plants when (Mitsch and Gosselink, 1993): a) insoluble phosphates
precipitate with ferric iron, calcium, and aluminum under aerobic conditions; b) chemical
sorption of phosphate to clay particles, organic peat, and other minerals occurs; and c) P
incorporates into the living biomass of wetland biota. Phosphorous is not particularly mobile in
soils and phosphate ions do not readily leach, thus P transport is mostly from plant uptake or
through soil transport (Novotny and Olem, 1994). Figure 5 details the basic transport modes and
reactions for P in a wetland.
Phosphorous occurs as insoluble and soluble complexes in both organic and inorganic
forms in wetland soils. The principal inorganic form is orthophosphate, which includes the ions
PO4-3, HPO4=, and H2PO4- (Mitsch and Gosselink, 1993). The phosphorous cycle is sedimentary
rather than gaseous (i.e., N); therefore, commonly a major portion of a wetlands P content is tied
up in organic peat and litter and in sediment (Mitsch and Gosselink, 1993). Removal efficiencies
range from 0 to 90% (Watson et al., 1989).
a.
vital role in determining P transport and concentrations. Dissolved P in both inorganic and
organic forms usually interacts with suspended and bed sediments. Many of these interactions
are heterogeneous in nature and it is therefore likely that the kinetics of the processes rather than
the chemical equilibrium determine the P division (Grobbelaar, and House, 1995). The nature of
specific interactions for many systems is still unknown, because (Grobbelaar and House, 1995):
The wide range of affinities of P for sediments, combined with the uncertainties in
sedimentary materials composition makes it difficult to identify the key processes,
23
In many wetlands, P cycling tends to follow sediment deposition and resuspension. This
is due to the high sorption rates associated with P. However, there is a common misconception
that wetlands provide P removal only through sorption processes on settling sediments.
Although most sediments do have sorptive capacity for P, this storage will become saturated
under constant P loading rates (Kadlec and Knight, 1996).
b.
Precipitation
Precipitation of P in wetland systems is very complicated and is highly dependent on pH
in the system. At higher pH values, the P precipitates mostly in combination with calcium.
Below a pH of 7, which is characteristic of soils with high clay and organic matter (such as
wetlands), P reacts predominately with the iron and aluminum ions in soils (Novotny and Olem,
1994). Depending on soil pH, the dissolved P concentrations may decrease to values of 0.01
mg/L or less.
24
c.
up by plants during a growing season. All wetland biota undergo a constant cycle of growth,
death and partial decomposition. This results in the decay of plant life and the subsequent
release of assimilated P. Therefore, increases in biomass should not be counted towards the
long-term sustainable P removal capacity of wetlands (Kadlec and Knight, 1996). Although
plants may temporarily remove P from the wetland water and soils, in the long term, it provides
very little retention.
Determining P removal is dependent upon the accretion of biomass residuals and
minerals because this is the only sustainable storage mechanism for P removal (Kadlec and
Knight, 1996). Burial of material removes P from the plant growth/death cycle; therefore the
more plant growth/death cycles, the more chances for burial. Turnover rate is defined as the
number of times the above ground biomass is replaced per year. In northern climates the
turnover rate is lower than in southern climates because southern areas have a longer growing
season (Kadlec and Knight, 1996). Since turnover rate is higher in southern climates, there are
increased chances for accretion and a higher probability of nutrient retention.
5.
Bacteria in Wetlands
Many nutrient transformations in wetlands are due to microbial metabolism and are
directly related to microbial growth (Tanji, 1982). There are theories that state that
decomposition and ammonification rates are linked to microbial energy requirements, the C:N
ratio of the organic matter and the growth rate of microbes in the substrate (Parnas, 1975; Fyock ,
1977; Patrick, 1982). Nitrogen and C are both necessary as a source of energy, while C is
required for building microbial biomass (Parnas, 1975). Growth rates of microbes are a function
of both the environmental conditions and substrate availability.
Energy is obtained by the transference of electrons from an electron donor to an electron
acceptor. Examples of electron donors would be complex organics and NH4+, while oxygen and
NO3- are acceptable electron acceptors (Gidley, 1995). Most of the treatment in wetlands is due
to heterotrophic and autotrophic bacteria (Mitsch and Jorgensen, 1989). Particulate and soluble
labile organics are used as a C source and electron donor by heterotrophic bacteria (Gidley,
25
1995). Equations 3, 4, and 5 show how the microbial transformations generate energy, whose
yield differs for each process. Aerobic degradation of organic materials yields more energy per
mass of electron donor, than either organics degradation or nitrification.
Microbes also utilize N and C to build cell mass. A common formula for microbes is
C5H7O2N (Parnas, 1975). Nitrogen comprises more than 12% of cells, while C accounts for
more than 50% of cell mass. Since microbes use C and N organics; growth of heterotrophs are
influenced by the C:N ratio of the materials they degrade ( Reddy and Patrick, 1983). Aerobic
heterotrophs require organics with a C:N ratio of about 23.5 (Parnas, 1975). Part of the C is used
as an electron donor, while the rest is incorporated into cell mass. Anaerobic decomposition is
not as efficient, therefore more C is required to generate equal amounts of energy. Anaerobic
heterotrophs optimize organic use at a C:N ratio of about 80. Consequently, ammonification is
greater under anaerobic conditions (Reddy and Patrick, 1983). If the C:N ratio is lower than 23.5
or 80, for aerobic and anaerobic conditions respectively, growth will be C limited and the excess
N is wasted as NH4+. If the C:N ratio is higher than these ratios, growth is N limited and the C:N
ratio of the organic materials increase as N is incorporated into cell mass. If the excess N is
NH4+, microbes utilize NH4+ and the C:N ratio remain the same (Parnas, 1975).
Microbial growth rate is determined by the availability of electron donors and acceptors,
the amounts of C and N, and environmental conditions (temperature, pH, space, etc.) (Grady and
Lim, 1980; Reddy and Patrick, 1983). While heterotrophs are responsible for ammonification,
nitrification is inhibited when the DO concentrations drop below 2 mg/L (Bowmer, 1987).
Conversely, the rate of denitrification is reduced in the presence of oxygen.
Optimal conditions for bacterial growth are generally reported as being between a pH of
six and nine, and at temperatures ranging from 15 C and 40C (Fyock, 1977; Reddy and Patrick,
1983; Bruno and Tomasso, 1991). Growth of microbes still occurs outside of these ranges but
the rates are reduced (Broderick et al., 1988). The pH of submerged soils is generally neutral
because the oxidation of organic material produces CO2, which buffers the system (Reddy and
Patrick, 1983). When organic loading is high, heterotrophs out-compete autotrophs and
nitrification is reduced (Grady and Lim, 1980).
The N and C cycles in a wetland are not mutually exclusive as other bacteria may require
the byproducts of one microbial process. For example, heterotrophic bacteria obtain energy from
organics and produce NH4+, which is in turn used by autotrophs as an energy source. The NO3-
26
cycle of growth and nutrient uptake, death, and lastly, decomposition, nutrient release, and soil
accretion (Gidley, 1995). Plant growth and death follow seasonal patterns, while processes such
as decomposition and soil accumulation may take years. Wetlands usually have a seasonal
pattern of nutrient retention in the summer, followed by a nutrient release in the fall and early
spring floods when lower temperatures reduce biological activity (Mitsch and Jorgensen, 1989;
Hantzsche, 1985).
During the summer, vegetation grows and uptakes nutrients. Boyd (1978) found the
mean C and N content were 45% and 1.01%, respectively, for Typha latifolia, and 48% and
1.36% for Juncus effusus, in a natural wetland. Tanner (1996) found the range of N and P
content were 1.5% to 3.2% and 0.13% to 0.34%, respectively, for eight emergent plant species in
a constructed wetland. The chemical composition in wetland biomass varies among plant parts
and changes seasonally; therefore, it is difficult to determine an average nutrient concentration.
Younger plants that have grown in nutrient enriched environments have the highest nutrient
content and aboveground parts usually have higher concentrations than below ground parts
(Heliotis and DeWitt, 1983; Mitsch and Jorgensen, 1989). Kadlec (1989) estimated the typical
N content of wetland biomass to be 2% of the dry weight. The primary productivity in wetlands
is greater than the best agricultural land (Hammer, 1986). The average above ground biomass
growth rates ranged from 0.003 m2-day at Porter Ranch peatland in Michigan (Hammer, 1984) to
0.04 m2-day of total growth of Typha Latifolia in a cultivated northern U.S. peatland on a dry
weight basis (DeBusk and Ryther, 1986).
Nutrients stored in wetland vegetation represent only a small fraction of nutrient input to
wetlands, despite the high primary productivity rates (Mitsch and Gosselink, 1993). The soil
27
nutrient stock is often 1-2 times higher than the biomass nutrient stock (Johnston, 1991).
Regardless, the small amount that is assimilated by biomass is returned to the wetland system at
the end of the growing season as emergent vegetation dies and is decomposed by microbes.
Over the winter and spring the standing dead plants fall to the ground and become litter.
There is an initial period during which nutrients leach from the dead vegetation (Heliotis and
DeWitt, 1983; Kadlec, 1986). Polunin (1982) observed a 13% initial weight loss during a
decomposition study in England, and attributed this degradation to physical factors since the loss
was unaffected by biological inhibitors.
Several researchers have described microbial decomposition as a two-stage process
(Kadlec, 1989b; Kulshrestha and Gopal, 1982). The first stage is an initial rapid weight loss over
the first 30-60 days that is caused by the biological utilization of starches and sugars. The loss of
C from litter is rapid and can exceed initial loss of total organic N and P (Morris and Bowden,
1986). The second stage is an exponential decomposition with the biological release of
additional nutrients. Litter degradation is dependent on particle refractability, microbial growth
and the availability of electron acceptors and nutrients (Heliotis and Dewitt, 1983).
A portion of the biomass is resistant to decomposition and accumulates as soil. Jansson
and Persson (1982) have described this soil organic matter as a heterogeneous mixture. There is
little information regarding the rate of this soil formation, although soil accretion rates for the
Houghton Lake wetland were measured at 2.5 mm/yr using carbon dating (Kadlec, 1989b).
7.
al., 1988). As the interest in wetlands has increased, so has the interest to model the processes
that occur within a wetland. A comprehensive understanding of processes is desired, so that the
construction of replacement wetlands will be equivalent or even better than natural wetlands; be
that for NPS pollution control or the creation of wildlife habitat.
A tremendous amount of wetland modeling has been drawn from previous works on
lakes and other large water bodies. This section will explore general modeling approaches and
techniques, present the specific modeling processes for wetlands used by researchers, and
conclude with descriptions of a few existing wetland models.
28
a.
29
Parameters
Simulated *
Model Purpose #
DY, DI, T
FWS wetlands
R, M
DY, L, T
FWS peatland
DY, DI, T
FWS wetlands
R, M, DE
DY, DI, T
FWS wetlands
R, M
Widener (1995)
DY, DI, T
FWS
H, N
DY, DI, M
FWS coastal
H, P, Sd
S, L, T
FWS
constructed
H, P, Sd
DE
H, N, P
M, DE
H,N, P
M, DE
H, N, DO,
C, BAC, BOD
M, DE
H, N, P
H, N, P,CL
M, DE
PHOSPHOROUS:
DY, L, M
DY, L, M
Gidley (1995)
DY, L, M
DY, DI, M
DY, DI, M
FWS bogs,
swamps
FWS
freshwater
SSF
constructed
FWS
reedswamp
FWS wetlands
30
Ideally, a model would require minimal input; and give output which is very accurate and
detailed, yet this is usually not the case. To design a model that is so esoteric, to the point in
which only a few specialists could use it, would be of limited use. There are models with a wide
spectrum of complexity yet it is best to select a model appropriate to available input data
(Jorgensen, 1995).
b.
There have been various approaches to describe wetland processes. Researchers have usually
focused on one or two of the following specific processes: hydrology, nutrient (N, P, and C)
transport and transformations, and vegetative growth patterns. Described in the following
sections are the more frequently used approaches to modeling specific wetland processes.
i.
Hydrology
31
distributed modeling approach is used, then either a link node method (Walton et al., 1996; Hales
et al., 1991) or finite-difference approach (McDonald and Harbaugh, 1988) is usually used. The
link node method divides the system into a series of finite volumes called nodes where the
stage is defined. Flows are defined along 1-d links between adjacent nodes that can be modified
to represent simple or complex geometry. Essentially the link node method is applied to
continuous segments of the modeled area and uses continuity and momentum equations for
surface water to confirm logic in the system. Detail on the continuity and momentum equations
is included in the surface water description.
The finite-difference scheme or block-centered scheme is used in the groundwater flow
model MODFLOW (Harbaugh and McDonald, 1996). In this approach, the nodes are associated
with known parameter values and equations are solved to obtain unknown values (Fetter, 1994).
Any model that is distributed and represents more than one vertical layer of hydrologic
representation usually uses this scheme.
Many of the parameters that are included in the overall water budget are not modeled but
are included as input values. Amounts of precipitation, river inflow, and river outflow are
measured or estimated from previously recorded data. These inputs are forcing functions or
external variables that influence the state of the system, which in this case is wetland volume.
An input that is not specified to a great extent is the catchment runoff. Many researchers
have used recorded input volumes or first order estimations to determine the amount of nutrients
associated with inflow from catchment runoff (Brown, 1988; Christensen, 1994). This approach
does not allow for considerations of changes in watershed activities that may increase or
decrease nutrient loads.
One method of estimating catchment runoff is the Soil Conservation Service (SCS)
runoff curve model; a method that has been developed to estimate rainfall excess without the
need to compute infiltration and surface storage separately (Novotny and Olem, 1994). In the
SCS method, excess rainfall, Q, depends on the volume of precipitation, PP, and the volume of
total storage, S. S includes both the initial abstraction and total infiltration Ia.
The relationship between the rainfall excess and total rainfall on a daily basis is (SCS,
1968):
32
( PP 0.2 S ) 2
Q=
( PP + 0.8S )
(6)
25400
254
CN
(7)
and
S=
where Q is the rainfall excess (mm); PP is the precipitation rate (mm); S is a storage parameter
(mm); and CN is the runoff curve number. The curve number depends upon a number of factors
that include: land-use type, land-use cover, hydrologic soil group, hydrologic conditions, and
watershed soil moisture conditions. Typical values for the curve number can be found in the
SCS (1968) reference.
Another method for estimating hydrologic and nutrient transport to an existing or
potential wetland is to use an existing comprehensive watershed model such as ANSWERS,
AGNPS, or BASINS for predictions. The output results can be used as input sources to a
wetland model. These predictions should be more accurate than using more generalized
methods; thereby, improving input data and consequently output predictions of the wetland
model.
Surface Water Flow
In a FWS wetland, surface water flow plays an integral part in determining the movement
of dissolved and particulate nutrients out of the system. This is because the velocity at which
water flows through a wetland affects the mixing and settling of the constituents in the water. In
addition, the faster the water exits the system, the less time the processes that transform nutrients
in the system have to perform, thereby decreasing a wetlands nutrient retaining capabilities.
Hydraulic structures, uneven topography, and vegetation cause frictional drag that decreases
surface water velocity. Accounting for the loss in surface water velocity due to obstacles will
allow for a better accounting of predicted residence times and outflows, thereby increasing the
accuracy of predictions for nutrient and sediment effluent.
A simple manner in which to determine the surface water flow velocity is:
33
v=
Q
A
(8)
where v is the water velocity (m/d), Q is the discharge from the wetland (m3/d), and A is the
cross sectional area of the wetland (m2). As Equation 8 shows, if the water velocity in the
system is over or under predicted, the determined outflow of the system is affected. Since a
lumped model assesses the system as a whole, the flow velocity rate is the same throughout the
entire wetland.
The modeling of surface water flow in a distributed system is more complicated. The
nodes in the system allow different velocities to be determined through the wetland based on the
site data. To assess logic and reasoning within a distributed modeling system, the continuity and
momentum equations are used to evaluate water movement. Walton et al. (1996) used the
equations for conservation of momentum and continuity:
y
Q ( Q 2 / A
+
+ gA = gA( S o S f )
t
x
x
(9)
and
N
V
= Q n + Qi
t n =1
(10)
where Q is the flow rate (L3/T); is a momentum correction factor; A is the cross sectional area
of the link (L2); g is the acceleration due to gravity (L/T2); y is water depth (L); So is the bed
slope; Sf is the friction slope; V is the nodal volume (L3); Qn is the flow rate in link n; N is the
number of channels entering nodes; t is time (T); x is the longitudinal distance along a link (L);
and Qi is all other inflows such as precipitation, bank inflows, etc. (L3/T). In each adjoining
node, the momentum and conservation equations must be satisfied. This process confirms that
water is flowing smoothly in the wetland model.
Kadlec (1990) presented a system that is based on a friction rate law equation:
34
v=
K *d * S
d'
(11)
where v is the water velocity through the wetland bed; K is a premultiplier constant; d is the
average wetland depth (L); S is the slope; is a depth exponent; is a slope exponent; and d is
the average depth of free water. This rate law is combined with mass conservation to determine
wetland surface water flow. The model takes into account the movement of surface water in
response to gradient and vegetation flow resistance.
Evapotranspiration
Evapotranspiration (ET) is the term used for the water lost to the atmosphere from both
evaporation and plant transpiration. It is difficult to differentiate the two processes when
measuring water loss from a system, thus they are combined for many wetland modeling
situations. It is mainly a function of climatic variables such as solar radiation, temperature,
vapor pressure deficit, and wind speed (Abtew, 1996).
Like any other system, there are simple and more complicated approaches to modeling
ET. One empirical method is the pan evaporation method in which measured values of
evaporation from a standard pan are multiplied by a coefficient to represent actual ET in a
system (Novotny and Olem, 1994).
Pierce (1992) recommends the use of the Thornthwaite method for calculating potential
ET. The Thornthwaite method assumes that the soil moisture is not limiting and that air
temperature is the primary controlling factor of ET. The Thornthwaite method equation for
potential evapotranspiration (PET) follows the form of (Thornthwaite and Mather, 1955):
D
10 * T a
PET = 2.54 * ( ) * (
)
I
12
(12)
where PET is the potential ET in cm/day; D is the number of daylight hours; T is the average
monthly air temperature (C); I is the heat index; and a is a coefficient. The heat index, I, and
coefficient, a, are calculated as follows:
35
ij = (
Tj
5
)1.514
j = 112,
(13)
I = i j , and
(14)
(15)
( R n G ) + * c p * (e A e d ) *
ET =
1
r
r
+ (1 + c )
r
(16)
where ET is the evapotranspiration (kw/m2); Rn is the net radiation flux at surface (kw/m2); G is
the water heat flux (kw/m2); is the atmospheric density (kw/m3); cp is the specific heat of moist
air (KJ/ kg C); (ea-ed) is the vapor pressure deficit (kPa); rc is the canopy resistance (s/m); ra is
the aerodynamic resistance (s/m); is the slope of vapor pressure curve (kPa/C); and is a
psychometric constant (kPa/C). The Penman Combination method is of the form (Allen et al.,
1989):
ET =
1 ( R n G ) + * 6.43(a w + bw u 2 )(e a e d )
(17)
where ET is the grass or alfalfa reference ET (m/d); u2 is the wind speed at 2 m height (m/s); aw
is an empirical coefficient for the study area; bw is an empirical coefficient for the study area; Rn
36
is the net radiation flux at surface (kw/m2); G is the water heat flux (kw/m2); (ea-ed) is the vapor
pressure deficit (kPa); and is a psychometric constant (kPa/C). Both approaches require large
data input, thus their use is limited as such data are not readily available.
Groundwater Flow
Groundwater flow can convey water and potentially nutrients to a wetland site. Some
wetlands are entirely fed and rely on groundwater flow for most hydrologic input. A few models
of constructed wetlands assume that groundwater inflow and outflow are negligible, compared to
the through-flow water volume, and are therefore ignored (Niswander and Mitsch, 1993; Mitsch
and Reeder, 1991; Hammer and Kadlec, 1986). Other researchers have assumed a constant rate
of infiltration to groundwater (Christensen et al., 1994; Brown, 1988).
The modeling of horizontal groundwater flow usually entails the use of Darcys law for
saturated groundwater flow (Fetter, 1994):
Qh = K h * A * (
dh
)
dl
(18)
where Qh is the horizontal flow (L3/T); Kh is the horizontal hydraulic conductivity (L/T); A is
the cross sectional area (L2); dh is the change in hydraulic head (L); and dl is the change in
distance (L). Vertical groundwater flow can be determined in the same manner with Kh being
replaced with Kv, the vertical hydraulic conductivity.
With these two calculations, the total potentiometric head may be calculated at each
subsurface node with the continuity equation (Equation 9). By taking the calculated heads at
various points in a wetland, the release of wetland water (infiltration) or input of groundwater
(percolation) may be deduced. Walton et al. (1996) and Dorge (1994) are among a few
researchers who have used this approach in their modeling.
ii.
Nitrogen
The modeling of N processes usually include nitrification, denitrification, mineralization,
37
deposition, and adsorption are either concluded to be negligible or are only modeled for specific
situations. The approaches to modeling the differing processes are very similar. There are
certain approaches that are used with their only differences being in parameter declaration and
values.
A commonly used empirical model for wetland design assumes plug flow hydraulics and
first order BOD removal kinetics with an adjustment for temperature effects (Gidley, 1995). The
equation follows the form:
C e = C o e KT t
(19)
where Co is the influent concentration (M/L3); Ce is the effluent concentration (M/L3); KT is the
rate constant (day1); and t is the hydraulic retention time (days). The rate constant is a function
of temperature and is determined by:
K T = K 20 T 20
(20)
where K20 is the rate constant at 20 degrees Celsius (day-1), and T is the design or operating
temperature (C).
There are theoretical models that emulate the transformations within a wetland. If the
reaction is independent of the substrate concentration, it is considered to be zero order. The rate
equation for this type of reaction is:
d [S ]
=k
dt
(21)
where S is the substrate (M/L3), k is the rate constant (M/L3/T), and t is time.
The first order equation is similar to the zero order form except it accounts for substrate
concentration. This equation is of the form:
d [S ]
= K [S ]
dt
(22)
38
M=
M max S
Ks + S
(23)
where M is the growth rate (L3/MT); Mmax is the maximum growth rate (M/L3T); Ks is the
substrate concentration when M= Mmax/2 half-saturation constant (M/L3); and S is the substrate
concentration (M/L3).
Very similar to the Monod kinetics approach is the Michaelis-Menten kinetics equation
(Dorge, 1994):
d [S ] Vmax [S ]
=
dt
K m + [S ]
(24)
where S is the substrate (M/L3); Vmax is the maximum rate of reaction (M/L3T); and Km is the
half-saturation constant (M/L3). The two approaches bear a striking resemblance, and are based
on the concept of conversion of a single substrate to a single product. They are fairly
interchangeable and are basically a matter of semantics; however, the Monod equation was
initially developed for bacterial growth, while the Michaelis-Menten was developed for enzyme
reactions.
Previous models usually ignore diffusion of dissolved N. There are two common choices
for modeling diffusion; Ficks law and a law with no formal name that involves a mass transfer
coefficient.
Ficks law is the most commonly cited in modeling diffusion. There are many forms of
Ficks law, the basic form is (Cussler, 1997):
39
J 1 = Aj1 = AD
c1
z
(25)
where J1 is the one dimensional flux of a constituent (M); A is the cross sectional area across
which diffusion occurs(L2); j1 is the flux per unit area (M/L2 ); c1 is the concentration of the
constituent (M/L3); z is the distance (L); and D is the diffusion coefficient (L/T).
Mass transfer is an approximate engineering idea that often gives a simpler description of
diffusion. Analyzing diffusion with mass transfer coefficients requires the assumption that
changes in concentration are limited to the small part of the systems volumes connecting
boundaries (Cussler, 1997). The common form used to describe mass transfer is (Cussler, 1997):
N 1 = k (c1i c1 )
(26)
where N1 is the flux at the interface (M/L2T); k is the mass transfer coefficient (L/T); c1 is the
concentration in bulk solution 1 (M/L3); and c1i is the concentration at interface (M/L3). The
concentration of c1i is in the bulk concentration of c1 but is usually in equilibrium with the
concentration across the interface in a second adjacent fluid. The product of the flux and cross
sectional area will result in the amount of mass that has been transferred.
As stated earlier, nitrification, denitrification, mineralization, assimilation and
immobilization are modeled with similar approaches, the only difference being parameter
declarations and the type of substrate being examined. For example, if nitrification was being
modeled with a first order equation, the substrate substance would be NH4+ with an associated
rate constant. If the process examination concerns denitrification, the substrate examined would
be NO3-. For mineralization, immobilization, and assimilation, the respective substrates to be
examined would be organic N, NH3, and both NH3 and NO3-.
These approaches do not cover all attempts taken to model the N cycle within a wetland,
yet they are the most often used. In the selected models description sections, differing
approaches will be more fully examined.
40
iii.
Phosphorous
Modeling of the P cycle would seem to be simpler than the N cycle because of the fewer
processes involved, yet this is not the case since processes in the P cycle are not entirely
understood.
For simplification, a high percentage of models simulate total P rather than its component
forms. A mass balance is implemented to account for inputs, outputs, and retention of P
amounts:
dPm
= Pin Pout Pret
dt
(27)
where Pm is the mass of phosphorous per unit wetland volume (M/L3); Pin is the influent
phosphorous (M/L3); Pout is the effluent phosphorous (M/L3); and Pret is the retention of
phosphorous (M/L3)
Various modeling approaches differ with respect to how they represent the mass balance
parameters. Mitsch et al. (1995) developed a simple model that accounted for retention by
incorporating the effect of the hydrologic loading on the wetland. The retention was calculated
as:
Pret = k (aLh + 1) PT
(28)
where Lh is the hydrologic loading of the wetland (L/T); k is the retention coefficient for P (1/L3;
a is the coefficient reflecting the magnitude of added hydrologic effect (T/L); and PT is the total P
in the system (M). Input is the phosphorous concentration of inflow and output is the
concentration value of wetland outflow. The retention coefficients were determined through
calibration.
Mitsch and Reeder (1991) used the same mass balance concept with more detail. The
model accounted for macrophyte and plankton phosphorous uptake and release, sedimentation
and resuspension velocities, and phosphorous concentrations in sediments. Christensen et al.
(1994) accounted for pools of dissolved TP in the water column, particulate TP in water, bottom
41
soil TP, and macrophyte uptake and release. Mineralization, sedimentation and inflows were all
modeled using first order equations.
Both models use simple adsorption rates that were determined from field data.
When adsorption is calculated, it is usually modeled using the Langmuir expression, which is
valid for monolayer adsorption (Jorgensen, 1988). The Langmuir model is of the form (Novotny
and Olem, 1994):
r=
Q o bC e
1 + bC e
(29)
where Qo is the adsorption maximum at a fixed temperature (g/g); b is a constant related to the
net energy of net enthalpy of adsorption (1/g); r is the adsorbed concentration of the
contaminate (g/g); and Ce is the dissolved (free) concentration of the contaminant water (g/1).
Jorgensen (1988) uses the Langmuir model to determine adsorption amounts while Dorge (1996)
uses a temperature-dependent Langmuir adsorption model.
There are other models to determine adsorption of P to particles. One approach is use of
the Freundlich equation (Novotny and Olem, 1994):
r = K * Ce
1/ n
(30)
where K and n are constants. The constants K and n are obtained by plotting r versus Ce on loglog graph paper, where the logarithmic intercept is K and the logarithmic slope equals 1/n. The
Freundlich isotherm is useful when the energy term in the Langmuir isotherm varies as a
function of surface coverage (Novotny and Olem, 1994).
For low concentrations of contaminants, the Freundlich and Langmuir isotherms can be
simplified to a linear isotherm:
r = Ce
(31)
42
Every model examined in the wetland review did not model the amounts of P
transformed through precipitation. This is due to the negligible amounts that are lost through
precipitation and also the difficulty involved in modeling the process.
iv.
Sediment
Wetland ecosystems are effective sediment traps, generally retaining more suspended
sediments than they export. Sediment deposition is important in improving a wetlands water
quality not only because sediments themselves are a water contaminant, but also because
sediments concentrate many toxic species and nutrients through sorption processes (Fennessey et
al., 1994). Sediment particles may possibly improve water quality as they settle and scavenge
suspended and dissolved contaminants from the water column (Hart, 1982), and can represent
permanent removal of contaminants through chemical breakdown, and burial in the bottom
(Johnson et al., 1984). Although intuitively wetland vegetation would augment sediment
deposition through filtration of the wetland waters, this is not necessarily the case. Hosokawa
and Horie (1992) observed no differences in TSS removal efficiency of vegetated and
unvegetated cells, suggesting that vegetation does not increase TSS removal.
Sediment deposition is one of the most difficult parameters to measure in wetland
ecosystems (Fennessey et al., 1994), because sedimentary processes are not well understood
(Mitsch and Gosselink, 1993). A constructed wetland in Florida retained over 80% of the
incoming total suspended solids (Knight, 1987), and a wetland bordering Lake Erie retained 55%
of the annual sediment inflow in a drought year (Mitsch and Reeder, 1992). The values reported
in the literature indicate sediment removal efficiencies between 80-95% (Daukas et al., 1989).
Modeling of sediment deposition entails determination of sedimentation velocities,
resuspension velocities, and the amount of available sediment material. Sedimentation and
resuspension velocities are usually modeled with zero order rate equations. The only differences
in amounts of sedimentation and resuspension is determined by the pools of available sediment
material which can increase and decrease due to catchment runoff and biological death. Mitsch
and Reeder (1991) included sediment coefficients for a plankton pool and for a macrophyte pool.
Christensen et al. (1994) used a mass balance approach to determine sediments in the water
43
column and sediments located on the wetland bottom. Sedimentation was modeled with a first
order equation.
One method used to increase the accuracy of the prediction of resuspension of material
from wetland bottoms is the determination of a critical velocity. The critical velocity is the
minimum velocity at which resuspension occurs. The theory of plain sedimentation predicts that
as the water velocity in a wetland increases, there is a point at which the shear stress will detach
particles from the wetland bottom. These stresses can be reflected with a Mannings coefficient
value (Kadlec and Hammer, 1996). Wetland flows simplify this determination because generally
flow is in the laminar regime. The critical velocity, u, can be determined with (Kadlec and
Hammer, 1996):
w1 / 3 H 1 / 6
u 7.2 *
2/3
nd
(32)
where d is the particle diameter (m); H is the water depth (m); n is the Mannings coefficient for
open channel (s/m1/3); u is the water velocity (m/d); and w is the particle settling velocity (m/s).
According to this theory, if the water velocity is less than the right-hand side of Equation 32,
resuspension will not occur. Since this result is based on laminar theory, it is necessary that two
Reynolds number criteria are satisfied:
Re1 =
d * * u 2 * H 1 / 6
n g
d * *w
1
Re 2 =
(33)
(34)
where is the water viscosity (kg/ms ~ .001 at 20 C), and is the water density (kg/m3 ~1000
at 20C). Although this theory places a constraint on when resuspension will and will not occur,
it does not predict how much will resuspend.
44
OR =
Q
w
A
(35)
where OR is the overflow rate (m/day), Q is the inflow to the basin (m3/day), A is the water
surface area (m2), and w is the particle settling velocity (m/day). Utilizing the ideal settling
basin assumption, all particles with settling velocities that are greater than the OR will be
trapped in the wetland. The theory assumes that all particles of a particular size will either be
completely retained (100%), or not at all (0%), by the wetland system. This may affect the
accuracy of the results.
Particle settling velocity may be estimated from Stokes equation (Novotny and Olem,
1994):
gd 2
w=
( s 1)
18
(36)
where d is the particle diameter (m), g is gravity acceleration (m/sec2), s is the specific gravity
of the particle with respect to water, and is the kinematic viscosity (m2/sec).
v.
Vegetation
According to Kadlec and Hammer (1988) a simple model of nutrient uptake is sufficient
for describing wetland processes, and highly sophisticated models are not warranted for
simulation of overall wetland features. This is beneficial as finely detailed models may require
parameter inputs for rarely collected data on leaf resistance to CO2, light intensity variations, leaf
mortality rate, and stem mortality rate (Dixon, 1974).
Various researchers employ the idea of a simple vegetation model. Dorge (1994) based
plant uptake on a ratio of daily to yearly irradiance multiplied by an empirical parameter for
45
yearly plant uptake. He assumed that over a yearly scale, plant uptake is equivalent to plant
death. Brown et al. (1988) used empirical data to fit a growth curve to the growing season for
both N and P. He assumed that there was a breakdown of 75% uptake from the soil
concentration and 25% from the water column. Christensen et al. (1994) used a simple mass
balance on macrophyte biomass which determines the change in biomass based on the
macrophyte death rate subtracted from the macrophyte growth rate. Macrophyte rates were
modeled using first order equations.
Output that describes hydroperiod and depth of inundation can be used to determine
proper vegetation for a specific wetland site. If the hydrologic characteristics for a wetland are
known, then based on known growth patterns for vegetation and climate, a user can determine
what would be the most appropriate vegetation to include in a wetland.
c.
approaches taken to model the hydrologic and nutrient dynamics in wetland systems. Far from
being comprehensive, a basic overview of the models, results of their application, ease of use,
and their deficiencies will be discussed, to present the scope of previous research.
Hammer and Kadlec (1986) developed a distributed, hydrologic model for the overland
flow of a wetland. The model takes into account the movement of surface water in response to
the hydrologic gradient and vegetation flow resistance. It employs the rate law equation
(Equation 11), combined with a mass balance to determine surface water flow in the system.
Input information includes site topography, porosities, ET, stream flows, groundwater
recharge/discharge, and pumped additions. Depths and flow rates are calculated as functions of
position and time. Evapotranspiration is estimated using the Thornthwaite method (Equations
12, 13, 14, and 15). The model accounts for snowmelt and freezing of the wetland water. The
model uses a Crank-Nicholson finite difference approximation technique to solve the partial
difference equations, along with a volume-differencing procedure to reduce cumulative errors.
The model was applied to data from the Porter Ranch site in Houghton, Michigan.
Simulation data from 1978 to 1979 were presented to detail the surface water elevation
predictions in the wetland. Visual analysis showed a very good match between predicted and
observed values. Input to the model is specific and may be difficult to obtain for it requires very
46
precise descriptions of the shape of the wetland and soil porosities and conductivities of the area.
The set of parameters chosen for the model is site-specific to the Porter Ranch system, and
generalization of parameters is necessary.
Feng and Molz (1997) developed a 2-dimensional, diffusion based wetland flow model
(WETFLOW) which allows the incorporation of spatial variations in flow resistance and
vegetation. Continuity equations are combined with a modified Mannings equation and solved
with a Picard-iteration scheme. The two basic parameter distributions required by WETFLOW
are land surface topography and flow resistance. Since the model is two-dimensional, the
amount of input needed to properly describe the site is large.
The model was applied to two situations: a laboratory experiment and a wetland pond.
The researchers related that the values were similar through visual analysis of experimental and
predicted data, and that WETFLOW performed well in a numerical sense. A decrease in grid
size significantly affects the representation of the land surface and hydrologic representation,
with smaller more detailed entries resulting in more accurate model predictions. For the
simulation concerning the wetland pond, 105,600 nodes was required to describe the site.
Guertin, Barten and Brooks (1987) developed a hydrologic model (PHIM) for wetlands
typical of those found in the Great Lakes region of the United States. PHIM is physically based
with input limited to climactic data (precipitation, maximum and minimum daily air temperature)
and wetland description information (e.g. vegetation type, canopy cover and soil depth). The
model accounts for three different types of wetlands; natural peatlands, mined peatlands, and
mineral soil upland peatlands. Inputs differ for each respective wetland system, but the basic
hydrologic inputs include net precipitation, snowmelt, upland inflow, groundwater inflow,
percolation, and ET.
All three wetland types were simulated with data from watersheds in northern Minnesota.
The short-term predictions for all three simulations were within one standard deviation of the
observed values. Long term evaluations were not completed.
Walton et al. (1996) developed the Wetlands Dynamic Water Budget Model, named
thusly because it provides magnitudes for the water budget components as well as water depths,
discharges and flow velocities throughout the entire wetland. There are three dynamically linked
modules which allow the model to represent the surface water, vertical processes and horizontal
groundwater movement in the wetland. The surface water module combines the equations for
47
conservation of momentum and volume (Equations 9 and 10). The vertical processes simulated
in the model are canopy interception and drainage, infiltration, surface water evaporation, soil
water evaporation, and transpiration. Horizontal movement of groundwater is based on Darcys
law (Equation 18). The model uses a link-node method that divides a wetland system into a
series of finite volumes where the water level is defined. The scheme is flexible because it can
easily represent simple or complex geometry, while the one dimensional flows used for the links
are simple to program and efficiently solve.
The model was applied to the Black Swamps wetland, located on the Cache River in
eastern Arkansas. A link node grid consisting of 66 nodes and 115 links were used to describe
the wetland area. The model was calibrated with data from the water year 1990 and validated
with data from the water years 1988 to 1991. The researchers presented the validation data
which visually matched the observed data points, but no statistical or sensitivity analyses were
presented. Although the model presents data that is accurate, the amount of input necessary to
generate the results are fairly extensive.
While all of the presented hydrologic models provide fairly, accurate predictions, their
use is limited as the number of input values is quite extensive. The amount of energy and
resources needed to collect the required input values make their use impractical. Although they
can be used to simulate the hydrologic functions of a wetland and the potential problems with
these designs, a wetland would not be designed specifically for water retention. If combined
with nutrient modeling, their use would be greatly enhanced.
Widener (1995) developed a mathematical model of the N cycle for FWS constructed
wetlands. The model is designed to represent the organic N, NH4+ and NO3- concentrations in a
wetland system to a very fine degree, supported by the entry of spatial examination in
centimeters. Ionic exchange of NH4+ with soil particles, nitrification, mineralization, and
diffusion of ammonium NH4+-N affect ammonium NH4+-N concentrations in the sediment.
Nitrification, denitrification, and diffusion processes affect NO3- concentration. Mineralization,
nitrification, and denitrification are modeled using first-order equations with temperature
dependent effects. Oxygen is an examined parameter, but not explicitly modeled. The model
assumes that there is active transport of oxygen to the wetland soil. The transport aerates the
wetland soil to a user-defined thickness, which the model assumes is the extent of root zone
penetration. The oxic zone is where nitrification occurs and where denitrification is inhibited.
48
The partial differential equations were solved using center-divided differences with
Taylor Series expansion. Inputs to the model are not extensive, and should be easily obtained
except for the determination of the wetland sediment aerobic zone. The model was not applied
to its intended target of NPS reduction due to lack of available data, but six scenarios were
examined to determine if model predictions are comparable to qualitative theory. The scenarios
tested the spatial distribution of the calculations, the temperature effects and the flow condition
effects. Widener (1995) concluded that the model predictions were sound from analyzing graphs
of the ammonium and nitrate movement.
The extensive time required for simulations during the study period was limiting, as the
computation to real time simulation ratio was 30:1. This eliminated any long term simulations as
it would have required 12.2 computing days to simulate one year of real world processes. This
problem may have been alleviated due to recent increases in computing capabilities.
Christensen, Mitsch, and Jorgensen (1994) developed an ecosystem model to examine the
retention of sediment and P at the Des Plaines, Illinois experimental wetlands. The hydrologic
model is a black box model and uses a simple water budget to account for runoff and pumped
inflow, runoff and pumped outflow, direct precipitation, and ET amounts. Two pools for
sediment are simulated, amounts located in the water and the bottom. There are three pools for
P; dissolved and particulate P in the water column, and bottom total P. The mass balance for
sediment in water is affected by influent amounts, net sedimentation of solids, sediment outflow,
and sediment production in the water which consists of macrophyte and algal production.
Sediments on the bottom are accounted for with additions from macrophyte death and decay of
organic bottom sediments. The dissolved P balance accounts for mineralization from the bottom
and particulate P pools, inflows, outflows, and sedimentation. Mineralization, and sedimentation
are modeled with first order rate equations. Particulate P balances inflows, outflows, and
sedimentation, while bottom P accounts for sedimentation, remineralization and contributions
from macrophyte death. Macrophyte biomass is modeled with a simple zero order growth and
death rate. Primary productivity is modeled with a dependence upon the solar radiation and
water temperature.
The model was calibrated with data collected at the Des Plaines wetlands, but not
validated with an independent data set. Calibration was determined through visual analysis of
the measured and predicted values, minimizing the differences between the two. The model
49
appeared to be the most sensitive to the sedimentation velocity in the wetland system. It does not
account for precipitation of P in the wetland system.
Mitsch and Reeder (1991) developed a model for a coastal wetland of Lake Erie called
Old Woman Creek, which included submodels for hydrology, productivity and P. The hydrology
submodel is capable of running independently from the other submodels. Factors affecting the
only hydrologic state variable, volume of water in the marsh, include rainfall, inflow, ET and
outflow. The productivity model simulates macrophyte cover as a function of water level, which
was determined through empirical observations. Plankton biomass is based on the gross primary
productivity, solar radiation and a zero order sedimentation rate. Incoming P concentrations were
based on empirical observations from similar areas and are a function of flow. The model
utilized two P storages, wetland waters and sediments. Sedimentation and resuspension
pathways are based on an average settling velocity and a function based on wetland depth. The
model differs from similar models in that it accounts for plankton biomass in the overall P
balance and determines the amount of P using a simple ratio between the P and growth rate base
on solar radiation.
The model assumes that there is no P limitation in the wetland, therefore no attempt was
made to divide the P in the water column into available and unavailable forms. Assumptions
were also made concerning the macrophyte growth, which was limited to the dominant
vegetation of Nelumbo lutea. The model uses a fourth order Runge-Kutta technique with and
integration interval of 0.1 day.
The model was calibrated with data collected from March, 1988 to November, 1988 in a
step-wise manner where first hydrology; then primary productivity, macrophyte biomass, and
chlorophyl data; and finally P were examined. Subsequent simulations were made for various
combinations of flow from the watershed to the wetland and varying water levels for Old Woman
Creek to determine optimal conditions of P retention and to test the stability and range of the
model. P retention varied from 17% to 52%, with highest retention of P by the wetland
occurring when high inflows from the watershed are coupled with high water levels in the
wetlands.
According to the researchers, results for the hydrologic and nutrient modeling were fairly
accurate. The hydrology submodel predictions were usually within 0.2 m of recorded field data,
but did not reflect dramatic changes in water level. Field measurements of gross primary
50
productivity were within one standard deviation of the field measurements 70% of the time.
Predicted average P concentrations followed the observed field data trends.
Brown (1988) developed a model in BASIC, for the hydrology and nutrient dynamics of
a wetland typical of those found in north and central Florida. The model predicts water depth,
recharge, and surface outflow as well as concentrations of P (total P) and three N species (NOX,
NH3, and organic N). The hydrology submodel may be run independently from the nutrient
dynamics and accounts for runoff, precipitation, evaporation, transpiration, and surface and
groundwater outflow. Evaporation is estimated as the average pan evaporation with adjustments
made for rain, temperature and humidity. Transpiration is described as the product of a growth
coefficient and solar radiation. The N cycle accounts for atmospheric, plant decomposition,
sedimentation, groundwater, and runoff inputs and for transformations such as nitrification,
denitrification , and mineralization. Transformations of N were modeled with zero order rate
equations based upon measured rate coefficients. Hydroperiod and depth of inundation are
calculated, which allow appropriate vegetation choices to be planted to optimize plant growth
and nutrient uptake. Nutrient uptake by vegetation is based upon the product of tissue nutrient
concentration and growth rate. The output consists of water levels and nutrient concentrations at
a daily time step.
Required input to the model are readily available or fairly easy to estimate. This is
because the model is lumped and requires mostly weekly, monthly and yearly, but not daily
inputs. Comparisons of field data to model predictions were not presented, instead simulation
results for hypothetical additions of treated effluent were presented. The simulation results
concluded that the addition of treated effluent increased the depth and duration of flooding in the
wetland. The results also showed a predicted decrease in the effluent concentration of the N and
P species. The decrease in concentration may stem from the extra water volume which enters
and subsequently leaves the wetland, but could also mean that the additional treated effluent may
have had an initial lower concentration of contaminants.
Drawbacks to the model include the failure to model diffusion processes, and the
assumption that the amount of organic C is non-limiting in the denitrification process. This
limits the application of the model to areas where these assumptions apply. In addition, Browns
model does not allow for seasonal parameter changes, which may vary considerably.
51
Jorgensen et al. (1988) modeled the nutrient removal capabilities of a wet meadow and
natural reedswamp in Denmark. The hydrology, N, and P submodels were created using the
language CSMP. The hydrology submodel accounts for precipitation, ET, inflows, outflows,
water uptake by plants, and the modeling of vertical and horizontal flows with Darcys equation.
The N submodel included N input from inflows and rainwater, ammonification, nitrification,
denitrification, adsorption, and plant assimilation processes. Ammonification and nitrification
are described with first order functions dependent on moisture content and temperature.
Volatilization is described as a first order equation that is dependent on moisture content,
temperature and pH. Denitrification is modeled using a Michaelis-Menton equation (Equation
24). Adsorption is modeled using a Langmuir expression (Equation 30), which is valid for
monolayer adsorption. Plant uptake is dependent on the root zone nutrient gradients and the
maximum concentration of nutrients within the vegetation. Plant decomposition is described
with a temperature dependent first order equation that allows for complete plant degradation
between autumn and the following springtime. The P submodel is described similar to its
parallel N processes with different parameters, being the only distinction.
The presented model was not calibrated or validated due to a lack of available data;
therefore no statistical or sensitivity analyses were presented. Since the model covers so many
processes, model input is fairly difficult to obtain. Although the model may be very
encompassing, the accuracy of the model is unknown.
Kadlec and Hammer (1988) developed a dynamic, compartmental, simulation model to
evaluate the effects of wastewater additions to a natural wetland. Flow rates and water levels are
calculated using a one-dimensional model for surface flow through vegetation (Hammer and
Kadlec, 1986). The ecosystem model took a different approach than other models by containing
nine compartments that accounted for annual and woody biomass; roots; standing dead; annual
and woody litter; and top, mid and deep soil layers. Solids exchange rates were computed as a
constant depletion rate multiplied by the peak pool amount. Summer decomposition and nutrient
transformation rates were 10 times the winter rates. The annual biomass pool is assumed to be
zero on the first day of the growing season, producing at a constant composition until it reaches
peak biomass on the last day of the growing season. Growth rates are a function of soil nutrient
limits and are constrained by the limiting nutrient. Mass balances were applied to chloride, N
and P amounts in the wetland.
52
The model was applied to the Porter Ranch wastewater treatment facility at Houghton
Lake, Michigan. Results were reported for simulations comparing the first two years of the
systems operation (1978 and 1979). The authors determined that the model predicted surface
water solute concentrations, as well as biomass, litter, and soil dynamics accurately based on a
visual analysis of recorded effluent and model predictions. No statistical or sensitivity analyses
were provided.
The model requires a number of plant growth and death parameters for input that may be
difficult to attain. In addition, the model does not allow for seasonal changes in parameter rate
values.
Dorge (1994, 1996) developed a simulation model describing the hydrology, N turnover,
and P turnover in freshwater wetlands. The WET model describes four components in the water
phase (NH4-N, NO3-N, PO4-P and Particulate P) and five attached to the sediment/peat (plant
biomass, labile organic matter, stabile organic matter, immobile N and adsorbed N). The N cycle
is dynamically modeled while the P cycle is modeled as a black box, i.e. as an inexhaustible pool
of P. The hydrologic submodel accounts for inflow, precipitation, evaporation, outflow, and
infiltration. Nitrification and denitrification are described with temperature dependent
Michaelis-Menton equations. Adsorption is determined with the Langmuir equation (Equation
30), while mineralization is a temperature dependent process coupled with the decomposition of
organic material. The plant uptake component assumes that there is no N limitation and that
growth is only dependent on the actual radiance intensity at the wetland surface. The model
accounts for microbial immobilization with a temperature dependent, first order assimilation rate
and assumes the subsequent release is based on a first order decay rate. Diffusion is modeled
based upon the hydrologic gradients in the system, where transport of dissolved constituents
travels between the surface water and peat water depending on the flow of the water. Diffusion
is not based on concentration gradients in the wetland. The P cycle accounts for mineralization,
immobilization and plant uptake of PO4-P. The particulate-P component accounts for
sedimentation and resuspension. Sedimentation has an empirically determined maximum and
minimum rate and is dependent on plant biomass in the wetland. Resuspension is an empirically
determined constant that is input as grams per meter squared.
An early version of the model that modeled only hydrologic and N processes was applied
to three wetlands in Denmark. The researchers reported results as being fairly accurate based on
53
visual analysis, with no statistical analysis. The model had difficulty in modeling extreme
weather events such as the colder winter months and seasonal flooding. The model is interfaced
with a Windows based system, allowing for easy entry of data. Input entries are minimal and
should be fairly easy to determine.
Gidley (1995) developed a mechanistic, compartmental simulation model for SSF
constructed wetlands, written in STELLA. The model consists of six submodels, one each for
water volume, N, C, dissolved oxygen, heterotrophic and autotrophic bacteria. The N cycle
includes pools for dissolved, immobilized and particulate organic N, ammonium and nitrate.
Processes modeled include both particulate and dissolved organic N ammonification, N uptake,
incorporation by biomass and microbes, nitrification and denitrification. The C cycle models the
balances of particulate and dissolved C, by accounting for contributions made from plant death,
leaching, decomposition, wastewater BOD, and microbial death; and removals due to
heterotrophic growth and peat accumulation. Bacteria are modeled using dual substrate Monod
kinetics with limitations due to electron donors and acceptors, along with temperature
considerations. The DO balance is computed assuming plant translocation and losses due to
microbial respiration. The water balance uses Darcys law to determine outflow, and
Thornthwaites method to estimate ET.
Gidleys model took a different approach to the typical modeling of wetland systems. It
accounted for the C and N interaction, and the effect of DO levels upon microbial growth. It also
directly linked microbial growth and death to the consumption of nutrients in the system.
Previous models accounted for these interactions with zero to first order rate equations that
assume rates are dependent only on initial concentrations.
The model was tested on a constructed wetland site in Maryland by examining model
predictions of DO, BOD5, and N concentrations against actual effluent values. Model results
reproduced seasonal trends and treatment efficiency well based on visual analysis of the output
data. A statistical comparison of measured and predicted values were not completed; however a
sensitivity analysis determined that the model was more sensitive to parameters affecting
microbial growth and substrate use directly.
54
D.
attempt to alleviate the impact of NPS pollution. Constructed wetlands are an attractive BMP
because in addition to controlling NPS pollutants, they provide other beneficial functions to the
environment such as wildlife habitat and recreation, while also being relatively low maintenance
and cost effective.
Many processes affect the N and P cycles in a wetland. To complicate matters, many of
these processes are poorly understood and are affected by a complex web of dependent
interactions. The hydrologic cycle controls all of the wetlands functions, but factors which affect
the N cycle include the vegetative, C, bacteria and dissolved oxygen cycles. The N cycle is
affected by transformations such mineralization, nitrification, denitrification, N fixation, and
assimilation; and transfer processes such as particulate settling and resuspension, diffusion,
litterfall, plant uptake and translocation, ammonia volatilization, adsorption, seed release, and
organism migrations. The P cycle is a sedimentary cycle and is affected by adsorption,
precipitation, desorption, mineralization, sedimentation, and resuspension.
One of the more effective ways to enhance the design and construction of wetlands is
with the use of models. Models provide an ability to make comparisons among alternative
designs and management strategies, allowing a wetland to be optimally utilized for its intended
purpose, be that for pollution control or wildlife interactions.
The problems with most existing models is that model development was geared towards
certain objectives and goals to fit the specific needs and functions of a certain wetland type. This
is exemplified by assumptions made during model development by Brown (1988), and Mitsch
and Reeder (1991). Assumptions initially made to ease model development restrict the
transferability of the model from one area to another, as assumptions for one location may not
hold in another. To develop a model that is applicable to many situations, generalizations will
need to be made. Generalizations usually decrease accuracy; nonetheless, these model types are
necessary so that an easy to use, widely applicable program can be developed which will not
require constant modifications. From experience, it is noted that a person can operate a model
better if they have a chance to utilize the model on a frequent basis. Therefore, a model that can
55
provide the ability to examine many different wetland designs at one time, would be the most
beneficial for a wetland designer.
Individually each model presented in the model review works for the intended purpose of
the described research. However, the question is whether others would use the model to actually
design and model wetlands. If a proposed wetland fits the assumptions made by previous
models, they might be able to, otherwise the alternative is to find a model that is similar to the
desired system and alter the modeling code.
The ability to model many designs is a key point missed by many existing models. Most
simulate either FWS or SSF wetlands, but usually never both. There is an inability to change
values of parameters after initial input, which does not allow the modeling of changes due to
seasonal variability. Most models require all components of a model (hydrologic, N and P) to be
run, even if only one cycle is desired. This can also apply to processes in individual cycles such
as N fixation or atmospheric deposition in the N cycle. Additionally, there is a lack of
documentation on how to operate most existing models.
Another overall problem with previous models is that the input requirements are
restricting. Most models are not linked to existing models that will accurately quantify input
values to the wetland. Known daily input values would be ideal as rapid changes in weather can
effect incoming loads; however, when predicting input values for a possible wetland these values
are unknown. The ability to predict incoming concentrations with one model or the ability to
input values directly from an existing model would be beneficial, as better input values usually
results in more accurate effluent predictions.
Additionally, one would like the ability to determine input values on a more theoretical
basis, because less laboratory and research efforts would be needed to determine how wetland
designs would function. For example, most wetland models have lumped the effects of the N
processes with first order (Widener, 1995) rate equations, which do not account for the
interactions of the bacterial, C, and DO cycles on these processes. Attempts were made by
Dorge (1994, 1996) and Jorgensen (1988) to use Michaelis-Menten equations to account for
bacterial affects, and Gidley (1995) directly incorporates actual bacterial growth, in addition to
the C and DO cycles. These approaches are a movement in a new direction towards modeling
wetland processes and relations.
56
From the review, it can be seen that there is a need to develop a model that is general
enough to be applied to various situations, but accurate enough to be useful and implemented. In
this study, attempts were made to develop such a model.
57
III:
Model Development
A.
MODEL OVERVIEW:
The developed simulation model, SET-WET, models both FWS and SSF wetlands. The
SET-WET model simulates the hydrologic, vegetative, N, C, DO, bacteria, sediment, and P
cycles of the wetland. SET-WET is designed so that the cycles may be modeled independently
with some constraints. The N, C, bacteria, and DO cycles are linked and may not be run
independently. These cycles will be referred to as the NCOB cycle to denote this dependence.
The P cycle is dependent on the sediment cycle, and all of the cycles except the hydrologic,
require that vegetative growth also be simulated. The sediment and P cycles may not be modeled
in conjunction with the SSF wetland option. In addition to the choice of which cycles may be
simulated, there are options within each modeling cycle that allow certain processes to be
simulated. More details will be given with each individual cycle description.
The SET-WET model simulates the wetland as a continuously stirred tank reactor. The
model user must enter the average wetland length, average wetland width, wetland substrate
bottom (HB, m), wetland substrate top (HT, m); the top of the wetland cell (HO, m) and the
initial water height (HI, m) to describe the physical description and initial hydrologic conditions
in the wetland system. The wetland levels (HB, HI, HO, HT) are all relative to a user-defined
datum; however, it is suggested that the user designate the substrate bottom as the reference level
(datum). Figure 6 represents how the FWS and SSF wetlands are dimensionally represented in
the SET-WET model.
The SET-WET model is designed as a flexible tool and could be applied to many
situations. Every wetland type could not be covered, yet there are many options in the model
that should allow the design and modeling of various new and old wetland systems. The SETWET model is also designed to be user-friendly. One feature of SET-WET which increases the
ease of model use is the lack of formatted model input. This style was chosen to simplify
inputting parameter values because it alleviates problems associated with incorrect value entries
due to differences in spacing and decimal places. It is suggested that inputs be entered similar to
the format the parameters are placed in the Fortran code (usually six entries per line) to facilitate
the comparisons of input parameter readings. Appendix A lists all possible entries to the model
58
along with the required units. Appendix B details the order and entry of inputs to the model for
each respective submodel. Appendix C lists the SET-WET FORTRAN code.
The SET-WET model is designed to accept two forms of data inputs. The first option
requires hydrologic, sediment and nutrient inputs which consist of daily input values for various
hydrologic and nutrient parameters (daily inflow (m3/d); NO3-, NH4+, DO etc., (mg/L)). These
values may have been collected or been generated by a NPS model such as ANSWERS or
BASINS. Output from the existing NPS models would be used as input to the SET-WET model.
The second option for data input is based on the SCS curve method (Equations 6 and 7). The
hydrologic inflow that is determined by the SCS method is combined with a runoff concentration
to determine the constituents inflow to the wetland system in the following manner:
CONIN = WATINPUT * RUNCON
(37)
where CONIN is the constituent input to wetland (g); WATINPUT is the water input due to
catchment runoff (m3); and RUNCON is the concentration in the runoff for the individual
constituent (mg/L). Runoff concentration coefficients may be determined from empirical
analysis of site data or estimated from similar watershed situations.
59
Two different models (Thornthwaite method or pan method) may be used to model ET.
Six different options to describe the wetland outlet may be used to simulate the hydrologic
outflow from the wetland system, including five types for FWS systems, and one for SSF
wetlands. For the N cycle, the atmospheric deposition inputs, N fixation and volatilization may
be simulated. These components are not mandatory inputs as various wetland sites may or may
not be significantly affected by these processes. For the P cycle, adsorption may be modeled in
two different manners, using the Freundlich isotherm, which may be simplified to a linear
isotherm through choice of parameters (Equations 30 and 31).
The model is designed to control time management with a season and time period format.
Each season period consists of a user-defined number of time periods, with each time period
representing one day. To conserve memory, the season periods are programmed for a maximum
length of 500 days; however, this length can easily be changed in the program code. The
season/time period format minimizes data entry by allowing stable parameter values to be
entered on fewer occasions. Instead of entering data for individual days, values that stay the
same for longer time frames need only be entered once every season period. However, a change
in season periods allows the user to input new values for these parameters. For example, four
season periods could be used to simulate the four seasons of the year; plant growth and microbial
parameters may differ for these time periods and can be changed accordingly. In between season
periods, certain flagged entries will determine whether parameter values need to be changed; a
1 entry signifying changes will be made, while a 0 entry signifies no changes are necessary.
1.
two pools of water simulated for FWS systems and one for SSF systems. To ease the transition
between coding for both systems, the same base parameter names have been used for both
wetland types soil/peat systems. For processes or pools that occur in both the surface water and
peat of the FWS system, the difference is signified in the programming code with a W to
represent surface water, and a B for wetland soil/peat bottoms, at the end of individual
parameter code name. Figure 6 shows how these respective pools are represented. For example,
the parameter code HETEROW is used to account for the number of heterotrophic bacteria in the
60
FWS wetland surface water, while HETEROB accounts for those in the wetland soil/peat of both
wetland systems.
The principal differences that exist between the modeling of FWS and SSF wetlands are
the inclusion of settling and resuspension rates for particulates, and the use of mass transfer rates
for dissolved constituents, to model movement in a FWS wetland. The fall rates of particles such
as particulate organic N, particulate organic C and sediment particles is based on the comparison
of the fall rate with the height of the water differential in the FWS surface (wetland water surface
minus wetland soil/peat top). The particles are considered to be homogeneously mixed through
the water surface volume and the settling amount is considered 100% if the fall rate of the
particles is greater than the difference between the wetland water surface and the top of the
wetland soil/peat layer. If the fall rate is less than the difference between the water surface and
the wetland soil then a proportion of the constituent, based on the outflow and water volume, is
assumed to settle.
Resuspension of sediment is based on a critical velocity, and is modeled using Equation
32. When the critical velocity is attained, a certain thickness of the soil/peat layer is affected.
From this thickness a user-defined percentage of the analyzed constituent will re-enter the water
column and be resuspended from the surface bottom. For these processes, the resuspension is
modeled as:
RESTHC
HT j 1 HB
(38)
where CONRE is the constituent resuspension amount (g constituents); CONB is the total
constituent amount in the wetland peat bottom (g constituents); CONRES is the percentage of
constituent that resuspends from the critical resuspension area (g resuspended cons./ g cons.
total); RESTHC equals the thickness of wetland soil to which the velocity resuspends
particles(m); HT is the top of the wetland soil/peat surface (m); and HB is the bottom of the
wetland soil/peat surface (m). The CONRES and RESTHC parameter values are determined
through calibration.
Mass transfer rates are used to model the transfer of constituents from the surface water
to the peat water. Mass transfer is modeled as:
61
MASSTC = 864 * K mt * (C s C w ) * A
(39)
where MASSTC is the mass transfer amount for a constituent (g/day); Kmt is the mass transfer
coefficient (cm/sec); Cs is the concentration of the constituent in surface water (mg/L); Cw is the
concentration of the constituent in the peat water (mg/L); and A is the wetland area (m2).
B.
MODEL COMPONENTS:
The following sections will describe the various submodels; their purpose and the
reasoning behind the equations and assumptions used in their development. For clarity and
conciseness, while explaining individual processes in the submodel descriptions, the W and B
indicators at the end of each parameter code name will be omitted from the parameter names.
However, to differentiate which variables have two pools for the FWS system, a parenthesized
W/B will be placed after the initial parameter description. If the process occurs in only one of
the FWS pools they will be distinguished with a parenthesized W or B. The described processes
will occur in the SSF system, unless otherwise stated.
Figure 7 shows the relationships between the main code and the various submodels for
SET-WET. As Figure 7 shows, the main program calls every submodel a various number of
times. The number of times and whether a sub-model is called are dependent on the length of
the simulation and which cycles are modeled.
Figures 8 through 22 represent the relationships between modeled processes affecting the
respective cycles in SET-WET. The figures represent the various stocks, parameters and flows
that are accounted for by SET-WET during a simulation run. Appendix D explains the symbols
used in the respective figures.
1.
when and if each submodel should be called, transfers data between time and season periods,
opens input and output files, and writes to the files. There are no inputs to the main program, but
it is the code that links all of the submodel components together.
62
BASE
( ONE TIME )!
DELTAH
( T )*,#
HYDROLOGY
( S + T )!
PHOSPHOROUS
( S + T )#
VEGETATION
( S + T )*,#
MAIN CODE
SEDIMENT
( S + T )#
CARBON
( S + T )*
BACTERIA
S + T )*
DISSOLVED
OXYGEN (S + T)*
NITROGEN
( S + T )*
!
*
#
Number of times the MAIN CODE calls each respective submodel for an entire simulation run.
Submodel is called whenever SET-WET runs a simulation
If the NCOB cycle is simulated, then submodel is called by the main code
If the phosphorous cycle is simulated, then submodel is called by the main code
One Time - submodel is called only once for the entire simulation run
T submodel is called once every time period
S + T submodel is called by main code once every season and time period
2.
Base Submodel:
The base submodel describes the basic design of the wetland system and determines
which nutrient cycles will be modeled. As stated before, the input values are not in a fixed
format; however, it is suggested that the values are entered in the same manner of the Fortran
code, or similar to the manner that it is displayed in Appendix B.
63
3.
Hydrology Submodel:
The hydrology submodel is active for every simulation. The model represents two pools
of water for the FWS simulations and one for the SSF wetland simulations (Figures 8 and 9).
The model assumes that the FWS wetland always has water lying upon the surface and is never
completely exhausted of free lying surface water. It further assumes that the water flow for a
SSF wetland is always beneath surface level. For FWS simulations, the soil and peat area is
assumed to be saturated at all times. A valid assumption for many constructed wetlands would
be that they are lined, thus infiltration/percolation is minimal; however, there is a percolation and
infiltration option that allows a daily set amount of water to be released or enter the system.
Water input could be derived from four sources; 1) watershed catchment runoff, 2) direct
precipitation input, 3) percolation additions, and 4) point source additions. Although initially
designed for constructed wetlands, the model may analyze any natural wetlands that follow these
conditions.
The total amount of water in the system is determined using a simple mass balance
similar to Equation 2. The water budget for the model may be written as:
dV
= Qc + Q p Qout + BMV + SDV + ( P PI ET ) * A
dt
(40)
where dV/dt is the change in surface storage (m3/day); Qc is the runoff from the catchment
(m3/day); Qp are the water flow from possible point source additions (m3/day); Qout is the outflow
rate (m3/day); BMV is the change in living biomass volume in the surface water (m3/day);
SDV is the change in standing dead volume in surface water (m3/day); P is the daily
precipitation rate (m/day); PI is the percolation/infiltration rate (m/day); ET is the
evapotranspiration rate (m/day); and A is the wetland surface area (m2), determined from the
input wetland length and width.
As stated earlier, there are two options available for determining runoff from the
watershed; the daily option and the SCS method. The daily option requires input of Qc on a daily
basis, where the amount is either known or determined with an existing NPS model. If the SCS
64
method is used, runoff will be determined from the daily precipitation and the SCS curve
number.
The outflow from the system is a function of the water volume, and the type of outflow
device associated with the wetland, while ET is a function of daily temperature and local climate.
There are six different options that can be chosen to represent the wetland outlet, five for FWS
wetlands and one for SSF wetlands.
Outlet option 1 is for rectangular weirs. A modified Francis equation (Benefield et. al,
1984) is used to describe the head-discharge relationship for weirs of this type and is determined
by:
Q = 1.843 ( L 0.2 H ) H
3
2
(41)
where Q is the discharge (m3/s); L is the weir length (m); and H is the head on the weir (m). The
head on the weir is the difference between the wetland water height (HI) and the bottom level of
the outflow weir (HOUT, m). The Francis equation allows both suppressed and contracted weir
outflows to be determined.
If the water height in the system is lower than HOUT, then it is assumed that outflow
from the system is zero. If the water height in the system is higher than the HOVER (water
height completely above the weir) mark, outflow is the addition of the overflow through the weir
and the water that spills over the weir height.
Outlet option 2 is for fully contracted v-notch weirs of any angle between 25 and 100
degrees (Kulin and Compton, 1975). A modified Kindsvater-Shen relationship is used to
determine outflow:
5/2
Q = 2.363 * C e *tan( ) * h1e
2
(42)
where Q is the discharge over the weir (m3/s); Ce is the effective discharge coefficient; is the
angle of the v-notch (degrees); h1 is the head on the weir (m); kh is the head correction factor (m,
function of ); and h1e is equivalent to the sum of h1 and kh.
65
Hydtype=1,
Watershed Area,Length,
Width,SC Curve Number
Biomass
Volume
Hydtype=1,
Daily Precipitation
Hydtype=0,
Water input
Standing Dead
Volume
% Standing dead
Underw ater
% Biomass
Underw ater
WATER VOLUME
SURFACE
Evapotranspiration
Point Flow
Outflow
Outlet=5;
Alpha,
Beta,
Kcoeff,
Hout
Evap=1;
Panevap
HT
Evap=0;
Evap=0;
Air Temp., Evap. Coef.,
Daylength, Heat Index
Daily Precipitation
Length
Percolation/Infiltration Flag
Surface Area
Outlet=1;
Hout,
Outw idth,
Hover,
Ncont
Outlet=2;
Angvnot,
Diseffc,
Khcoef,
Hout
HI
Outlet=3;
Hout,
Areapipe,
Contc
Outlet=4;
Flow out,
Toppump
Width
HB
Percolation/Infiltration Rate
Peat Porosity
Percolation/Infiltration
WATER
VOLUME
BOTTOM
FIGURE 8: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE SUBMODEL FOR FWS WETLANDS IN THE SETWET MODEL
66
Hydtype=1,
Daily Precipitation
Hydtype=1,
Watershed Area,Length,
Width,SC Curve Number
Hydtype=0,
Water input
Peat Porosity
Outflow
WATER VOLUME
BOTTOM
HT
Evap=1;
Panevap
Precipitation
Percolation/
Infiltration
Daily Precipitation
Evapotranspiration
Point Flow
HB
Evap=0;
Air Temp.,
Daylength,
HI
Evap=0;
Evap. Coef.
Heat Index
Hout
Percolation/Infiltration Rate
Porosity
Hydraulic
Conductivity
Bed slope
Percolation/Infiltration Flag
Width
Length
Surface Area
FIGURE 9: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE SUBMODEL FOR SSF WETLANDS IN THE SETWET MODEL
67
Outlet option 3 is for a fixed non-perforated pipe that removes water from the wetland
according to the hydraulic pressure on the outlet opening.
OUTFLOW = CONTC * AREAPIPE * 2 * 9.81* ( HI HOUT )
(43)
where CONTC is the contraction coefficient for the pipe; AREAPIPE is the area of opening for
the outflow pipe (m2); HI is the water level in the wetland relative to a zero datum (m); and
HOUT is the height (m) of the outflow opening relative to a zero datum.
Outlet option 4 represents pumped discharge that removes a constant amount of water
from the system per day. The input values are TOPPUMP and FLOWOUT and represent
respectively the height necessary in the wetland for water to flow from the system (m, with
respect to zero datum) and the amount removed per day when outflow is continuos through the
day (m3). Flow is continuos when the wetland water level is above the TOPPUMP location.
For each of these outlet options, the water velocity in the wetland system is determined
after the outflow in the system is determined. Water velocity is calculated by dividing the
outflow through the system by the wetland cross-sectional area (Equation 8).
For option 5, water velocity is determined first and is then used to calculate outflow.
Water velocity is determined using the rate law equation (Equation 11). In the rate law equation
when flow is in the turbulent range, (slope component) is approximately 0.5, and if flow is in
the laminar range, =1.0. The (depth component) value is usually in the range of 24, but is
dependent on microtopography as well as the stem-density depth distribution of the wetland
(Hammer and Kadlec, 1986). The calculated velocity in the wetland system is multiplied by the
wetland cross-sectional area to determine outflow.
Outlet option 6 is for SSF wetlands. To determine the outflow from the SSF wetland the
water flow through the wetland substrate must be determined. Darcys equation (Equation 18) is
used to model this type of flow (Kadlec, 1989; Jorgensen et. al, 1988). Although flow through
the SSF wetland may be unsteady due to factors such as precipitation and ET, it is assumed that
it is steady for short time periods, which is a requirement for Darcys law. The system porosity
is considered constant throughout each season period. Changes in porosity may be accounted for
by the user during changes in season periods; however, the effects of peat accumulation on
hydraulic conductivity and porosity are not determined by the SET-WET model. The hydraulic
68
gradient in the system is assumed to be the greater value between the difference in elevation
between the wetland water surface and the outflow pipe height, or the bed slope.
The model runs on a daily time step, but to improve the accuracy of the model, the
outflow data is calculated on an hourly basis. The amounts of individual water inputs and
outputs (except outflow) are first divided by twenty-four to give an hourly, instead of daily rate.
These values are used to determine outflow for a one-hour time period and a new wetland water
surface height (HI) is determined. The new water surface is then used in the following outflow
calculation and continued until 24 one-hour time steps are conducted. Outflows for each hourly
time step are then added to determine the total daily outflow from the system. No other
component of the model is run on a time step shorter than a day.
Evaporation and transpiration losses are difficult to determine separately, and thus are
combined as an ET parameter. Various models are available for estimating ET. Thornthwaites
method (Equations 12, 13, 14 and 15) and the pan method were used in SET-WET because their
required input data are readily available. For the Thornthwaite method, daily ET approximations
are determined by using the daily average air temperature instead of the monthly average
temperature and then dividing by 30 days:
a
DL 10 *AT W * L
ET =
*
*
12 HINDEX 30
(44)
where DL is the daylength (hours daylight/12 ); AT is the daily air temperature (C); HINDEX is
the heat index based on the historical average monthly temperatures (unitless); a is a determined
coefficient based on the heat index (unitless); W is the average wetland width (m); and L is the
average wetland length (m). ET is considered zero when temperatures are below freezing.
4.
Vegetation Submodel:
Solar radiation, temperature, water, and available nutrients are a few of the factors
affecting plant growth. Solar radiation data is not readily available, therefore, a simple model of
nutrient uptake was used in SET-WET to model vegetative growth (Figure 10). A simple
wetland model is sufficient for overall wetland features (Hammer and Kadlec, 1988) which is
69
important because most proposed plant models are very complex. The model is similar in
structure and form to the one previously described by Gidley (1995).
There are two components to the vegetation submodel, BIOMASST and STANDDT.
The total amount of living plant biomass in the wetland system is defined as BIOMASST, while
STANNDT refers to the dead aboveground plant parts before they have had a chance to fall to
the ground and become litter.
The total amount of plant biomass in the system is determined by:
BIOMASSTJ = BIOMASSTJ-1 + (BIOMGROWJ
- BIOMDTHJ - BIOMOUTJ)*dt
(45)
where BIOMASST is the amount of live plant biomass in wetland (g biomass); BIOMGROW is
the amount of plant growth in the system (g biomass/day); BIOMDTH is the amount of biomass
death in the system (g biomass/day); BIOMOUT is the amount of biomass removal from system
(g biomass/day); J equals the time period number (day from start of season period), and dt is the
daily time step (day).
The amount of standing dead in the system is accounted for by:
STANDDTJ = STANDDTJ-1 + (BIOMDTHJ
+ PHYDEGJ - SDEADOUTJ)*dt
(46)
where STANDDT is the amount of standing dead biomass in wetland (g biomass); PHYDSEG is
the amount of physical degradation of standing dead (g biomass/day); and SDEADOUT is the
amount of standing dead removal from the wetland (g biomass/day).
Biomass growth is simulated with a simple zero order equation (Equation 21) that
multiplies the wetland area by a biomass growth rate. At the beginning of the growing season,
biomass of an assumed constant composition throughout the wetland system, grows at a linearly
increasing rate for a user-specified number of days, until the maximum growth rate is reached.
This growth continues throughout the growing season until winter, at which time the growth rate
linearly decreases to zero at a user-defined rate. The growth rates are ramped to represent the
gradual changes within the system and to reduce model instabilities. The start and end of the
70
Time period
LENGTH
WIDTH
Biomass grow th
Biomass grow th
Biomass
Death
TOTAL
BIOMASS
TOTAL
STANDING
DEAD
Physical
Degradation
WINKCO
Winter flag
Amount of STANDD
Living
biomass
removal
FIGURE 10: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE VEGETATION CYCLE SUBMODEL OF THE
SET-WET MODEL
71
growing season will differ by location; but a general rule would be that the start and end of the
growing seasons corresponds with the first and last frosts in the area. The time periods over
which plant growth increases and decreases are estimated, as literature values are unavailable. It
is assumed that biomass growth is not limited by lack of nutrients or water supply. Since little is
known about the dynamics of below ground parts of plants, it is not modeled.
Biomass growth and death are assumed to be dependent and can not occur
simultaneously. Biomass death only occurs in the winter and is simulated with a simple first
order equation (Equation 22). When biomass death begins it becomes standing dead plant
material at an exponential rate. The percentage of which biomass degrades (DEGBIO) and the
length of time (DAYSDEG) over which biomass degradation occurs is input by the model user.
The standing dead physically degrades to other sources at an exponential rate over the
course of one year. Since this plant mass is not lying on the ground, degradation occurs due to
physical processes such as rain and wind rather than microbial processes.
An option to account for removal of plant material from the system is available. This
option can be used to represent anthropomorphic removal, or any other removal from the system.
The day from the start of the season period, and the amount of removal are input by the user and
removed from the system accordingly.
5.
Nitrogen/Carbon/DO/Bacteria Relations:
As examined in the literature review, the N cycle is not an independent process, but one
whose reactions are dependent upon many factors. The C, DO, and microbiological cycles all
play a role in the rates and reactions of N processes which occur within the wetland system.
Each cycle was modeled separately, and the relationships between the respective cycles are
expressed in Figures 11 through 22. Each cycle will be examined individually, but the
relationships between them must always be accounted for. The work in the N, C, DO, and
bacteria relations are based on the work by Gidley (1995) and Parnas (1975).
72
6.
Carbon Submodel:
The C cycle consists of either five (SSF) or seven (FWS) state variables (Figures 11 and
12). The state variables are; biomass C (W), standing dead C (W), particulate organic C (POC,
W/B), dissolved organic C (DOC, W/B), and refractory C.
Biomass C and standing dead C components are intertwined heavily with the vegetative
submodel as the biomass C is calculated as:
BIOMASSJ = BIOMASSJ-1 + ((BIOMASSTJ
BIOMASSTJ-1) * BIOCCONT)*dt
(47)
where BIOMASS is the total plant C biomass (g C), and BIOCCONT is the fraction of C in
plants (g C/g biomass). The growth, death and removal of biomass C have already been
accounted for by the BIOMASST term from the vegetative cycle. The standing dead C values
are determined similarly:
STANDDCJ = STANDDCJ-1 + ((STANDDTJ STANDDTJ-1)
* BIOCCONT) DOCLEACHJ - PHYSDEGCJ)*dt
(48)
where STANDDC is the dead biomass that has not become litter (g C); DOCLEACH is the
leaching of standing dead to DOC (g C/day); and PHYSDEGC is the conversion of standing
dead C to homogeneous litter (POC) (g C/day). Standing dead physically degrades to other
sources at an exponential rate over the course of one year. Physical degradation is an input
source to the POC pool, which Jansson and Persson (1982) define as a homogenous organic
mixture of decaying plant litter, sloughed microbial cells and particulate influent wastewater
BOD. Over the first 30 days of winter, 15% of the plant nutrients are quickly lost by leaching to
DOC and DON (Heliotis and DeWitt, 1983; Kadlec, 1986).
73
Percolation/Infiltration Total
DONB
DOC percolation
DOC min\imob B
DOC Percolation/
Infiltration Total
HT grow th B
POC min\imob B
TOCB
DOC MT coef.
Runoff inflow
Length
DO Bottom
BOD runoff
influent
Water Bottom
HT yield B
WATER VOLUME
BOTTOM
HT death B
DOCB
TONB
BOD C fraction
Soluble BOD inflow
Width
DOC leaching
Leaching Rate
NS death B
BIOMASS
CARBON
POCB
Microbial death B
HT
Microbial C content
DOCW
Water Surface
DOC outflow
TOTAL
BIOMASS
Physical
Deg. Carbon
HI
Water Surface
DOC min\imob W
Outflow
POC size
POC settlling
STANDING
DEAD CARBON
Microbial C content
Biomass C content
DON SURFACE
TOCW
Water velocity
TONW
HB
HT grow th W
Microbial death W
POCW
DO SURFACE
POC min\imobw
BOD C fraction
HT yield W
Particulate BOD inflow
Peat accumulation
WATER VOLUME
SURFACE
Peat C content
POC outflow
PONW
REFC
FIGURE 11: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET
MODEL
74
BOD runof f
Runof f inf low
TOTAL BIOMASS
BOD C
f raction
W ater Volume
Bottom
DOC Percolation/
Inf iltration Total
DOC leaching
BIOMASS
CARBON
DOCB
Width
Leaching Rate
DOC min\imob B
Outf low
Length
STANDING DEAD
CARBON
DON BOTTOM
TOCB
Physical Deg.
Carbon
HT death B
TONB
Microbial C content
NS death B
HT grow th B
Microbial death B
POCB
DO BOTTOM
POC min\imob B
BOD C f raction
Peat accumulation
HT yield B
W ATER VOLUME
BOTTOM
PONB
REFC
HT=0; BOD runof f
inf luent conc.;
HT=1; BOD runof f
coef .
FIGURE 12: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE SUBMODEL FOR SSF WETLANDS OF THE SET-WET
75
MODEL
The mass balance for POC depends on which type of wetland is being examined. For
FWS wetlands the mass balances for the two pools are:
POCWJ = POCWJ-1 + (PBODINJ + MICRODWJ
- PEATCACWJ POCMINIWJ POCSETJ
+ POCREJ + PHYSDEGCJ POCOUTJ)*dt
(49)
and
POCBJ = POCBJ-1 + (POCSETJ + MICRODBJ
- POCMINIBJ PEATCACBJ -POCREJ)*dt
(50)
(51)
where POC(X) is the amount of POC in water volume (g); PBODIN is the particulate BOD
influx from catchment runoff (g/day); MICROD(X) is the sloughing of dead microbial cells to
POC (g/day); PEATCAC(X) is the accumulation of refractory solids from surface water (g/day);
POCMINI(X) is the conversion of POC to microbial biomass and carbon dioxide (g/day);
POCSET is the amount of POC that settles from POCW (g/day); POCRE is the amount of POC
that resuspends from POCB (g/day); POCOUT is the effluent POC amount (g/day); and
PHYSDEGC is the conversion of standing dead to POC (g/day).
The mass balance of DOC also differs for the two wetland types. For FWS wetlands the
mass balance pools are:
DOCWJ = DOCWJ-1 + (SOBODINJ DOCOUTJ
- DOCMINIWJ + DOCMTJ)*dt
(52)
and
DOCBJ = DOCBJ-1 + (DOCLEACHJ DOCMINIBJ
- DOCMTJ - DOCPERCJ)*dt
(53)
76
(54)
where DOC(X) is the amount of DOC in the water volume (g); SOBODIN is the soluble BOD
influx from catchment runoff (g/day); DOCOUT is the effluent DOC (g/day); DOCMINI(X) is
the conversion of DOC to microbial biomass and carbon dioxide (g/day); DOCMT is the amount
of DOC that diffuses at the interface (g/day); DOCLEACH is the leaching of DOC from standing
dead (g/day); and DOCPERC equals the amount of DOC lost/gained from infiltration/percolation
(g/day).
Since the model is designed for NPS pollutants, a general assumption of the chemical
composition of the material coming into the wetland system is difficult to determine. This value
is site-specific and can be estimated with the approach presented by McCarty (1975).
Microbial sloughing contributes to the POC content and is the product of the microbial C
content and the sum of heterotroph and autotroph death. Microbes are not considered to be a part
of POC until they die and are accounted for in the bacteria counts. It is assumed that microbial
death contributes only to POC, not DOC, since most microbes in wetland systems are associated
with plant litter and soil organic matter. Particulate BOD inflow is determined as:
PBODINJ = 1.4 * BODPFRAC * BODCFRAC * (BODINFCO
* WATINPUTJ + APFLOW * BODCONC)
(55)
where PBODIN is the particulate BOD flux from waste additions (g C/day); BODPFRAC is the
fraction of influent BOD that is in particulate form (g part. BOD5/g total BOD5); BODCFRAC is
the fraction of C in influent BOD (g C/g BOD5); BODINFCO is the BOD5 concentration in
catchment runoff (mg BOD5/l); WATINPUT is the amount of catchment runoff (m3); APFLOW
is the amount of point flow contributions (m3); and BODCONC is the BOD5 concentration in
point source (mg BOD5/l). Five day BOD is converted to ultimate BOD assuming k=0.25
(Tchobangolous and Burton, 1991).
77
For SSF wetlands, removal of particulate constituents such as PON and POC is assumed
to be 100%. This simplification is necessary for there are no mechanistic models available for
estimating particulate removal in porous substrates (Gidley, 1995). SSF wetland effluent TSS
values are generally low, thus minimal error should be introduced. For FWS wetlands,
particulate constituents of this type are subject to settling and resuspension like any other
particle. Settling and resuspension of POC is modeled as described by Equation 38.
The submodel of C and N immobilization, mineralization, and ammonification is based
on the work of Gidley (1995) and Parnas (1975). The total amount of C that is required for
heterotroph growth (HT) is calculated as heterotrophic growth (HTGROW) divided by the
heterotrophic yield (HTYIELD) where HTGROW has units of g microbes/day and HTYIELD
has units of g cells/g C. Heterotrophic yield is a function of DO concentration (Henze et al.,
1986; Tchobongolous and Burton, 1991).
Whether fractions of DOC and POC (or DON, PON, and NH4+ to be discussed later), are
utilized for microbial growth and energy are determined by the required substrate C:N ratio, as
defined by the microbe total C:N (MICTCN). The MICTCN is directly related to the DO
concentration (DOXYC) as (Parnas, 1975):
MICTCN = 80. - 10. * DOXYC
(56a)
(56b)
78
microbial growth is N limited and the amount of PON and POC utilized depends on the relative
fraction of PON to other N sources (PON/(TON +NH4)). The fraction of required C and N
derived from dissolved organic materials depends on the ratio of DON and TON + NH4. If
TOC/TON is less than MICTCN, microbial growth is C limited and the amount of PON and
DON that is utilized will depend on the relative proportions of POC and TOC. The fraction of
required C and N derived from dissolved components is proportional to DOC divided by TOC.
The respective C fraction utilized is proportional to the organic N utilization. The amount of
each N source that is immobilized depends on its relative fraction, with the excess organic N
(ON) being converted to NH4+ by ammonification. These guidelines help establish the following
relationships for DOC mineralization/immobilization (DOC min/imob) and POC
mineralization/immobilization (POC min/imob):
DOC min/imob = DON/TON * HTGROW/HTYIELD
(57a)
(when TOC/TON>MICTCN)
DOC min/imob = DOC/TOC * HTGROW/HTYIELD
(57b)
(when TOC/TON<MICTCN)
and
POC min/imob = PON/TON * HTGROW/HTYIELD
(58a)
(when TOC/TON>MICTCN)
POC min/imob = POC/TOC * HTGROW/HTYIELD
(58b)
(when TOC/TON<MICTCN)
A part of POC is considered refractory and is thus permanently lost to the soil through
peat accumulation. Peat accumulation is the product of the peat accumulation rate and the peat C
content. DOC is assumed to be 100 percent labile (Gidley, 1995). DOC accumulation occurs
because of DOC leaching, and soluble BOD inflow. Soluble BOD inflow is the contributions of
TOC that are not associated with POC and can be determined by changing the BODPFRAC
parameter in Equation 55 to (1-BODPFRAC). Microbes utilize DOC for energy and cell growth.
Outflow of DOC is calculated by multiplying the DOC concentration by the outflow.
79
7.
Nitrogen Submodel:
Processes that are always modeled by the N submodel include ammonification (W/B),
immobilization (W/B), nitrification (W/B), denitrification (W/B), and peat accumulation (W/B).
As stated before, there are flags that signify whether ammonia volatilization (W), atmospheric
deposition (W) and N fixation (W) will also be included in the overall N cycle modeling. These
are user-defined options, whereby all or none of the processes have to be included in the
simulation. Sorption of NH4+ is not modeled as it is assumed that sorbed NH4+ is still available
to attached microbes. The state variables include dissolved organic N (DON; W/B), particulate
organic N (PON; W/B), NH3 and NH4+-N (NH4; W/B), nitrate N (NO3; W/B), immobilized N
(W/B), and refractory N (Figures 13 and 14).
Influent DON enters the wetland through catchment runoff, point source inflow, percolation,
and atmospheric deposition (dry and wet):
DONINJ = (1 ONPARTF) * WATINPUTJ * ORGNINCJ
+ APFLOW * DONCONC
(59)
where DONIN is the incoming DON to wetland (g N/day); ONPARTF is the fraction of organic
N in particulate form (g PON/g TON); ORGNINC is the influent ON from catchment runoff (mg
ON-N/L); and DONCONC is the ON concentration from point sources (mg ON-N/L).
If the atmospheric deposition option is chosen and there is no precipitation, then the total
dry deposition for that day is the dry deposition rate (g DON/ m2) multiplied by the wetland
surface area. If precipitation does occur, then the input to the wetland is the total precipitation
amount multiplied by the concentration of DON in the rain.
Dissolved organic N also accumulates by leaching of N from the standing dead, due to
the physical degradation of standing dead biomass. It is assumed that accumulation occurs based
on a constant proportion of C to N in the plant biomass. Dissolved organic N leaching is
calculated as the product of the DOC leaching rate and the biomass C:N ratio (BIOMASS C:N).
Dissolved organic N outflow is determined by multiplying the wetland DON concentration by
the outflow. As previously described, a part of DON is reincorporated into microbial biomass
during the degradation of organic C; while the rest is wasted as ammonium. These relations are
as follows:
80
WATER
VOLUME
BOTTOM
Outflow
HT NO3 yield W
Point NO3 conc
Anaerobic HT grow th W
NO3 inf luent
NO3 SURFACE
Biomass C:N
biomass grow th
NS grow th W
nitrate uptake W
Nitrif ication W
IMMOBILIZED N
SURFACE
Plant up division
NS Yield W
Microbial N content
NH4 influent
HT NH4 immob. W
ammonium uptake W
NH4 SURFACE
PON
NO3 MT coef.
A. D. flag
pH
Direct precip.
pH factor
runoff inflow
NH4 outflow
Perc./Inf
Denitification B
Volatization
Wet A.D.
Dry A.D.
Wet A.D. DON
NO3 BOTTOM
Surface
Area
Ana
HT NO3 yield B
NO3 mass
transf er
Length
HT g
Water Surface
Volatilization rate
Water Bottom
Outflow
Nitrif ication B
WATER VOLUME
SURFACE
HT yield W
Plant
DOC SURFACE
NS Yield B
NH4 MT coef .
NH4 MT
DON ammoniafication W
Point ON conc.
TON W
DON influent
NH4 BOTTOM
DON outflow
TOCW
HT yield B
DON SURFACE
NH4 Pe
Inf. Co
Surface area
DON ammoniafication B
Length
DON immobilization W
ON particulate fraction
TON B
Nitrogen Fixation
DON MT coef .
DON BOTTOM
Microbial N Content
Biomass C:N
Width
HT grow th W
N leaching
Water Surface
TONW
Water Bottom
HT NH4 immob. W
NH4 SURFACE
DON SURFACE
IM M OBLIZED N
SURFACE
Perc./Inf. Total
DON Per
PON SURFACE
DON Perc./
Inf. Conc.
FIGURE 13: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
81
WATER
VOLUME
BOTTOM
HT death W
NS death W
physical degradation C
Biomass C:N
POC SURFACE
death
IMMOBILIZED N
SURFACE
PON immobilization
mmob. W
HT grow th W
A. D. flag
Width
TOCB
PON ammonification W
oef.
Direct precip.
TONW
HT yield W
HT grow th W
PON SURFACE
Length
PON influent
Point ON conc.
ON Particulate fraction
Anaerobic HT grow th B
O3 yield B
POC SURFACE
Perc./Inf. Total
fication B
NO3 Perc./Infil.
HT death B
NO3 BOTTOM
NS death B
physical degradation C
Biomass C:N
biomass grow th
POC BOTTOM
nitrate uptake B
death B
Nitrification B
IMMOBILIZED N
BOTTOM
Plant up division
HT NH4 immob. B
ammonium uptake B
NS Yield B
HT grow th B
Peat N content
Microbial N content
NH4 BOTTOM
dB
PON immobilization B
NH4 Perc./
Inf. Conc.
PON ammonification B
Perc./Inf. Total
NH4 Perc./Infil.
DOC BOTTOM
TONB
HT yield
PON BOTTOM
REFRACTORY N
Peat accumulation rate B
HT grow th B
POC BOTTOM
PON settling
tion B
TOCB
PON resuspension
DON immobilization B
HB
DON BOTTOM
Microbial N Content
Water Surface
Water Bottom
HT grow th B
HT
Water velocity
TONB
HT NH4 immob. B
HI
PON falling rate
MANNC
PON size
Perc./Inf. Total
DON Perc./Infil.
DON Perc./
Inf. Conc.
NH4 BOTTOM
DON BOTTOM
PON BOTTOM
FIGURE 13 (CONT.): RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN
CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
82
NO3 Percolation/
Infiltration Conc.
WATER BOTTOM
HT NO3 yield B
Anaerobic HT growth B
Perc./Inf. Total
NO3 Perc./Infil.
HT
NO3 BOTTOM
NO3 outflow
NO3 influent
HT=0; influent
runoff NH4 conc.
HT=1; NH4 runoff
coeff.
biomass growth
NS Yield B
nitrate uptake B
Volatization
pH factor
Microbial N content
NH4 BOTTOM
A. D. flag
PON
NH4 outflow
NH4 Perc./
Inf. Conc.
Direct
Point NH4 conc.
Perc./Inf. Total
NH4 Perc./Infil.
HT yield B
WATER BOTTOM
DON ammonification
DOC BOTTOM
Point ON conc.
TOCB
TON B
DON influent
HT NH4 immob. B
ammonium uptake B
NH4 influent
runoff inflow
IMMOBILIZED N
BOTTOM
Plant up division
Volatization rate
Length
Biomass C:N
NS growth B
Nitrification B
pH
Width
WATER VOLUME
BOTTOM
Outflow
Denitrification B
DON BOTTOM
DON immobilization B
DOC leaching
ON particulate fraction
DON outflow
Biomass C:N
DON Perc./
Inf. Conc.
Nitrogen Fixation
N leaching
Length
HT growth B
DON Perc./Infil.
T
HT NH4 immob. B
Perc./Inf. Total
Width
NH4 BOTTOM
WATER BOTTOM
FIGURE 14: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE SUBMODEL FOR SSF WETLANDS OF THE SETWET MODEL
83
HT growth B
WATER VOLUME
BOTTOM
HT death B
NS death B
physical degradation C
Biomass C:N
biomass growth
POC BOTTOM
ake B
death B
IMMOBILIZED N
BOTTOM
PON immobilization B
HT NH4 immob. B
Peat N content
HT growth B
bial N content
TONB
PON ammonification B
Perc./Inf. Total
HT yield B
PON BOTTOM
REFRACTORY N
Peat accumulation rate B
HT growth B
POC BOTTOM
Outflow
PON influent
A. D. flag
DOC BOTTOM
be Total C:N B
WATER VOLUME
BOTTOM
Point source flow
DON immobilization B
Width
Point ON conc.
catchment
runoff inflow
DON outflow
ON Particulate fraction
Direct precip.
Length
Microbial N Content
T growth B
TONB
HT NH4 immob. B
Perc./Inf. Total
NH4 BOTTOM
DON BOTTOM
PON BOTTOM
84
FIGURE 14 (CONT.): RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE SUBMODEL FOR SSF WETLANDS
OF THE SET-WET MODEL
(60a)
(when TOC/TON>MICTCN)
DONIM = DOC/TOC * (MICRONC * HTGROW - HTNH4IM)
(60b)
(when TOC/TON<MICTCN)
PONIM = PON/(TON+NH4)*MICRONC*HTGROW
(61a)
(when TOC/TON>MICTCN)
PONIM = POC/TOC *(MICRONC * HTGROW - HTNH4IM)
(61b)
(when TOC/TON<MICTCN)
HTNH4IM = MICRONC *HTGROW *NH4 / (TON +NH4)
(62)
DONAM = 0
(63a)
(63b)
*DON/DOC DONIM
(when TOC/TON < MICTCN)
PONAM = 0
(64a)
(64b)
85
86
assumed that the movement of the N2 gas from the soil system is instantaneous, as the diffusion
of the product is not modeled. Denitrification is modeled as the quotient of anaerobic
heterotroph growth and heterotroph NO3- yield. Other intermediates such as NO and N2O are
produced by microbes, but are not considered in the simulation.
8.
Oxygen is added to the wetland through influent runoff, influent point sources, DO
concentrations in precipitation, percolation, by plants and re-aeration. Re-aeration with the
atmosphere is modeled only for FWS wetlands as it is assumed that oxygen transfer with the
atmosphere is negligible in SSF systems because the water surface is below the substrate. The
mass balance for DO in the FWS wetland surface is:
DOXYWJ = DOXYWJ-1 + (DOINFJ + MTDOXYJ
+ MTFWSJ HTRESPWJ NSRESPWJ - DOOUTJ)*dt
(65)
(66)
(67)
where DOXY(X) is the amount of DO in water volume (g); DOINF is the oxygen additions from
runoff and point inflow (g/day); MTDOXY is the diffusion of DO from one pool to the other
(g/day); MTFWS is the DO re-aeration from atmosphere (g/day); HTRESP(X) is the oxygen loss
by heterotroph growth (g/day); NSRESP(X) is the oxygen loss due to autotroph growth (g/day);
DOOUT is the effluent DO (g/day); DOPERC are additions from percolation (g/day); and
BIOFLUXB is the oxygen flux from rootzone aeration by plants (g/day).
87
Precipitation input
Runoff inf low
HT=0, inf luent DO conc.;
HT=1, DO runoff conc.
Outf low
At. DO conc.
Inf luent DO
DISSOLVED
OXYGEN
SURFACE
Aerobic HT
growth W
DO outf low
NS growth W
WATER
VOLUME
SURFACE
HT respiration W
NS respiration W
HT DO
y ield W
NS DO
y ield W
DO mass transf er
Surf ace area
DO MT coef .
Biomass f lux
Aerobic HT
growth B
DISSOLVED
OXYGEN
BOTTOM
NS growth B
HT respiration B
HT DO
y ield W
WATER
VOLUME
BOTTOM
DO Percolation/
Inf iltration Conc.
DO Percolation/
Inf iltration
NS respiration B
Percolation/
Inf iltration Total
NS DO
y ield B
FIGURE 15: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
88
Precipitation input
Runoff inflow
HT=0, influent DO conc.;
HT=1, DO runoff conc.
Surface area
Biomass flux
DO Percolation/
Infiltration
Percolation/
Infiltration Total
At. DO conc.
Influent DO
Aerobic HT
growth W
Point flow DO conc.
DO Percolation/
Infiltration Conc.
DISSOLVED
OXYGEN
BOTTOM
DO outflow
Outflow
NS grow th W
WATER
VOLUME
BOTTOM
HT respiration W
HT DO
yield W
NS respiration W
NS DO
yield W
FIGURE 16: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE SUBMODEL FOR SSF
WETLANDS OF THE SET-WET MODEL
(SEE APPENDIX D FOR EXPLANATION OF FIGURES SYMBOLS)
89
Re-aeration to the wetland is modeled using a general two-film theory that is based on
mass transfer. The mass transfer of oxygen from air to water can be presented as (Jorgensen,
1983):
dC K L A
=
* (C S C )
dt
V
(68)
where KL is the mass transfer coefficient (m/day); A is the surface area through which diffusion
takes place (m2); V is the volume of water being re-aerated (m3); Cs is the oxygen saturation
concentration (g/m3); and C is the oxygen concentration in surface water volume (g/m3).
Biomass flux is the product of the biomass oxygenation rate and the wetland surface area.
It is assumed that there is a uniform vegetation stand throughout the wetland system and that
plants transport of oxygen to the wetland bottom occurs at a constant rate through the growing
season. This assumption is based on the theory that processes such as Knudsen diffusion are
responsible for rootzone aeration (Grosse, 1989). The biomass oxygenation rate is linearly
increased from zero to the maximum biomass oxygenation rate starting from the first day of the
growing season and decreasing linearly to zero at the end of the growing season. Although not
linked through any proportionality ratios, the oxygenation rate is ramped to correspond with the
vegetative growth model.
Chemical oxidation of reduced iron and manganese, and autotrophic and heterotrophic
respiration are two of the processes that consume dissolved oxygen in a wetland (Reddy and
Patrick, 1983). Heterotroph (HTRESP) and Nitrosomonas (NSRESP) respiration are modeled as
proportional to microbial growth:
HTRESP =
AEHTGRW
HTDOY
(69)
NSRESP =
ANHTGRW
NSDOY
(70)
where HTRESP is the oxygen loss due to heterotrophic growth (g O2/day); AEHTGRW is the
growth of aerobic heterotrophs (g microbes/day); HTDOY is the oxygen yield of aerobic
90
heterotrophs (g microbes/g O2); NSRESP is the oxygen loss due to Nitrosomonas growth (g
O2/day); ANHTGRW is the growth of anaerobic autotrophs (g microbes/day); and NSDOY is
the oxygen yield of Nitrosomonas bacteria (g microbes/g O2).
Most constructed wetlands have gravel substrates and secondary wastewater additions are
generally aerobic; therefore reduced iron, manganese and sulfide concentrations are considered
negligible. The dissolved oxygen outflow from the system is the product of the wetland DO
concentration and the effluent volume from the system.
9.
Bacteria Submodel:
The bacteria submodel accounts for the microbial growth and associated activities in the
model. To simplify the description, the model is described according to the natural breakdown
between autotrophic (AT) and heterotrophic (HT) bacteria.
a.
Autotrophic Dynamics
The state variable NITROSO (NS, W/B) represents the AT population within the
wetland. Although there are other AT species that carry out nitrification, Nitrosomonas and
Nitrobacter are the dominant species responsible for the rate limited reaction (Wheaton et al.,
1991). Changes in the population of NS are due to NS growth and NS death (Figures 17 and 18).
Growth rate of NS is described using Monod dual substrate limitation kinetics (Gidley,
1995):
NH 4
= max *
NH 4 + K NH
4
DO
*
DO + K DO
(71)
where is the actual NS specific growth rate (day-1); max is the NS maximum growth rate
(day-1); NH4+ is the ammonium concentration (mg/L); KNH4 is a NS NH4+ half saturation
constant (mg/L); DO is the wetland dissolved oxygen concentration (mg/L); and KDO is the NS
DO half saturation constant. This expression modifies the NS growth rate when oxygen (the
electron acceptor) or ammonium (the electron donor) is limiting.
91
NS grow th W
NS death W
NITROSOMONAS
SURFACE
Water Temp. W
NS Temp. Factor W
NH4 SURFACE
NS max.
grow th
rate W
NS DO
half sat.
con. W
NS death rate W
WATER VOLUME
SURFACE
NS NH4
half sat.
con. W
DISSOLVED
OXYGEN
SURFACE
FIGURE 17: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC BACTERIA
CYCLE IN FWS WETLAND SURFACE WATER
NS grow th B
NS death B
NITROSOMONAS
BOTTOM
Water Temp. B
NS Temp. Factor B
NH4 BOTTOM
NS max.
grow th
rate B
NS DO
half sat.
con. B
NS death rate B
WATER VOLUME
BOTTOM
NS NH4
half sat.
con. B
DISSOLVED
OXYGEN
BOTTOM
FIGURE 18: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC BACTERIA
CYCLE IN FWS AND SSF WETLAND SUBSTRATE
92
Monod models are steady state and do not describe microbial growth well in unsteady
conditions (Boyd, 1978). Benefield and Molz (1984) used the Monod model to develop a
biofilm model for an idealized plate assuming pseudo steady state conditions. Gidley (1995)
also assumed pseudo steady state conditions were adequate for Monod kinetics in wetlands.
Besides substrate limitations, temperature and pH also limit microbial growth. The
optimum temperature range for NS is from 15 to 35 C (Reddy and Patrick, 1983; Wheaton et
al., 1991), and while growth still occurs outside of this range, it is at a reduced rate. To account
for these temperature effects, the NS temperature factor is included. In the optimum temperature
range, the temperature factor equals 1.0. The factor value linearly decreases to 0 C (the lower
limit) and to 40 C (upper limit), once outside of the optimum range. Fyock (1977), and Reddy
and Patrick (1983) have reported that the optimum pH range for NS growth is 6-9. Wetlands
tend to drive pH toward neutrality (pH = 7), therefore a pH factor is not included. NS growth is
calculated by:
NSGROW j = * NSTEMPF * NITRSOMONAS j 1
(72)
Heterotrophic Dynamics
Heterotrophic (W/B) dynamics is also modeled in the bacteria submodel (Figures 19 and
20). Monod kinetics is used to model the changes in microbial growth rate caused by substrate
limitations. For HT bacteria, the electron acceptor is either oxygen or NO3- while the electron
donors are the C compounds in the system. Heterotrophs can use electron acceptors other than
oxygen, including NO3-, sulfate, iron and manganese (Gidley, 1995); however, only NO3- is
modeled and is assumed to be the only other available electron acceptor when DO concentrations
drop below 1-2 mg/L.
93
HT grow th W
Aerobic max.
grow th rate W
HT death W
HETEROTROPHS
SURFACE
Aerobic HT
grow th W
HT death rate W
Anaerobic HT
grow th W
TOC
NO3 SURFACE
WATER VOLUME
SURFACE
Water temp. W
HT NO3 half
sat. con. W
HT temp. fac. W
HT organics
half sat. con. W
HT DO half sat
constant W
DISSOLVED
OXYGEN
SURFACE
Anaerobic max.
grow th rate W
Anaerobe
fraction W
FIGURE 19: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER
HT grow th B
Aerobic max.
grow th rate B
HT death B
HETEROTROPHS
BOTTOM
Aerobic HT
grow th B
HT death rate B
Anaerobic HT
grow th B
TOC
NO3 BOTTOM
WATER VOLUME
BOTTOM
Water temp.
HT NO3 half
sat. con. B
HT temp. fac.
HT organics
half sat. con. B
HT DO half sat
constant B
DISSOLVED
OXYGEN
BOTTOM
Anaerobic max.
grow th rate B
Anaerobe
fraction B
FIGURE 20: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE
94
HT bacteria are facultative, meaning they can survive under both aerobic or anaerobic
conditions. SET-WET determines the fraction of the HT bacteria which utilize either aerobic or
anaerobic conditions based on the DO concentrations in the system (Gidley, 1995):
ANFRAC = 0.8 * ( DOXYC * 0.75)
(73a)
(73b)
TOC
AH = max, AH *
TOC + K TOC
DO
*
DO + K DO
K DO
TOC
*
(74)
(75)
where maxis the maximum growth rate (day-1); TOC is the total organic C concentration (mg/L);
KTOC is the HT organics half saturation constant (mg/L); DO is the dissolved oxygen
concentration (mg/L); KDO is the HT DO half saturation constant (mg/L); AH are aerobic
heterotrophs; and ANH are anaerobic heterotrophs.
The same optimum temperature and pH rules apply to HT bacteria, as they did for the AT
bacteria. Total HT growth is the sum of the aerobic HT and anaerobic HT growth. The HT
95
growth and death are modeled in the exact manner as the AT bacteria, with the differing
respective parameter values.
10.
Sediment Submodel:
The sediment cycle simulates sediment particles and any other desired suspended solid in
the wetland system (Figure 21). As stated earlier, the sediment model does not work in
conjunction with the SSF wetland simulation. For the SSF system, it is assumed that the
retention of particulates in the system is 100%, yet there is no account to the loss in porosity for
the wetland gravel substrate. Between season periods, there is an ability to change the porosity
values in the SSF system. If the system did have high loading of sediment to the system, the
model may have difficulty in modeling the NCOB cycle in the SSF system due to the clogging of
the wetland substrate. It is advised that the SSF system be used in conjunction with another
BMP designed to lower sediment concentrations, such as a detention basin, if using a SSF
system to control NPS pollution, to prevent clogging of the SSF wetland substrate.
The sediment model has been designed to accept up to five particle size categories. Each
category requires inputs for sediment diameter (SEDSIZE, mm), fall rate (SEDFALL, m/s),
initial amount in the water surface (SEDINITW, g), initial amount in the wetland bottom
(SEDINITB, g), and the percentage of the total incoming amount based on weight (SEDPER).
The final input category may be used to represent the organic suspended solids in the wetland
system. All organic material must be represented in this single category, as there is an associated
decomposition component for the final category. The decomposition component in the sediment
submodel does not run concurrently with the NCOB cycle. Decomposition is not modeled based
on the bacterial consumption modeled in the NCOB cycle, but with a simple user defined first
order rate equation. This was a necessary step to allow SET-WET the ability to simulate the
nutrient components independently.
Total incoming sediment amounts are either directly input or determined with a
combination of the SCS curve method and runoff concentrations (Equation 37). The total
incoming amount from catchment runoff and point flow is multiplied by SEDPER to determine
the incoming amounts for each respective category (SEDINW).
The balances for sediment in the wetland are (adapted from Christensen et al., 1994):
96
WATER VOLUME
SURFACE
Outlet = 1,2,3,5
Outflow height
Outlet = 4
Pump outlet
Outflow
Physical Degradation
Outlet
SEDIMENT
SURFACE, X
Sed. X outflow
Sed. X falling
rate
Sed. Resusp. rate
HI
Sed. X,
Resuspension
HT
Sed. X, deposition
HI
HT
Manning's coeff.
HB
Decomposition
Rate
Sed. X, Reynold's
number 1 check
Water velocity
Sed. Reynold's
number 1b check
SEDIMENT
BOTTOM, X
Sediment Decomposition
FIGURE 21: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE SEDIMENT CYCLE SUBMODEL FOR FWS WETLANDS OF THE SETWET MODEL
97
(76)
and
SEDQTYBC,J = SEDQTYBC,J-1 + (SEDDEPC,J
+SDECOMPSEDCAT,J RESUSPC,J)*dt
(77)
where SEDQTYW is the amount of sediment in the surface water (g), SEDQTYB is the amount
of sediment in the surface bottom (g), SEDINW is the amount of incoming sediment (g/day);
RESUSP is the amount of sediment resuspended from the surface bottom (g/day); SEDOUT is
the sediment outflow (g/day); SEDDEP is sediment deposition (g/day); C is the sediment
category; and SEDCAT is the total number of sediment categories.
Resuspension and settling of sediment particles are determined with Equation 38.
Sediment outflow is based on the fall rate and is treated like other particle constituents in the
wetland. If the fall rate exceeds the free water surface height, then sediment outflow is
considered to be zero, while for rates less than the free water surface height, a ratio of removal is
based on outflow and surface water volume.
11.
Phosphorous Submodel:
The P cycle (Figure 22) is dependent on the sediment cycle, and does not operate if the
sediment cycle is not included in the simulation. The basis of the model is that all of the
suspended sediment particles provide surface area to which P can be attached and consequently,
settled, resuspended, or transformed. There are four pools for the P cycle; particulate and
dissolved for both the surface and bottom of the wetland. The budgets for the pools are as
follows (adapted from Christensen et al., 1994):
For surface water dissolved P:
DTPHOSWJ = DTPHOSWJ-1 + (DISPHOSIJ
+ PMINPPTJ DPHOSOUTJ + PHOSMTJ)*dt
(78)
98
(79)
(80)
(81)
where DTPHOSW is the total dissolved P in surface water (g P); DTPHOSB is the total
dissolved P in wetland bottom (g P); PPHOS is the particulate P in surface water (g P); BTPHOS
is the particulate P in the wetland bottom (g P); DISPHOSI is the influent dissolved P (g P/day);
PMINPPT is the P mineralization from the particulate pool (g P/day); MASSTP is the mass
transfer of dissolved P (g P/day); DPHOSOUT is the effluent dissolved P outflow (g P/day);
PRMINBPT is the P remineralization from bottom particulate pool (g P/day); PPHOSRES is the
resuspension of particulate P from bottom (g P)/day; PPHOSOUT is the particulate P effluent
outflow (g P/day); PPHOSSET is the particulate P settling from surface (g P/day); and
DEADPHOS is the contribution of P from physical degradation of standing dead (g P/day).
Incoming particulate P amounts are modeled using either the Freundlich isotherm
(Equation 30), the Linear isotherm (Equation 31) or as direct input. If using either the Linear or
Freundlich isotherms, the dissolved P concentration is used to determine the particulate P
concentration. Knowing the weight of sediment, the surface area for each particle class is
determined and the particulate P is assumed to be partitioned among each particle class
99
DP inf luent
Point DP conc.
DISSOLVED
PHOSPHOROUS
(DP) SURFACE
PARTICULATE
PHOSPHOROUS
(PPS) SURFACE; X
SEDIMENT
SURFACE, X
Length
Sed. X., deposition
W ATER VOLUME
SURFACE
DP Mass transf er
SEDIMENT
SURFACE, X
SEDIMENT
BOTTOM, X
PPS,Xsettling
PP,X resuspension
Sed. X Resuspension
# of sed. categories
W ATER VOLUME
BOTTOM
Percolation/
Inf iltration Total
DP Percolation/
Inf iltration Rate
DISSOLVED
PHOSPOROUS
(DP) BOTTOM
Plant DP uptake
Biomass Grow th
PARTICULATE
PHOSPHOROUS
(PPB) BOTTOM; X
PPB ratio
DP Percolation/
Inf iltration
Physical degradation
FIGURE 22: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE PHOSPHOROUS CYCLE SUBMODEL FOR FWS WETLANDS OF THE
SET-WET MODEL
100
according to the incoming surface area ratio for the particle class, divided by the total of all
particle surface areas.
Mass transfer of dissolved P is modeled in the form of Equation 39. Mineralization and
remineralization are modeled as first order equations for each particular category of particles.
Resuspension and settling of P corresponds to the amount of sediment particles and is related to
the ratio of each category that resuspends and settles. This also applies to the attached P for P
outflow. The contributions of P made by physical degradation (DEADPHOS) is based on the
plant:phosphorous ratio (BIOMPP). The quotient of physical degradation and BIOMPP
determines DEADPHOS. As the suspended particles are broken into categories, the addition of
each category amount allows the determination of the total amounts in each pool.
12.
Deltaht Submodel:
The DELTAHT submodel determines the change in height, respective to the zero datum,
of the wetland surface bottom. This applies only to FWS wetlands and is activated whenever the
sediment or N cycles are being simulated.
If only the N cycle is being modeled the changes in the wetland bottom surface height
(HT) are determined by:
PEATACRB J
PEATDENS *1000
DELTAHT =
(LENGTH *WIDTH )
(82)
where DELTAHT is the change in height of the soil top layer (m); PEATACRB is the bottom
peat accumulation rate (g/day); and PEATDENS is the peat density (kg/m3). If only the sediment
cycle is being modeled the changes in HT are determined by:
SEDTOTAL
SEDSPG *1000000
DELTAHT =
(LENGTH * WIDTH )
(83)
101
where SEDSPG is the sediment specific gravity (kg/m3), and SEDTOTAL is the total change in
sediment mass of the wetland bottom (g). If both options are activated then DELTAHT is
determined by:
SEDTOTAL
PEATACRBJ
+
(84)
removed from the flow chart for sake of clarity. As Figure 7 indicates, the main program calls
every submodel and thus the number of connections required to represent these interactions
would make Figure 23 impossible to view or understand. Instead, the main program is not
shown in Figure 23, and the connections shown represent the flows from one submodel to
another without referring to the actual flow from the main program for each section.
C.
vegetative, P and sediment cycles of a wetland system. The N, C, DO, and bacteria cycles are
linked and cannot be run independently. These cycles are referred to as the NCOB cycle to
denote their dependence. The P cycle is dependent on the sediment cycle, and neither the
sediment nor P cycles may be simulated with the SSF wetland option. Any simulation run that
includes modeling the NCOB, sediment or P cycles requires that vegetative growth be simulated.
The SET-WET model is designed so that the hydrologic component may be run independently or
concurrently with any or all of the NCOB, sediment and P cycles.
The SET-WET model development was based on the work of Gidley (1995), Parnas
(1975) and Christensen et al. (1994). The SET-WET model is written in Fortran 77 to facilitate
linking with existing NPS models, such as ANSWERS. The SET-WET model is based on
options. The program is developed such that a main program calls and manages the cycles and
options that need to be simulated, based on input values. SET-WET may use two different forms
102
START OF
SIMULATION
BASE
HYDSTR
VEGSTR
CARSTR
NITCYCLE=0
NITCYCLE =1
SEDSTR
BACSTR
PCYCLE=1
NITSTR
PHOSSTR
SEDCYCLE=1
OXYSTR
PCYCLE=0
SEDCYCLE=0
HYDTIME
VEGTIME
CARTIME
NITCYCLE=0
NITCYCLE =1
SEDTIME
BACTIME
PCYCLE=1
NITTIME
PHOSTIME
SEDCYCLE=1
OXYTIME
PCYCLE=0
SEDCYCLE=0
WETTYPE=0
DELTAHT
J<NUMTMPER
J=NUMTMPER
M<NUMSTPER
M=NUMSTPER
END OF
SIMULATION
FIGURE 23: FLOW CHART FOR CALLING ORDER OF SET-WET MODEL FROM MAIN CODE THROUGH
SUBROUTINES
of input: 1) input on a daily basis with known values or, 2) input estimated with a combination
of the SCS runoff curve method and runoff concentration coefficients. There are options to
include modeling of groundwater interactions, and point source additions for the hydrologic
cycle, and also N fixation, volatilization and atmospheric deposition for the N cycle. Time steps
are managed with season and time period loops, where each season period can be designated a
103
certain number of daily time periods. Designed to act like a continuously stirred tank reactor, the
model assumes all incoming nutrients are evenly mixed throughout the entire volume. Dissolved
nutrient effluent concentrations from the wetland system are determined with a ratio of the
hydrologic outflow and the total amount of water volume in the system. Particulate effluent
concentrations are determined with the ratio of the fall rate and water depth in the wetland, in
conjunction with the ratio of the hydrologic outflow and the total amount of water in the wetland.
The model allows for both free water surface (FWS) and sub-surface flow (SSF) wetlands to be
modeled and, although initially developed to help with the design of constructed wetlands, SETWET may also be applied to model the functions of natural wetlands.
SET-WET differs from many existing wetland models in that it uses a systems approach
and links the many interactions of the various nutrient cycles in a wetland system. It accounts
for C and N interactions, as well as for effect of DO levels upon microbial growth. It also
directly links microbial growth and death to the consumption and transformations of nutrients in
the wetland system. Many previous models have accounted for these interactions with zero and
first order rate equations, which assume that rates are dependent only on initial concentrations.
SET-WET was developed to be as general as possible, but there were assumptions made
during model development which might not apply to all wetland locations. The model assumes
that the FWS system always has free lying surface water, and that the SSF system never floods.
This may limit its ability to model extreme climactic occurrences during drought or flood
periods. It does not account for snow or snow melt, which may limit its use to warmer climate
areas. The model assumes that the dissolved and particulate concentrations in both the surface
water and substrate water are evenly mixed through the respective volumes, which is known to
not physically be the case.
Plant growth is assumed to be of constant composition throughout the wetland, and is not
limited by lack of nutrients or water supply. In addition, growth and death cannot occur at the
same time, therefore, the model does not account for turnover of plant material during the year.
Oxygen transfer by plants is considered to be constant through the growing season, and it is also
assumed that biomass and bacteria prefer NO3- rather than NH4+. The use of a simple plant
model will affect the release rate of nutrients to the system from plant decomposition, but more
complicated plant growth models input requirements are too data intensive to be used.
104
Modeling of adsorption of NH4+ is not conducted as it is assumed that the amount that is
adsorbed is still available for microbial use. The model assumes that P is attached to all of the
particulates in the system based on the surface area of the particles. In addition, SSF wetland
aeration is assumed to be negligible because the water surface is below the substrate, and SSF
wetland PON and POC removal is assumed to be 100%.
In addition to assumptions made directly concerning wetland interactions and processes,
there are assumptions made by previous researchers that were incorporated into the model. The
Thornthwaite method for estimating ET assumes that the soil moisture in the area is not limiting
and that air temperature is the primary controller of ET. Another assumption is induced by using
mass transfer coefficients to model diffusion, which assumes that changes in concentrations
between the two pools (surface and substrate water) are limited to the small part of the systems
volume connecting boundaries.
Although SET-WET does incorporate a few assumptions into model development, it still
offers the flexibility of modeling various designs for a desired wetland. However, these
assumptions are noted for the benefit of the user.
105
IV.
Model Evaluation
Another objective of this study was to evaluate the proposed model by comparing its
predictions with field data collected from representative constructed wetland site(s). This
objective was difficult to satisfy because there are a lack of field data related to the control of
NPS pollution using FWS constructed wetlands. Wetlands have been created (Silverman, 1989;
Daukas et al., 1989; Teague et al., 1997) whose intent is to control NPS pollution, but data
collection has not included all relative parameters (NH4+, BOD5, DO, etc) for substantial time
periods (1-2 years) that would allow the use of data for calibration and validation of SET-WET.
The SET-WET models performance in simulating the functions of FWS wetlands was
evaluated with data collected at a wetland site located in Benton, Kentucky. This was not an
ideal data set because the Benton wetlands treat municipal waste, but there were no available
NPS pollution wetland data sets. The models performance for SSF wetlands was not evaluated;
however, previous research by Gidley (1995) examined SET-WETs basic approach towards
modeling SSF wetlands. Model evaluation procedures included the calibration and validation of
the model, performing two types of statistical analyses, conducting a sensitivity analysis, and
using SET-WET to demonstrate its application for design of wetlands. Overall the model
performed well, even though the data used for evaluation were somewhat limited.
A.
1.
wetland located in Benton, Kentucky. As described by Choate et al. (1990) and summarized by
Kadlec and Knight (1996), the Benton wetlands were designed to upgrade and polish municipal
effluent from an existing lagoon for 5000 people. There are three parallel wetland cells (two
FWS, one SSF) of equivalent size (333 m by 45m) located in Benton, but only one of the FWS
cells (cell 2) was used to examine SET-WETs performance.
The wetland cell consists of substrate made of native clay and impermeable clay lining of
3 to 4.5 m thickness, which eliminated infiltration or percolation effects. It was planted with
106
Scripus cyperinus (L.) Kunth (woolgrass bulrush), Scripus validius Vahl. (softstem bulrush), and
Typha latiofolia L. (cattail). Influent and effluent samples were collected either monthly or
bimonthly for a variety of parameters. Measurements for most of the parameters were conducted
from March 30, 1998 until September 6, 1990; however, total suspended solids (TSS), and
dissolved and total P were discontinued after April 26, 1989. Listed in Table 5 is the data set
used as input for the calibration and validation of the SET-WET model.
Incoming input daily water flow, nutrient, and sediment values were determined by linear
interpolation between the monthly data points. This was the case for all daily parameters,
excluding the daily weather data such as precipitation and air and water temperature.
Precipitation and air temperature were obtained from NOAA (National Oceanographic and
Atmospheric Administration) records at a site located in Padukah, Kentucky; approximately 25
miles northeast of Benton (NOAA, 1988: NOAA, 1989). Water temperature was estimated from
a linear regression of the air temperature and measured water temperature. With the known air
temperature, the daily water temperature was determined.
2.
Model Calibration:
Data for one year were used to calibrate and validate the SET-WET model. This time
period was chosen because the nutrient and hydrologic input data needed for SET-WET were
available. During this time period, only thirteen data points for the hydrologic and nutrient
TABLE 5: MEASURED INFLOW VALUES TO WETLAND CELL 2 IN BENTON, KENTUCKY USED FOR
VALIDATION AND CALIBRATION OF SET-WET MODEL.
Flow
Date
(m^3/d)
4/27/88 213.2
5/25/88 567.9
6/29/88 339.0
7/27/88 350.4
8/30/88 472.5
9/28/88 403.6
10/25/88 478.0
11/29/88 1263.1
12/13/88 523.9
1/24/89 877.1
2/22/89 1816.1
3/28/89 1016.4
4/26/89 633.0
DO
mg/l
2.4
8.1
2
8.3
16.8 *
2.4
6.2
10.8
19.1 *
13
10
8.3
9
BOD5
mg/l
30
34
18
19
20
14
23
22
38
24
23
24
26
NH3-N
mg/l
12.0
12.0
5.2
3.1
0.1
4.1
9.8
3.8
4.8
3.1
1
2.0
3.0
NO3-NO2
mg/l
0.02
0.02
0.02
0.01
0.09
0.01
0.36
0.74
0.4
0.51
0.57
0.25
0.24
Org-N
mg/l
5
8
11
12.3
9.9
9.6
7.1
6.5
6.3
1.5
4.3
6
12
TKN
mg/l
17
20
16.2
15.4
10
13.7
16.9
10.3
11.1
4.6
5.3
8
15
TSS
mg/l
30
49
63
110
59
54
31
54
31
9
29
53
Dis-P
mg/l
3
5.5
6.8
4.1
5.4
5.6
2.3
2.9
0.7
0.9
1.1
2.3
Tot-P
mg/l
4.5
6.8
7.8
5.6
9.7
5.7
6.6
3.4
3.3
3.7
1.4
2.6
2.9
107
conditions were collected, which may have limited the models predictions as explained later.
For a couple of the parameters (TSS and DP), data collection was missing, and a few of the DO
measurements were removed for values that were physically impossible (See Table 5). The data
were divided into three groups, two (4/27/88 to 7/27/88 and 1/24/89 to 4/26/89) of which were
used for calibration, while the final time frame (7/27/88 to 1/24/89) was used to validate the
model. The calibration periods were selected to evaluate how the model performed during a
warmer and colder time period.
Listed in Table 6 are the input values used for the calibration and validation periods for
the simulations performed at the Benton wetland site. The initial parameter values used in the
model simulations, as well as the values used for the calibrations and validation are listed in
Table 6. Model input parameters were estimated from domestic wastewater, wetland and
microbiology literature. It was assumed that the make-up of the water entering the system was
similar to that of domestic wastewater since the wetland cells were designed for municipal
treatment. More information on the determination of the bacterial parameters may be found in
Gidley (1995). The hydrologic component of SET-WET was calibrated first, followed by the
NCOB, the sediment and then the P cycles.
The design of the wetland as described by Choate et al. (1990) allowed the basic design
and hydrology of the wetland system to be easily described in the model. No detailed
description of the wetland outflow system was available, except it was known to consist of an
outflow pipe. The size and contraction of the pipe was determined through calibration of the
hydrologic cycle.
To estimate the current solids content of the wetland substrate, it was assumed that 5% of
the pore space was already filled and that the accumulated material had the same bulk density as
peat (0.11 Mg/m3) (Reed et al., 1994; Johnston, 1991). This gives a total mass of accumulated
solids of 19,320 kg. These solids were then broken into labile and refractory fractions with 10%
assumed to be labile, and the remaining 90% to be refractory and unavailable to microbial
decomposition. Each of these fractions were divided into C (0.80) and N contents (0.025)
respectively (Gidley, 1995). These values were used as the initial values of POC, PON, REFC,
and REFN. The BOD particulate fraction was estimated based on the ratio of TSS to BOD5 site
influent data.
108
TABLE 6: INPUT PARAMETER VALUES AND SOURCES FOR CALIBRATION AND VALIDATION PERIODS.
Parameter
BASE
LENGTH
WIDTH
HO
HB
HTI
HII
SO
HYDROLOGY
PORPEAT
OUTLET
HOUT
AREAPIPE
CONTC
WATINPUT
PRECIPR
AIRTEMP
DAYLEN
Initial
Value
Value
1st. Cal.
Value
2nd Cal.
Value
Typical
Validation Range
333.5
43.9
2
0
0.6
0.001
333.5
43.9
2
0
0.6
1.115
0.001
333.5
43.9
2
0
0.6
1.127
0.001
333.5
43.9
2
0
0.6
1.01
0.001
0.3
0.3
0.3
0.3
3
1.056
0.0105
0.8
3
1.057
0.0095
0.9
3
1.056
0.01
0.85
3,800,000
8,000,000
300
2,000
32
5,000
8,000,000
300
2,000
32
8,635,000
6,000,000
300
2,000
32
110
0.7
50
50
0.4
0.2
0.7
50
50
0.4
0.2
0.99
20
3
0.7
50
50
0.4
0.2
0.99
20
3
5.-15.
Site Data
Site Data
Site Data
Site Data
Site Data
Site Data
.01-.0001 Reed , 1988
.2-.4
BIOMASS
BIOINIT
STANDIN
PEATACRW
PEATACRB
PEATDENS
300
2,000
110
PRATEUP
BIODENS
STDDENS
PBIOUW
PSTDUW
DEGBIO
DAYSDEG
BIOMGRR
0.5
50
50
0.3
0.3
0.99
20
10
BACTERIA
NITROSIW
NITROSIB
HETEROIW
HETEROIB
NDRATEW/NDRATEB
18,300
6,000
1,000
1,000
183,000
18,000
2,400
2,400
5,300
40,000
10,000
25,000
53,000
250,000
230,000
240,000
.1 / .1
.002 / .002 .002 / .002 .002 / .002
.05-.15
NDOHSATW/NDOHSATB
1. / 1.
1. / 1.
1. / 1.
1. / 1.
.4-1.3
NMAXGRW/NMAXGRB
NNH4HSCW/NNH4HSCB
1
1
.005 / .005
1
.005 / .005
1
.005 / .005
1
.3-2.0
.2-5.0
Reference
See discussion
"
Johnston, 1991
Johnston, 1991
Reed et al., 1994
Johnston, 1991
Calibration
Kadlec and Knight, 1996
Kadlec and Knight, 1996
Calibration
Calibration
See discussion
See discussion
Kadlec, 1991
Hammer, 1984
See discussion
"
"
"
Henze et al., 1986
Brune and Tommasso, 1991
Henze et al., 1986
Fritz et al., 1979
Henze et al., 1986
Henze et al., 1986
Brune and Tommasso, 1991
Grady and Lim, 1980
109
Table 6 (cont.): Input parameter values and sources for calibration and validation periods.
Parameter
AEMAXGRW/AEMAXGRB
ANMAXGRW/ANMAXGRB
HTDRW/HTDRB
Initial
Value
6
4
0.05
Value
1st. Cal.
.05 / .01
.05 / .01
0.001
Value
2nd Cal.
.05 / .01
.05 / .01
0.001
Value
Validation
.05 / .01
.05 / .01
0.001
HTDOHSCW/HTDOHSCB
0.5 / 0.5
0.5 / 0.5
0.5 / 0.5
.1-2.0
HNO3HSCW/HNO3HSCB
0.15
0.15 / 0.15
0.15 / 0.15
0.15 / 0.15
.1-.2
HORGHSCW/HORGHSCB
50
50 / 50
50 / 50
50 / 50
15-100
.1-.9
0.0-1.0
.2-.8
0.0-1.0
.2-.8
0.0-1.0
.2-.8
0.0-1.0
0.0-0.4
-
20,000,000
20,000
30,000
75,000
3,000,000
0.47
0.8
0.5
0.01
16,000,000
50,000
60,000
200,000
2,500,000
0.47
0.8
0.5
0.01
ANFRACW/ANFRACB
HTTEMPFW/HTTEMPFB
WATTEMPW
WATTEMPB
CARBON
REFCINIT
DOCINITW
DOCINITB
POCINITW
POCINITB
BIOCCONT
BODCFRAC
BODPFRAC
LEACHR
MICROBEC
PEATCC
POCFALL
POCRES
MANNC
RESTHC
POCSIZE
MTCDOC
POCCOUT
BODINFCO
NITROGEN
DONINITW
DONINITB
IMMINITW
IMMINITB
NH4INITW
NH4INITB
NO3INITW
NO3INITB
PONINITW
PONINITB
REFNINIT
BIOMCN
BIOMPN
HTNO3YW/HTNO3YB
MICRONC
13,900,000 13,900,000
*
30,000
*
40,000
*
107,000
1,545,000
2,000,000
0.47
0.47
0.8
0.8
0.5
0.5
0.01
0.53
0.8
0.7
0.1
2
0.02
0.2
0.00005
1
0.53
0.8
0.45
0.001
2
0.01
0.25
0.00004
0.75
0.53
0.8
0.45
0.001
2
0.01
0.15
0.00004
0.3
0.53
0.8
0.45
0.001
2
0.01
0.2
0.00004
0.3
*
*
*
*
*
*
*
*
*
48,000
435,000
23.5
95
3.29 / 3.29
0.125
14,000
14,000
50,000
2,000,000
27,000
20,000
2,100
5,000
16,250
250,000
435,000
23.5
95
3.29 / 3.29
0.125
5,500
10,000
700,000
7,500,000
20,250
10,000
6,000
11,000
10,000
100,000
565,000
23.5
95
3.29 / 3.29
0.125
5,000
10,000
700,000
9,900,000
30,000
55,000
400
4,000
15,000
300,000
500,000
23.5
95
3.29 / 3.29
0.125
Typical
Range
3.-12.
2.-8.
.02-.08
Reference
Grady and Lim, 1980
Grady and Lim, 1980
Henze et al., 1986
Brune and Tommasso, 1991
Fritz et al., 1979
Henze et al., 1986
Strand et al., 1985
Henze et al., 1986
Brune and Tommasso, 1991
Henze et al., 1986
Grady and Lim, 1980
Brune and Tommasso, 1991
Henze et al., 1986
See discussion
Site Data
Site Data
See Discussion
"
"
"
"
Boyd, 1978
Stoichiometry
Stoichiometry
Kulshretha and Gopal, 1982
Polunin, 1982
Stoichiometry
Brady, 1984
Calibration
Calibration
Cussler, 1997
Calibration
Site Data
See Discussion
"
"
"
"
"
"
"
"
"
"
Boyd, 1978
Boyd, 1978
Stoichiometry
Stoichiometry
110
Table 6 (cont.): Input parameter values and sources for calibration and validation periods.
Parameter
NSYIELDW/NSYIELDB
Initial
Value
ONPARTF
PEATNC
PONRES
PONFALL
MTCDON
MTCNH4
MTCNO3
PONSIZE
RESTHN
PONCOUT
ORGNINC
NH4INC
NO3INC
DISSOLVED OXYGEN
DOINITW
DOINITB
HTDOYW /HTDOYB
NSDOYW/NSDOYB
DOCONCP
MTDOX
MTFWSDOC
DOXYCSAT
BIOOXRB
DOCONCIN
SEDIMENT
SEDCAT
SEDRES
SEDSIZE (SEDCLASS)
SEDFALL (SEDCLASS)
*
*
0.81
0.084
3
1
.25-.5
.3-.8
SEDINITW (SEDCLASS)
SEDINITB (SEDCLASS)
SEDSPG (SEDCLASS)
SEDPER (SEDCLASS)
RESTHICK
MANNC
DECOMPR
PSEDDEP
SEDINT
PHOSPHOROUS
DTPHOSIW
DTPHOSIB
BTPHOSI
PPHOSI
PMINPPC
PRMINBPC
ADSORP
MTCPHOS
BIOMPP
PHOSCON
DISPHOSC
PPHOSCAL
2.65
1
0.1
2
0.1
0.5
Value
1st. Cal.
0.3 / 0.3
Value
2nd Cal.
0.3 / 0.3
0.6
0.025
0.01
0.25
0.00002
0.00006
0.00006
0.05
0.01
0.375
0.6
0.025
0.01
0.25
0.00002
0.00006
60,000
15,000
.15 / .15
.02 / .02
0.001
0.0001
0.00008
8.5
22,000
15,000
.15 / .15
.02 / .02
0.001
0.0001
0.00008
8.5
2
0.1
.25-.5
.3-.7
80,000300,000
1,000,00023,000,000
2.65
1
0.0015
2.5
0.1
0.5
21,000
25,000
250,000
10,000
0.05
0.001
2
0.00006
300
0
0.05
0.01
0.375
Value
Validation
0.3 / 0.3
0.6
0.025
0.01
0.25
0.00002
0.00006
0.00006
0.05
0.01
0.375
15,000
15,000
.15 / .15
.02 / .02
0.001
0.0001
0.00008
8.5
2
2
0.1
0.1
.25-.5
.25-.5
.3-.6
.3-.7
10,00010,00020,000
190,000
12,000,000- 1,100,00028,000,000 26,000,000
2.65
2.65
1
1
0.0005
0.001
2.5
2.5
0.1
0.1
0.5
0.5
.
10,000
15,000
250,000
20,000
0.05
0.0005
2
0.00006
300
0
Typical
Range
.1-.34
38,000
40,000
250,000
7,150
0.05
0.00075
2
0.00006
300
0
Reference
Henze et al., 1986
Tchobangolous and Burton, 1991
Site Data
Gidley, 1995
Calibration
Cussler, 1997
Cussler, 1997
Cussler, 1997
Calibration
Calibration
Site Data
Site Data
Site Data
0.81
0.084
See discussion
"
Stoichiometry
Stoichiometry
Cussler, 1997
Cussler, 1997
Cussler, 1997
Site Data
Site Data
See Discussion
"
"
"
site data
site data
Site Data
Calibration
Calibration
Calibration
Site Data
See Discussion
"
"
"
Calibration
Calibration
Site Data
Cussler, 1997
Kadlec and Knight, 1996
Site Data
Site Data
Site Data
111
The initial amount of biomass in the system was determined using the time period (days)
between the start of the growing season and the start of the simulation, multiplied by the
assumed constant growth rate of the system. There were no collected estimates on biomass
amounts in the system, therefore a conservative growth estimate of 3 g/m2-day was assumed
(Kadlec, 1991). The initial amount of standing dead material in the system was similarly
determined using a decay rate of .0126 day 1 multiplied by the length of time between the start
of simulation and the end of the previous growing season (Gidley, 1995).
The initial values of DOC, DON, NH4+, NO3-, DO, BOD5, DP, TP, and TSS were
determined using the initial water volume and the site effluent concentrations at the start of each
simulation period. The water volume in the system was multiplied by the respective effluent
concentrations to determine the nutrient loads in each system. Since no site data were collected
concerning the concentrations of the nutrients in the wetland substrate, these amounts were
determined through calibration.
The initial amount of heterotroph and autotroph cells in the wetland substrate were
estimated based on values given in Stevenson (1982). NS populations are in the range of 106
107 cells/gram of soil. It was assumed that NS were present at a density of 106 cells/gram peat.
Generally there are a larger number of heterotrophs present than autotrophs due to their higher
growth rate, thus a density of 108 heterotroph cells/gram peat was assumed (Stevenson, 1982).
An average cell diameter of 1 m and a density of 1 g/cm3 were used to determine the mass of
bacteria cells. It was assumed that microbes had roughly the same density as water and that they
were spherical. To convert from the number of cells per soil mass to an estimate of local initial
mass of cells, an average bulk density (1.25 Mg/m3) was used (Brady, 1984). To estimate the
number of bacteria in the wetland surface water it was assumed that the number of bacteria in the
water was 10% of the amount in the substrate. It has been shown that up to 30% of the amount
of bacteria in the substrate, may be located in the surface water (Hatano et al., 1994) for
wastewater constructed wetlands.
There was no site data describing the particle size distribution of the incoming sediment
to the wetland cell; however, since the influent to the wetland system initially passed through one
treatment system, it was assumed that many of the larger particulate particles and aggregates had
112
already settled. Since the breakdown or size of the particles was unknown, only one small
particle category (0.2 mm) was used to describe sediment.
The model was determined to be calibrated through visual analysis of the graphical
output (Figures 24a to 32b) of the predicted and observed values for the hydrology and nutrients.
Since SET-WET models the interactions between various nutrient cycles, it can be difficult to
simultaneously calibrate all of the cycles. This is because a parameter change for one cycle can
affect up to eight other cycles. For example, the rate of mineralization will affect the amount of
organic N and NH4+ in the system. In addition, this rate will affect the ratio of TON to TOC in
the system, which in combination with the MICTCN ratio determines if the system is N or C
limited. These effects will be further explored with the sensitivity analysis.
Table 7 lists the differences between the observed and predicted values for the Benton
site. The data shows that the predicted values of SET-WET mostly follows the trends for the two
calibration periods. From visual analysis, the hydrology (Figures 24a and 24b), BOD5 (Figures
29a and 29b), dissolved P (Figures 31a and 31b), and total P (Figures 32a and 32b) predictions
values followed the observed values trends the best.
Ammonium concentrations were over-predicted for the first calibration period, while the
NO3-, organic N, and DO effluent concentration predictions did not completely match the trends
for the observed values. The objective of the calibration process was to examine the effects of
both warmer and colder seasons in the study area. Since there were limited data for each
calibration period (4 data points each), each calibration period was modeled as one season
period, eliminating the ability to change parameter values during the simulation. The high NH4+
predictions may be attributable to the extremely high influent concentrations on April 27, 1988
and May 25, 1988 for the Benton wetlands. These two data points were averaged over the entire
time period, meaning for a one-month period, the model was receiving input concentrations of 12
mg/L for NH3-N. These measurements may have been spikes in the observed measurements,
which would contribute to the high NH4+ predictions by the model. Any errors in the input
measurements will be amplified for all nutrients because so few data points were available.
For the second calibration period (1/24/88 to 4/26/89), the DO concentration predictions
were over predicted. Once again this can probably be attributed to the lack of sufficient data
points used for input. The measured incoming DO counts, for the entire second calibration
period were at high levels. Another reason DO values may have been over-predicted is due to
113
800
Predicted Hydrologic Outflow
700
Outflow (m 3/d)
600
500
400
300
200
100
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 24A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
2500
Outflow (m 3/d)
2000
1500
1000
500
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 24B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
114
12
10
Predicted Ammonium
0
4/24/88
Measured Ammonium
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 25A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
12
10
Predicted Ammonium
Measured Ammonium
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 25B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
115
0.6
Predicted Nitrate
Measured Nitrate
Effluent Nitrate
Concentration (mg/l)
0.5
0.4
0.3
0.2
0.1
0
4/24/88
5/4/88
5/14/88 5/24/88
6/3/88
6/13/88 6/23/88
7/3/88
7/13/88 7/23/88
Date
FIGURE 26A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
1.4
Predicted Nitrate
Measured Nitrate
1.2
1
0.8
0.6
0.4
0.2
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 26B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
116
0
4/24/88
5/4/88
5/14/88 5/24/88
6/3/88
6/13/88 6/23/88
7/3/88
7/13/88 7/23/88
Date
FIGURE 27A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
4.5
4
3.5
3
2.5
2
1.5
1
Predicted Organic Nitrogen
0.5
Measured Organic Nitrogen
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 27B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
117
15
12
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 28A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
10
9
8
7
6
5
4
3
2
Predicted Dissolved Oxygen
1
0
1/12/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 28B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
118
20
18
16
14
12
10
8
6
4
Predicted BOD5
Measured BOD5
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 29A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
18
16
14
12
10
8
6
4
Predicted BOD5
2
Measured BOD5
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 29B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
119
60
Predicted TSS
Measured TSS
50
40
30
20
10
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 30A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
25
Predicted TSS
Measured TSS
20
15
10
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 30B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
120
8
7
6
5
4
3
2
Predicted Dissolved Phosphorous
1
Measured Dissolved Phosphorous
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 31A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
4.5
4
3.5
3
2.5
2
1.5
1
Predicted Dissolved Phosphorous
0.5
Measured Dissolved Phosphorous
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 31B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
121
7
6
5
4
3
2
Predicted Total Phosphorous
1
Measured Total Phosphorous
0
4/24/88
5/4/88
5/14/88
5/24/88
6/3/88
6/13/88
6/23/88
7/3/88
7/13/88
7/23/88
Date
FIGURE 32A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
1/12/89
2/1/89
2/21/89
3/13/89
4/2/89
4/22/89
5/12/89
Date
FIGURE 32B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
122
TABLE 7: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED VALUES
FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR THE CALIBRATED,
PREDICTED VALUES.
Me a sure d
Pre dicte d
Diffe re nce
Me a sure d
Pre dicte d
NH4
Hydrologic
Hydrologic
in
NH4
Outflow
Outflow
Hydrologic
Efflue nt
Efflue nt
Diffe re nce
Da te
(m^3)
(m^3)
Outflow
Conc. (mg/l) Conc. (mg/l)
NH4
657.52
696.22
38.70
5
3.58
-1.42
4/27/88
621.05
518.56
-102.49
3.4
9.92
6.52
5/25/88
318.32
190.90
-127.42
0.4
10.57
10.17
6/29/88
305.97
332.33
26.36
5.3
7.72
2.42
7/27/88
1/24/89
2/22/89
3/28/89
4/26/89
864.97
1853.03
1071.55
598.78
858.29
1815.93
816.99
530.25
-6.68
-37.10
-254.56
-68.53
3.1
2.7
2
11
2.62
1.36
1.86
2.54
-0.48
-1.34
-0.14
-8.46
Da te
4/27/88
5/25/88
6/29/88
7/27/88
Me a sure d
NO 3
Efflue nt
Conc. (mg/l)
0.37
0.54
0.01
0
Pre dicte d
NO 3
Efflue nt
Conc. (mg/l)
0.279
0.013
0.013
0.013
Diffe re nce
NO 3
-0.092
-0.527
0.003
0.013
Me a sure d
Orga nic-N
Efflue nt
Conc. (mg/l)
3.45
4.5
4.2
1.5
Pre dicte d
Orga nic-N
Efflue nt
Conc. (mg/l)
2.67
2.96
3.87
4.34
Diffe re nce
Orga nic-N
-0.78
-1.54
-0.33
2.84
1/24/89
2/22/89
3/28/89
4/26/89
0.82
0.34
0.15
0.01
0.778
0.537
0.323
0.039
-0.042
0.197
0.173
0.029
1.5
1
3
4.2
1.20
1.74
2.55
4.27
-0.30
0.74
-0.45
0.07
Da te
4/27/88
5/25/88
6/29/88
7/27/88
Me a sure d
DO
Efflue nt
Conc. (mg/l)
8.75
11.50
0.30
0.20
Pre dicte d
DO
Efflue nt
Conc. (mg/l)
7.10
5.01
4.13
4.63
Diffe re nce
DO
-1.65
-6.49
3.83
4.43
Me a sure d
BOD5
Efflue nt
Conc. (mg/l)
16.50
18.00
6.00
19.00
Pre dicte d
BOD5
Efflue nt
Conc. (mg/l)
11.16
15.11
15.04
11.07
Diffe re nce
BOD5
-5.34
-2.89
9.04
-7.93
1/24/89
2/22/89
3/28/89
4/26/89
4.80
3.80
1.20
0.20
5.14
7.82
5.77
6.48
0.34
4.02
4.57
6.28
9.00
9.00
12.00
16.00
4.92
11.85
12.84
13.33
-4.08
2.85
0.84
-2.67
Da te
4/27/88
5/25/88
6/29/88
7/27/88
Me a sure d
TSS
Efflue nt
Conc. (mg/l)
40.00
53.00
23.00
27.00
Pre dicte d
TSS
Efflue nt
Conc. (mg/l)
39.79
22.07
21.36
21.45
Diffe re nce
TSS
-0.21
-30.93
-1.64
-5.55
Me a sure d
Dis.-P
Efflue nt
Conc. (mg/l)
3.20
5.80
5.40
3.10
Pre dicte d
Dis.-P
Efflue nt
Conc. (mg/l)
4.61
6.25
5.96
2.79
Diffe re nce
Dis.-P
1.41
0.45
0.56
-0.31
1/24/89
2/22/89
3/28/89
4/26/89
2.00
2.00
20.00
14.00
2.59
6.14
6.61
3.34
0.59
4.14
-13.39
-10.66
1.40
1.20
3.70
4.10
1.30
2.83
2.74
3.34
-0.10
1.63
-0.96
-0.76
123
the estimation of water temperature in the system. Water temperature was estimated from the air
temperature, with the assumption that the water temperature in the surface and substrate water
were the same. These values may have been incorrect, due to the limited data points used in the
regression. In addition, air temperature fluctuates appreciably more than water temperature.
Water temperature is a factor in bacterial growth. If the water temperature is below 0 C,
bacterial growth will cease, limiting the amount of oxygen consumption in the wetland. There
were no bacterial counts made in the system thus it is unknown if bacteria counts are close to
actual values.
Additionally, matching the predicted and simulated values for sediment was difficult for
the calibration periods. The hydrologic outflow from the system is used to determine the water
velocity in the wetland. The wetland water velocity is then used to determine if resuspension in
the system occurs based on the water velocity being greater than the critical velocity (Equation
32). There is no function controlling the amount of sediment that is lifted up (resuspended) from
the wetland substrate based on the speed above the critical velocity. A set amount is lifted up
whenever the critical velocity is exceeded; this makes the predictions less accurate.
The calibrated values input did not greatly deviate from the values previously
determined through literature, except for parameters directly involving bacterial growth and
oxygen transfer. For the microbial parameters, the most dramatic changes involved the
maximum growth rates, which were changed by two order of magnitudes (from 6.0 and 4.0. to
0.05 and 0.05 for HT bacteria; 1.0 to 0.005 for AT bacteria). The literature values are typical for
conventional wastewater treatment systems, therefore it is not unreasonable for microbial growth
to be slower in a less controlled environment. Additionally, the heterotroph and autotroph yields
and death rates were each decreased significantly. These values would correspond with the
changes associated with the lowering of the growth parameters. If the death parameter were not
lowered to correspond to the decrease in microbial growth parameters, then the bacteria in the
system would completely die out. Furthermore, the microbial yields needed to be changed as the
growth of the microbes is decreased with the smaller growth parameter; therefore, more oxygen
per bacteria is consumed since there is less competition.
124
3.
Model Validation:
SET-WET was validated by comparing the models predicted results with the observed
data for the period 7/27/88 to 1/24/89. In the validation process, the initial amounts of each
respective nutrient in the system were determined in the same manner as the calibration
procedure, by multiplying the effluent concentrations with the initial water volume in the system.
The input parameter values for the validation period were determined by taking the average of
the two calibration period parameter values (Table 6). On 08/30/88 (16.8 mg/L) and 12/13/88
(19.1 mg/L), the measured DO concentrations were above the maximum DO solubility under
atmospheric conditions (14.62 mg/L at 0 C; Reed et al., 1994). These values were removed
from the validation process. Figures 33-41 display the model predictions and the observed data
for the outflow and the effluents of NH4+, NO3-, organic N, DO, BOD5, TSS, DP and TP
concentrations for the validation period. Table 8 lists the measured values, predicted values, and
the difference between the two values for the hydrology outflow, and the NH4+, NO3-, organic N,
DO, BOD5, TSS, and DP effluent concentrations.
1600
1400
Outflow (m 3/d)
1200
1000
800
600
400
200
0
7/16/88
8/5/88
8/25/88
9/14/88
10/4/88
10/24/88 11/13/88
12/3/88
12/23/88
1/12/89
2/1/89
Date
FIGURE 33: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
125
10
Predicted Ammonium
Measured Ammonium
8
7
6
5
4
3
2
1
0
7/16/88
2/1/89
Date
FIGURE 34: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
1
0.8
Predicted Nitrate
Measured Nitrate
0.6
0.4
0.2
0
7/16/88
8/5/88
2/1/89
Date
FIGURE 35: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR NITRATE EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
126
8/5/88
8/25/88
9/14/88
10/4/88
10/24/88 11/13/88
12/3/88
12/23/88
1/12/89
2/1/89
Date
FIGURE 36: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
7
6
5
4
3
2
8/5/88
8/25/88
9/14/88
10/4/88
10/24/88 11/13/88
12/3/88
12/23/88
1/12/89
2/1/89
Date
FIGURE 37: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
127
20
16
12
Predicted BOD5
0
7/16/88
Measured BOD5
8/5/88
8/25/88
9/14/88
2/1/89
Date
FIGURE 38: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
30
Predicted TSS
Measured TSS
25
20
15
10
0
7/16/88
8/5/88
8/25/88
9/14/88
10/4/88
10/24/88 11/13/88
12/3/88
12/23/88
1/12/89
2/1/89
Date
FIGURE 39: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL SUSPENDED
SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
128
7
6
5
4
3
2
1
0
7/16/88
8/5/88
2/1/89
Date
FIGURE 40: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
10
8
7
6
5
4
3
2
1
0
7/16/88
8/5/88
8/25/88
9/14/88
10/4/88
10/24/88
11/13/88
12/3/88
12/23/88
1/12/89
2/1/89
Date
FIGURE 41: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
129
TABLE 8: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED VALUES
FOR THE HYDROLOGY AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR THE VALIDATED,
PREDICTED VALUES.
Date
7/27/88
8/30/88
9/28/88
10/25/88
11/29/88
12/13/88
1/24/89
Measured
Hydrologic
Outflow
(m^3)
332.33
473.75
332.33
560.24
1355.81
864.91
858.29
Predicted
Hydrologic
Outflow
(m^3)
345.64
424.41
371.21
492.28
1274.87
631.60
856.48
Date
7/27/88
8/30/88
9/28/88
10/25/88
11/29/88
12/13/88
1/24/89
Measured
NO3
Effluent
Conc. (mg/l)
0.01
0.01
0.02
0.17
0.55
0.76
0.82
Predicted
NO3
Effluent
Conc. (mg/l)
0.06
0.01
0.00
0.03
0.54
0.54
0.46
Date
7/27/88
8/30/88
9/28/88
10/25/88
11/29/88
12/13/88
1/24/89
Measured
DO
Effluent
Conc. (mg/l)
0.20
1.40
3.60
2.50
4.00
3.30
4.80
Predicted
DO
Effluent
Conc. (mg/l)
3.94
7.83
5.04
5.62
6.59
9.34
7.42
Date
7/27/88
8/30/88
9/28/88
10/25/88
11/29/88
12/13/88
1/24/89
Measured
TSS
Effluent
Conc. (mg/l)
27.00
11.00
9.00
2.00
4.00
3.00
2.00
Predicted
TSS
Effluent
Conc. (mg/l)
27.62
2.47
2.53
2.60
4.15
3.89
3.34
Difference
in
Hydrologic
Outflow
-13.31
49.34
-38.88
67.96
80.94
233.31
1.81
Measured
NH4
Effluent
Conc. (mg/l)
5.30
8.60
8.40
9.00
4.60
3.80
3.10
Predicted
NH4
Effluent
Conc. (mg/l)
4.36
3.54
2.83
6.22
5.53
5.02
4.99
Difference
NH4
0.94
5.06
5.57
2.78
-0.93
-1.22
-1.89
Difference
NO3
-0.05
0.00
0.02
0.14
0.01
0.22
0.36
Measured
Organic-N
Effluent
Conc. (mg/l)
1.50
4.90
7.20
2.40
1.00
2.40
1.50
Predicted
Organic-N
Effluent
Conc. (mg/l)
1.54
4.28
3.95
3.26
2.85
2.77
1.22
Difference
Organic-N
-0.04
0.62
3.25
-0.86
-1.85
-0.37
0.28
Difference
DO
-3.74
-6.43
-1.44
-3.12
-2.59
-6.04
-2.62
Measured
BOD5
Effluent
Conc. (mg/l)
19.00
11.00
9.00
5.00
8.00
7.00
9.00
Predicted
BOD5
Effluent
Conc. (mg/l)
14.28
11.85
9.24
9.48
11.78
15.62
14.19
Difference
BOD5
4.73
-0.85
-0.24
-4.48
-3.78
-8.62
-5.19
Difference
TSS
-0.62
8.53
6.47
-0.60
-0.15
-0.89
-1.34
Measured
Dis.-P
Effluent
Conc. (mg/l)
5.40
5.30
5.80
4.30
2.80
2.80
1.40
Predicted
Dis.-P
Effluent
Conc. (mg/l)
5.52
6.36
5.62
4.59
2.56
2.64
1.24
Difference
Dis.-P
-0.12
-1.06
0.18
-0.29
0.24
0.16
0.16
130
The data set used to calibrate and validate the SET-WET model was not ideal, as the use
of monthly linear interpolation between the observed data points contradicts the principle idea of
NPS pollutions randomness and dependence upon climatic occurrences. Since the Benton
wetland site is developed for municipal waste treatment, however, this problem may have been
mitigated. The ideal data set would consist of daily data points for all concerned hydrologic and
nutrient parameters over a minimum of two-years, but no such data were available. This type of
data set would allow the seasonal as well as long term comparison of model performances.
As evident in Figures 33-41, the predicted trends for the validated period match the
observed values fairly well for various parameters, except for the NH4+ (Figure 34) and DO
(Figure 37) predictions. The trend for predicted NH4+ concentrations does not follow the
measured values for the period of middle of August 1988 to the end of October 1988. During
this period, the models predictions are low, thus implying that the conversion of organic N
through mineralization is not sufficiently handled by the model for this period. However when
one examines the organic N concentration (Figure 36), there is not an overestimation of values;
therefore, the model may be under-predicting organic N amounts due to a lack of accounting for
biomass degradation and possible additions from N fixation. Another possible explanation for
the extremely low NH4+ predictions is the very low influent NH4+ concentrations during this time
frame. The values are comparably low and as stated earlier, since the data points are so few, any
major errors in measurements during data collection will have a significant effect on model
input.
The inability of the model to match DO concentrations is a cause for concern. This
concern stems from the fact that many of the bacterial growth and bacteria uptake rates are
directly and indirectly affected by oxygen concentrations in the system. It is possible that the
assumption of a constant oxygen transfer rate by plants during the growing season may be
incorrect, as the DO concentrations are much higher than observed.
The dissolved concentrations associated with the total P effluent (Figures 40 and 41)
comprise a significant percentage of the total P. These values suggest that there is a significant
amount of particulate P buried in the system, which is in accordance with findings of previous
researchers (Kadlec and Knight, 1996).
To confirm that the model is performing proper mathematical calculations, a mass
balance of various hydrologic and nutrient pools of the models predictions were performed.
131
The cumulative inflows and outflows from the system were determined for every pool in the
system. The difference between these two values was compared with the change in storage in
the system, which is the difference between the final and initial amounts for each pool. The
largest percent difference between these values was 1%, indicating that the model sufficiently
accounted for the interactions of each pool.
From visual analysis of the data points it appears that the data matches very well
considering the data limitations. Further support will be lent by the statistical analyses.
B.
Statistical Analysis:
Statistical analyses were performed on the differences between the observed
measurements and models predicted values for the hydrology, and nutrient parameters. The
SAS software package proc univariate procedure was used to perform the Wilcoxon signed
rank test on the data (Ott, 1993). In the test, the null hypothesis (H0) was that there are no
differences between the observed and predicted values, while the alternative hypotheses (H1) is
that there are significant differences between the values. Using a two-sided test, with an alpha
value of 0.10, it was determined whether the model predictions were similar to the measured
values.
Ordinarily, the objective of a statistical test is to try to reject the null hypothesis, giving
fairly strong support to the alternative hypothesis. The Wilcoxon test is not designed in this
manner, therefore, we do not want to reject the null hypothesis. Failure to reject the hypothesis
means acceptance of H0, indicating that the observed and predicted values are not statistically
different from each other. A p-value that is greater than the alpha cut-off value of 0.10 would
result in the failure to reject the null hypothesis. The confidence level in this comparison
increases as the reported p-value approaches the maximum value of 1.0.
Table 9 presents the p-values for the differences between the observed measurements and
predicted model values for selected outflow parameters. For the hydrology, NO3-, NH4+, organic
N, BOD5, TSS, dissolved and total P outflows, we failed to reject the null hypothesis, as their pvalues are greater than 0.10 indicating that the measured and predicted values were statistically
similar. The only parameter for which the null hypothesis was rejected was DO, whose p-value
was 0.016. These results must be examined closely however, as there were only seven values
132
TABLE 9: P-VALUES AND RESULTS OF THE WILCOXON SIGNED RANK TEST PROCEDURE FOR DIFFERENCES
BETWEEN THE MEASURED AND VALIDATED, PREDICTED VALUES OF WETLAND EFFLUENT IN BENTON,
KENTUCKY.
Outflow Parameter
Hydrology
Nitrate
Ammonium
Organic Nitrogen
Dissolved Oxygen
BOD 5
Total Suspended Solids
Dissolved Phosphorous
Total Phosphorous
P-value
0.156
0.375
0.297
0.938
0.016
0.109
0.938
0.999
0.688
H0
Accept
Accept
Accept
Accept
Not Accept
Accept
Accept
Accept
Accept
in the data set used for each comparison. The Wilcoxon signed rank test is more reliable when
more observations are used. For the BOD5 (p=0.109), and hydrology (p=0.156) comparisons,
the p-values are barely above the cutoff value, meaning that caution should be used in failing to
reject the null hypothesis. On the other hand, the p-values for the TSS (p=0.938), ON (p=.0938),
and DP (p=0.999) are close to one, indicating a higher confidence in the acceptance of the null
hypothesis because the values are symmetrical around the observed values. Results of the
statistical analyses indicate that the SET-WET model predicted values that are similar to the
observed data.
In addition to the Wilcoxon signed rank test, linear regression was also performed on the
output values. The regression analysis was used to compare the measured and predicted values
for the nutrients and hydrology of the wetland. For this analysis, our null hypothesis (H0) was
that the slope coefficient between the observed and predicted values is zero, while the alternative
hypothesis (H1) was that the slope coefficient is not equal to zero. The ideal situation, which is
referred to as the 1:1 line, would be a line that crosses the y intercept (B0) at zero and has a slope
(B1) of one. This line represents the situation in which the observed and predicted values are
exactly the same.
Table 10 lists the results of the linear regression analysis. The values for B0 and B1 are the
linear regression of the simulated (independent variable) and observed (dependent variable) data
points, respectively. Because SET-WET is a deterministic model, the simulated values are used
as the independent values since the model will always predict these values with the same data
input. There is variability and uncertainty in the observed values since there are differences in
the observed measurements depending on how, when and who takes and analyses the samples
133
(Brannan, 1999). The human error introduced into the observed values due to the collection and
analysis of a sample dictate that the observed values are stochastic and therefore should be the
dependent variable.
Analysis of the linear regression results is not as straight forward as the non-parametric
analysis. P-values of less than 0.05 indicate that the analysis of the observed and simulated
results do not have a slope coefficient equivalent to zero. Independently, these results tell us
little, but if analyzed in conjunction with the slope coefficient and R2 value, we may be able to
distinguish which parameters were statistically similar. If the slope value is near 1.0 (idealized),
and the R2 value is close to 1.0, indicating a good fit of observed values to the regressed line, we
can be more confident in the statistical equivalence of the observed and simulated values. Table
10 indicates that the parameters which follow these constraints are hydrology, NO3-, DO, TSS,
DP, and total P. The other parameters either have p-values above 0.05 (Org-N, NH4+, BOD5 ),
slope coefficients distant from 1.0 (NH4+, organic N), or have low R2 values (NH4+, BOD5).
Another conclusion that may be drawn from the linear regression analysis, stems from the
95% confidence intervals (CI). In the expected operating range (real world values), CI
predictions for all nine parameters, overlap zero for B0 and one for B1. This indicates that SETWET has shown a good agreement between the predicted and observed values, since the range of
CI values include the idealized line of equivalent predictions. However, there are large amounts
of variance in the data sets for each parameter, indicated by the large range in CI predictions;
therefore, results need to be accepted with caution.
TABLE 10: LINEAR REGRESSION DATA FOR OBSERVED (Y-AXIS) AND PREDICTED (X-AXIS) WETLAND
EFFLUENT.
Hydrology
NH4+
B0
B1
R2
p-value
Lower 95%
C.I. B0
Upper 95%
C.I. B0
Lower 95%
C. I. B1
Upper 95%
C. I. B1
9.820
9.45
1.070
-0.710
0.945
0.105
0.001
0.480
-199.30
-2.05
218.93
20.94
0.77
-3.10
1.37
1.68
NO3-
0.03
1.300
0.881
0.002
-0.16
0.22
0.75
1.86
Org-N
-1.23
1.480
0.560
0.053
-5.81
3.36
-0.03
3.00
DO
BOD5
-2.72
1.69
0.933
0.650
0.574
0.126
0.048
0.434
-8.32
-22.96
2.88
26.35
0.01
-1.32
1.86
2.61
TSS
2.478
0.873
0.812
0.006
-2.77
7.73
0.39
1.36
DP
TP
0.535
0.693
0.843
0.869
0.956
0.887
0.001
0.002
-0.39
-0.90
1.46
2.28
0.63
0.51
1.05
1.23
134
Figure 42 depicts an example of the observed and predicted values for dissolved P plotted
against each other, along with the linear regression and the 1:1 line. As the figure depicts, the
prediction interval does contain the 1:1 line indicating that there is strong agreement between the
model predictions and observed data. Since the p-value (.001) is also low, the R2 value is high
(0.956), and the C.I.s cover both B0=0 and B1=1, there is strong agreement between the observed
and predicted values. The error bars represent the 95% prediction interval band for the linearly
regressed line. Appendix E contains the plots for the hydrology, ammonium, nitrate, organic N,
dissolved oxygen, BOD5, TSS, and total P concentrations.
FIGURE 42: SIMULATED AND OBSERVED VALUES FOR DISSOLVED PHOSPHOROUS CONCENTRATIONS,
PLOTTED WITH THE DETERMINED LINEAR REGRESSION, AND IDEAL 1:1 LINE.
135
C.
Sensitivity Analysis:
A sensitivity analysis was conducted to determine which parameters would require the
most scrutiny in future simulations. The relative sensitivity (RS) of each parameter was
determined with the following equation (Heatwole, 1998):
O Ob
S R =
P Pb
Pb
*
Ob
(85)
where SR is the relative sensitivity; O is the model output variable of interest; P is the parameter
value; and b is a subscript which represents the parameter values and output of the base scenario.
Each examined parameter was adjusted a total of six times, with changes of (+/-) 10%,
(+/-) 25%, and (+/-) 50%. The cumulative average value for the base scenario was compared
with the average change in model response for the hydrologic and nutrient components. If the
absolute value of the unitless RS is equal to one, then the same percent change occurs in the
output and input; if the RS is less then one, the model response is damped; while an RS value
greater than one indicates the model response is inflated. If the RS value is equal to zero, then
the selected parameter has no effect on the outcome of the components predicted values.
Additionally, if the RS value is positive or negative, it can be determined if there is an indirect (-)
or direct (+) relationship between the parameter and the output (Heatwole, 1998).
Table 11 presents the relative sensitivity analysis results for changes of (+/-) 50% to the
base value. Appendix F contains the tables for the relative sensitivity changes of (+/-) 10% and
(+/-) 25%. The data is presented for the hydrologic and nutrient components of SET-WET. The
results of the (+) 50% change are presented first followed by the results of the () 50% change.
The (+/-) sign designates whether the RS value relation is direct or indirect, while the letters
represent the one of seven categories into which the RS value falls. A note is made of the NC
category, which indicates that the simulation run was incomplete due to the generation of
unintelligible results. Parameter values could not be changed (+) 50% for certain situations in
which the parameters were fractions. A (+) 50% change resulted in a fractional value greater
than one, which is impossible. For these cases (PEATCC, BODCFRAC, and PONCOUT), Table
11 presents RS values for a change of +25% from the base. For SET-WET, the results of the
136
TABLE 11: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-) 50% CHANGE IN BASE VALUES
P a ra m e te r
137
B A C T ER IA
AEM A X G RB
AEM A X G RW
ANM AX GRB
ANM AX GRW
H ET R O IB
H ET ER O IW
HNO 3HS CB
HNO 3HS CW
HORGHSCB
HORGHSCW
HTDOHSCB
HTDOHSCW
HTDRB
HTDRW
NDOHSATB
NDOHSATW
NDRA TEB
NDRA TEW
N IT R O S IB
N IT R O S IW
NM AXGRB
NM AXGRW
NNH4HS CB
NNH4HS CW
N IT R O G EN
B IO M C N
B IO M P N
D O N IN IT B
D O N IN IT W
H T N O 3YB
H T N O 3YW
IM M IN IT B
IM M IN IT W
M IC R O N C
M TCDON
M TCNH4
M TCNO3
N H 4 IN IT B
N H 4 IN IT W
N O 3 IN IT B
B a s e V a lu e
NH4
NO3
DON
0.01
0.05
0.01
0.05
240000
25000
0.15
0.15
50
50
0.15
0.5
0.00125
0.001
1
1
0.002
0.002
2400
1000
0.005
0.005
1
1
(N C )/ (+ E )
(+ E )/ (+ D )
(+ D )/ (+ D )
(+ B )/ (+ B )
(N C )/ (+ E )
(+ D )/ (+ D )
(-C )/ (-D )
(-B )/ (-B )
(-C )/ (-D )
(-C )/ (N C )
(-D )/ (N C )
(-C )/ (-C )
(-D )/ (-E )
(-C )/ (-C )
(-C )/ (-C )
(-B )/ (-B )
(-C )/ (-C )
(-B )/ (-B )
(+ D )/ (+ D )
(+ C )/ (+ C )
(-D )/ (N C )
(+ C )/ (+ C )
(-C )/ (-C )
(-B )/ (-B )
(N C )/ (-D )
(-C )/ (-C )
(-D )/ (-D )
(-C )/ (-C )
(N C )/ (-E )
(-C )/ (-C )
(+ D )/ (+ D )
(+ B )/ (+ B )
(+ C )/ (+ C )
(+ C )/ (N C )
(-C )/ (N C )
(-C )/ (-C )
(+ D )/ (+ D )
(+ B )/ (+ B )
(-C )/ (-C )
(-C )/ (-C )
(-C )/ (-C )
(-C )/ (-C )
(+ D )/ (+ D )
(+ D )/ (+ D )
(-F )/ (N C )
(+ D )/ (+ D )
(-C )/ (-C )
(-C )/ (-C )
(N C )/ (-C )
(-D )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(N C )/ (-C )
(-C )/ (-C )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ C )/ (N C )
(+ C )/ (N C )
(+ B )/ (+ B )
(+ B )/ (+ C )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(+ E )/ (N C )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
2 3 .5
95
10000
5000
3.29
3.29
9900000
700000
0.125
0.00002
0.00006
0.00006
55000
30000
4000
(+ D )/ (-D )
(+ E )/ (+ E )
(+ B )/ (+ C )
(-B )/ (+ C )
(+ C )/ (+ D )
(-B )/ (+ B )
(-B )/ (+ C )
(-B )/ (+ B )
(-C )/ (-D )
(-B )/ (+ B )
(-C )/ (-C )
(+ C )/ (+ C )
(+ E )/ (+ E )
(+ D )/ (+ D )
(+ C )/ (-D )
(+ F )/ (-B )
(+ D )/ (+ D )
(-B )/ (+ B )
(-B )/ (+ B )
(+ D )/ (+ D )
(+ C )/ (+ C )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (-B )
(+ C )/ (+ D )
(+ C )/ (+ C )
(+ B )/ (+ C )
(+ D )/ (-F )
(-C )/ (-C )
(-B )/ (-B )
(+ D )/ (+ D )
(+ D )/ (+ D )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(+ C )/ (+ C )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (+ C )
PON
DO
BOD5
TSS
DP
TP
(N C )/ (-C ) (N C )/ (-E )
(-C )/ (-C )
(-D )/ (-D )
(-B )/ (-B )
(-C )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(N C )/ (-C ) (N C )/ (-E )
(-B )/ (-B )
(-D )/ (-D )
(+ B )/ (+ B ) (+ B )/ (+ B )
(+ B )/ (+ B ) (+ B )/ (+ B )
(+ B )/ (+ B ) (+ C )/ (+ C )
(+ B )/ (N C ) (+ D )/ (N C )
(+ C )/ (N C ) (+ D )/ (N C )
(+ B )/ (+ B ) (+ C )/ (+ C )
(+ B )/ (+ C ) (+ D )/ (+ D )
(+ C )/ (+ C ) (+ C )/ (+ C )
(+ B )/ (+ B ) (+ C )/ (+ C )
(+ B )/ (+ B ) (+ C )/ (+ C )
(+ B )/ (+ B ) (+ C )/ (+ C )
(-B )/ (-B )
(+ C )/ (+ B )
(-C )/ (-B )
(-D )/ (-D )
(-B )/ (-B )
(-C )/ (-C )
(-D )/ (N C ) (+ D )/ (N C )
(-B )/ (-B )
(-D )/ (-D )
(+ B )/ (+ B ) (+ C )/ (+ (C )
(+ B )/ (+ B ) (+ C )/ (+ C )
(N C )/ (-C )
(-D )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(N C )/ (-C )
(-C )/ (-C )
(+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ C )/ (N C )
(+ C )/ (N C )
(+ B )/ (+ B )
(+ B )/ (+ C )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-E )/ (N C )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A / (-C )
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
(-C )/ (+ C )
(A / -C )
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
(-C )/ (+ C )
(-D )/ (-E )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ D )
(-C )/ (+ C )
(+ D )/ (+ D )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-C )/ + B )
(-C )/ (-D )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-C )/ (-C )
(-C )/ (-C )
(-B )/ (+ C )
TABLE 11 (CONT.) SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-)
50% CHANGE IN BASE VALUES
P a ra m e te r
B a s e V a lu e
NH4
NO3
PON
DO
BOD5
TSS
DP
TP
N S Y IE L D B
N S Y IE L D W
ONPARTF
PONCOUT
PONFALL
P O N IN IT B
P O N IN IT W
P O NR ES
P O N S IZE
R E F N IN IT
R E S T H N (N I T )
V EG ET A T IO N
B IO D EN S
B IO IN IT
P B IO U W
P EA T ACRB
P EA T ACRW
P EA T DENS
P RAT EUP
PSTDUW
S T A N D IN
S T D DENS
CARBON
B IO C C O N T
BODCFRAC *
BODP FRAC
D O C IN IT B
D O C IN IT W
LEACHR
M IC R O B EC
M TCDOC
P EA T CC *
POCCOUT *
POCFALL
P O C IN IT B
P O C IN IT W
P O CR ES
P O C S IZE
R E F C IN IT
0.3
0.3
0.6
0.75
0.25
300000
15000
0.01
0.05
500000
0.01
(-B )/ (+ C )
(-C )/ (+ C )
(+ D )/ (+ D )
(-C )/ (-C )
(+ C )/ (+ C )
(+ E )/ (-D )
(+ D )/ (-D )
(-B )/ (+ B )
(-B )/ (+ B )
(+ D )/ (-D )
(-B )/ (+ B )
(-D )/ (-D )
(-D )/ (-D )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(+ F )/ (-F )
(+ F )/ (-F )
(-B )/ (+ B )
(-B )/ (+ B )
(+ F )/ (-F )
(-B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-F )/ (-F )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
DON
(-C )/ (+ C )
(-C )/ (+ C )
(+ E )/ (+ E )
(+ E )/ (+ E )
(-E )/ (-F )
(+ C )/ (+ D )
(+ D )/ (+ D )
(+ D )/ (+ D )
(-C )/ (+ C )
(-C )/ (+ C )
(+ D )/ (+ D )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-C )/ (+ C )
(-C )/ (+ C )
(-B )/ (+ B )
(-B )/ (+ B )
(-C )/ (+ C )
(-B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
A /A
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B / (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
A /A
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B / (+ B )
(-B )/ (+ B )
5000
8635000
0.4
2000
300
0.7
0.7
0.4
6000000
5000
(+ B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ C )/ (+ C )
(+ B )/ (+ B )
(-B )/ (-B )
(-D )/ (+ B )
(+ B )/ (+ B )
(-D )/ (-D )
(+ B )/ (+ B )
A / (-B )
(-B )/ (-B )
(+ B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(+ E )/ (+ D )
(+ B )/ (+ B )
(-C )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(+ D )/ (+ D )
(+ B )/ (+ B )
(-B )/ (-B )
(-C )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ D )/ (+ D )
(-B )/ (-B )
(-B )/ (-B )
(-C )/ (-C )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-C )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(+ D )/ (+ D )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ D )/ (+ D )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
A /A
(+ B )/ (+ C )
A /A
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
A /A
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
A /A
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
0.47
0.8
0.5
60000
50000
0.01
0.53
0.00004
0.8
0.3
0.45
3000000
200000
0.001
0.2
15000000
(-D )/ (-D )
(-C )/ (-C )
(-D )/ (-D )
(-B )/ (+ B )
(-B )/ (+ C )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-C )/ (+ B )
(+ B )/ (+ C )
(-C )/ (-C )
(-D )/ (-D )
(-C )/ (-C )
(-C )/ (+ B )
(-B )/ (+ B )
(+ B )/ (+ B )
(-C )/ (-C )
(-C )/ (-C )
(+ C )/ (+ C )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ C )
(-C )/ (-C )
(-B )/ (-B )
(+ B )/ (+ B )
(-B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ B )
(+ C )/ (+ C )
(-B )/ (+ B )
(-B )/ (-B )
(-B )/ (+ B )
(-B )/ (-B )
(+ B )/ (-B )
(+ B )/ (+ C )
(+ B )/ (+ B )
(+ B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(+ D )/ (+ E )
(-D )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-D )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ D )
(-C )/ (+ C )
(-C )/ (+ C )
(-C )/ (+ C )
(+ C )/ (+ C )
(-C )/ (-C )
(-D )/ (-D )
(+ D )/ (+ C )
(-C )/ (-C )
(-C )/ (-C )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(-B )/ (-B )
(+ B )/ (+ B )
(+ B )/ (+ D )
(-C )/ (-C )
(-C )/ (-C )
(+ B )/ (-B )
(-B )/ (-B )
(-B )/ (+ B )
(+ D )/ (+ D )
(-E )/ (-E )
(-E )/ (-E )
(+ D )/ (+ D )
(+ D )/ (+ D )
(+ D )/ (+ C )
(+ B )/ (+ B )
(+ C )/ (+ C )
(+ B )/ (+ B )
(+ E )/ (+ E )
(-E )/ (-E )
(+ B )/ (+ C )
(+ C )/ (+ C )
(-E )/ (-B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
(-B )/ (+ B )
138
TABLE 11 (CONT.) SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-)
50% CHANGE IN BASE VALUES
P a ra m e te r
B a s e V a lu e
D IS S O L V ED O X YG EN
0 .0 0 1
DOCONCP
15000
D O IN IT B
15000
D O IN IT W
8 .5
DO X YCS AT
0 .1 5
HTDO YB
0 .1 5
HTDO YW
0 .0 0 0 1
M TDOX
0 .0 0 0 0 8
M TFW SDOC
0 .2
NS DO YB
0 .2
NS DO YW
PHOSPHOROUS
300
B IO M P P
250000
BTPHOSI
40000
D T P H O S IB
38000
D T P H O S IW
0 .0 0 0 0 6
M TCPHOS
0 .0 5
P M IN P P C
0 .0 0 0 0 5
P R M IN B P C
0 .5
P S EDD EP
S E D IM E N T
0 .1
DECO M P R
2 .5
M A N N C (S E D )
0 .5
P S EDD EP
0 .0 0 1 1
R E S T H I C K (S E D )
0 .3
S EDFALL1
0 .7
S EDFALL2
26000000
S E D IN IT B 1
10000000
S E D IN IT B 2
190000
S E D IN IT W 1
10000
S E D IN IT W 2
0 .1
S EDRES
0 .2 5
S E D S IZE 1
0 .5
S E D S IZE 2
1 .1
S EDS P G 1
2 .6 5
S EDS P G 2
NH4
NO3
( - B ) / (+ B )
( - D ) / (- D )
( - C ) / (- C )
( - E ) / (N C )
( - E ) / (N C )
( - C ) / (N C )
( - D ) / (N C )
( - D ) / (N C )
( - C ) / (- E )
( - C ) / (- D )
( - B ) / (+ B )
(+ C )/ (+ C )
(+ C )/ (+ C )
(+ D )/ (N C )
(+ E )/ (N C )
(+ C )/ (N C )
(+ D )/ (N C )
(+ D )/ (N C )
(+ C )/ (+ D )
(+ B )/ (+ C )
( - B ) / (+ B )
(+ C )/ (+ C )
(+ B )/ (+ C )
( - B ) / (+ C )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
)
)
)
)
)
)
)
)
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - B ) / (+
( - C ) / (+
C)
D)
C)
C)
C)
C)
B)
C)
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (- B )
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
)
)
)
)
)
)
)
)
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - E ) / (- F )
(+ C )/ (+ D )
(+ C )/ (+ C )
(+ E )/ (+ E )
(+ E )/ (+ E )
(+ C )/ (+ C )
(+ C )/ (+ C )
(+ C )/ (+ C )
( - C ) / (- C )
(+ C )/ (+ D )
(+ C )/ (+ C )
(+ E )/ (+ E )
(+ E )/ (+ E )
(+ C )/ (+ C )
(+ C )/ (+ C )
(+ C )/ (+ C )
( - C ) ( -C )
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (+ B )
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
( - C ) / (+
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
C)
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (- B )
(+ B )/ (+ B )
( - B ) / (( + B )
( - B ) / (( + B )
( - B ) / (+ B )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (( + B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (+ B )
( - B ) / (+ B )
( - B ) / (+ E )
( - E ) / (- F )
(+ E )/ (+ E )
( - E ) / (- F )
( - B ) / (- F )
(+ E )/ (+ E )
(+ B )/ (+ B )
(+ E )/ (+ E )
( - B ) / (+ B )
(+ E )/ (+ E )
( - B ) / (+ E )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (- B )
(-B ))/ (+ C )
( - C ) / (- C )
(+ C )/ (+ C )
( - C ) / (- C )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (- B )
(+ C )/ (+ C )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (- B )
( - B ) / (+ C )
( - C ) / (- C )
(+ C )/ (+ C )
( - C ) / (- C )
( - B ) / (- B )
( - B ) / (+ B )
( - B ) / (- B )
(+ B )/ (+ B )
( - B ) / (- B )
(+ C )/ (+ C )
( - B ) / (+ B )
(+ B )/ (+ B )
( - B ) / (- B )
(+ B )/ (+ B )
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
DON
(-B / (+ B )
(+ C )/ (+ C )
(+ B )/ (+ C )
(+ C )/ (N C )
(+ C )/ (N C )
(+ B )/ (N C )
(+ C )/ (N C )
(+ C )/ (N C )
(+ B )/ (+ C )
(+ B )/ (+ B )
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
B
B
B
B
B
B
B
B
PO N
DO
BOD5
( - C ) / (+ C )
( - C ) / (+ D )
( - C ) / (+ D )
( - C ) / (N C )
( - C ) / (N C )
( - C ) / (N C )
( - C ) / (N C )
( - C ) / (N C )
( - C ) / (+ D )
( - C ) / (+ C )
(+ B )/ (+ B )
(+ D )/ (+ D )
(+ D )/ (+ D )
(+ E )/ (N C )
(+ E )/ (N C )
(+ D )/ (N C )
( - C ) / (N C )
( - D ) / (N C )
(+ D )/ (+ D )
(+ C )/ (+ D )
( - B ) / (+ B )
(+ C )/ (+ C )
(+ C )/ (+ C )
(+ C )/ (N C )
(+ C )/ (N C )
(+ B )/ (N C )
(+ B )/ (N C )
(+ C )/ (N C )
(+ B )/ (+ C )
(+ B )/ (+ B )
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
B
B
B
B
B
B
B
B
TSS
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
139
- Results presented as the RS values for the +50% change in base value followed by the 50% change.
- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.
- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible results)
- * RS values presented for +25% due to fraction value greater than one.
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
B
B
B
B
B
B
B
B
B
B
DP
)
)
)
)
)
)
)
)
)
)
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
TP
B
B
B
B
B
B
B
B
B
B
)
)
)
)
)
)
)
)
)
)
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
(-B
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
) / (+
B
B
B
B
B
B
B
B
B
B
)
)
)
)
)
)
)
)
)
)
sensitivity analysis indicated that most of the RS values approached zero and thus model
response is extremely damped, implying that the effect of most parameters on model predictions
will be minimal.
Generally, the NCOB cycle is insensitive to changes to a single parameter due to the
complexity and number of interactions in the wetland system; however, there are exceptions.
The NCOB cycle is most sensitive to changes that affect microbial growth and oxygen; which is
exemplified by the situations in which the model was unable to complete the simulations
(HTDOYB, AEMAXGRB, HETROIB, MTDOX, NMAXGRB, HORGHSCW, MTDOX,
MTFWSDOC, HTDOHSCB, and HTDOYW). Since SET-WET is driven by the interactions
and processes of the bacteria cycle, it makes sense that the model is most sensitive to the
parameters affecting the growth and number of bacteria in the system. The faster the growth of
bacteria in the system, the quicker organics degradation, ammonification, nitrification and
denitrification occurs. Changing bacterial parameter values can considerably change the
bacterial growth, rates of nutrient uptake, and transformation rates determined by SET-WET in
the wetland system. As the changes accumulate, the modeling system may be unable to properly
account for the values, producing incomprehensible results, which include any negative values
for a stock variable. Since, the DO concentration interacts greatly with the bacteria cycle inSETWET, it is not surprising that the model is sensitive to changes in the DO parameter values.
Bacterial growth is affected by the DO concentrations because various rates in the bacteria cycle
that determine bacterial nutrient utilization, the determination of the anaerobic fraction
(ANFRAC), and growth, are a function of DO. The bacteria in turn consume the oxygen,
forming a carefully balanced cycle.
Trends in which parameters were the most significant did materialize. The BOD5
simulated values were sensitive to changes involving the C cycle parameters, while the NH4+ and
NO3- output were sensitive to changes associated with bacteria and NO3- parameters. Any
attempt to vary the fall rate (SEDFALL) or resuspension percentage of particles, significantly
affected particle outflow. These results are expected as changes in these parameters directly
affect the respective cycles. In addition, parameters associated with the bottom of the wetland
system have a greater effect on predictions, which is attributable to the more prevalent number of
bacteria in the wetland bottom. Since the NCOB, and sediment and P cycles were designed to
run independently, changes in associated parameter values had little effect upon the predictions
140
of the other independent cycles. The model is generally insensitive to changes affecting
sediment or P parameters.
D.
Modeling Application
SET-WET was developed to assist with the design of constructed wetlands for optimizing
the control of NPS pollution. To examine the potential use of SET-WET, it was applied to a
hypothetical situation where a constructed wetland might be desired. The objective of this
analysis was to observe the potential use of SET-WET for long-term simulations, to analyze the
design capabilities of SET-WET, and determine its strengths and weaknesses. The wetland
designs emphasized the capture of NPS pollutants from a watershed area, with the goal of
optimizing pollutant removal. This is a simplification of the wetland design process because
many factors must be accounted for. Besides the need to control pollution; ecological,
economics, and landowner concerns must also be addressed. The designer must determine the
proper balance between these factors.
1.
Study/Application Area
Data that was collected by Mostaghimi et al. (1998) in the Nomini Creek watershed was
used as input values for the simulation. This site was selected because long term field data that
contained most of the required input to the SET-WET model was collected from the watershed.
Any other data set that contains the hydrologic and nutrient values from a watershed area could
be used for simulation with the SET-WET model.
The Nomini Creek watershed is located in Westmoreland County, Virginia, and is 80-km
northeast of Richmond (Figure 43). In its entirety, the watershed is 1463 hectares large, but for
the purpose of this analysis a subwatershed, entitled QN2, was used (Figure 44). QN2 is 214
hectares in size. The watershed has typical Coastal Plain land use characteristics as 49 % is used
for cropland, 47% woodland, and 4% for homesteads and roads, with no significant point sources
in the area (Mostaghimi et. al, 1998).
The climate in the area is temperate and humid. Average annual precipitation is 101.6
cm, with a large percentage of the rainfall occurring between April and September; while the
141
FIGURE 43: LOCATION OF THE NOMINI CREEK WATERSHED IN VIRGINIA WITH RESPECT TO RICHMOND,
VA AND THE CHESAPEAKE BAY.
FIGURE 44: NOMINI CREEK WATERSHED (QN1) WITH SUBWATERSHED (QN2; SHADED)
average annual snowfall is about 10 cm. The average summer and winter temperatures are 25 C
and 3 C, respectively (Mostaghimi, et al., 1998).
Data collection in the watershed occurred from 1986 to 1997, and various hydrologic
(streamflow, precipitation), nutrient (ammonium, nitrate, DON, PON, COD dissolved ortho P,
dissolved organic P, Particulate P, TSS) and climactic (air temperature) parameters were
142
collected. Specific BMPs were implemented in the watershed including: strip cropping;
vegetative filter strips; no-till cropping systems; critical area stabilization; drop structures;
diversion structures; nutrient management; and IPM (Mostaghimi et al., 1998). The use of
BMPs should theoretically decrease the amount of nutrients that leave the watershed outlet,
thereby leaving smaller amounts of pollutants for the hypothetical wetland to treat.
Data for a two-year period (March 26, 1992 to March 25, 1994) were used for these
simulations. This time period was chosen because data collection during this period was
complete, with very few missing data points. Data collection of the hydrologic and nutrient
parameters occurred on a weekly and storm-event basis, excluding the water temperature and
influent BOD5. These values were extended to daily input points by linear interpolation between
the weekly and storm values. Water temperature was estimated from the average daily
temperature in the system by subtracting 2 C from the average daily air temperature, while BOD
influent was estimated from the periodically taken COD measurements. The determination of
the daily BOD influent required an estimate drawn from the COD measurements in the
watershed. It was assumed that the ultimate BOD/COD ratio approached one (Perrich, 1981).
Fifty-five measurements of COD were taken over the eleven-year research period. These fiftyfive measurements were plotted with the recorded influent flow as shown in Figure 45. A linear
regression was calculated from these points which allowed a daily estimation of BOD5
(dependent variable; y) to be made based upon the hydrologic influent measurements
(independent variable x) for the simulation period. Although the R2 value of the regression line
was not that high, it was the best estimate for the site. Data for DO was not collected, therefore
it was assumed that the concentration of DO entering the site was 5.7 mg/L.
2.
Simulation Runs
Simulation runs were divided into eight season periods with each season period
corresponding to the seasonal changes in the area. The growing season was designated as the
time period between the average first and last freeze of the year. For the Nomini Creek
watershed, data from Richmond and Norfolk, Virginia (Wood, 1996) were averaged to determine
the growing season. There is a mean freeze free period of 237 days in the area meaning 128 days
could be designated as winter. The winter time period was from Nov. 17th to March 25th.
143
FIGURE 45: LINEAR REGRESSION OF RECORDED TOTAL BOD5 AND HYDROLOGIC INFLOW TO QN2
SUBWATERSHED OF NOMINI CREEK WATERSHED FOR MARCH 26, 1992 TO MARCH 25, 1994.
Spring, occurred from March 26th to June 21st, summer, from June 22nd to September 21st, and
fall, from September 22nd to November 16th. The Thornthwaite method was used to determine
ET, with average monthly temperature values being estimated from data collected at Chatham,
Virginia (Ruffner, 1980).
It was assumed that the wetland would be placed at the outlet of the QN2 subwatershed.
It was placed at the outlet since this would allow the highest capture rate of pollutants from the
area, as all surface flows must leave from this area. Since the watershed outlet consists of a
perennial stream, and groundwater flow is prevalent in the area, it is assumed that the wetland is
lined. Water may flow from the watershed to the wetland, but due to the lining, direct
groundwater interactions with the wetland water cycle is assumed negligible. It is assumed that
there are no point source additions to the wetland and that it is fully established for the
simulation periods, where full plant and bacteria colonies exist.
144
The first step in determining the design of a wetland is to size the wetland for proper
hydrologic outflow and properties, because improper hydrologic design will not allow a
wetlands vegetation to establish, and the system will become a detention basin rather than a
wetland. The Water Pollution Control Federation (1990; Table 3) suggests that the desired
minimum hydraulic residence time in the wetland system should be 5 days with a maximum
water depth of 50 cm. The residence time is suggested to allow the various nutrient processes
enough time to complete their respective reactions, while the maximum depth is suggested for
human safety reasons. The average hydraulic flow into the system is 2700 m3/day, which would
require that the wetland area be at least 27,000 m2 (2700m3/day*5days/.5m) or 2.7 hectares. The
second step in the wetland design would be to decide on the configuration of the wetland. Since
the Water Pollution Control Federation suggests a desired length to width ratio of at least 2 to 1,
the wetland systems length was input as 270 meters with an input width of 100 meters.
Technically, since the wetland is modeled as a continuously stirred tank reactor (CSTR), any
length to width ratio (1:1, 1:2, 5:1) whose product was a total wetland area of 2.7 hectares, would
have resulted in the same output values. This is because the SET-WET model accounts for the
volume and mass relations, and not space relations.
Once the proper hydrologic design was designated, SET-WET was used to model the
NCOB, sediment and P cycles. It was assumed that the input parameters were similar to those at
the Benton wetland site, as no other information was available for comparison. Similar input
values were necessary to complete model simulations; however, since the model has not been
calibrated with proper NPS pollution data, it is not certain if the parameter values are correct.
Table 12 lists the initial input parameters to the SET-WET model for the simulation run. This
simulation is referred to as the 2.7 hectare, wetland from this point on. Occasionally,
parameter values were changed between season periods through the simulations to better
represent the processes occurring in the wetland. These changes were made to the more sensitive
parameters, mainly those that affect microbial growth rates, and oxygen transfer.
Another suggestion by the Water Pollution Control Federation, for the design of a FWS
constructed wetland is to have the wetland be at least 2 hectares in size for every 1000m3 of
inflow per day. This would require the wetland design to be at least 5.4 hectares in size (2700
m3/day * 2 ha/1000 m3/day). The wetland for this shape was designated as 540 meters in length
145
with a width of 100 meters. Once again this design is to the discretion of the model user. This
simulation is referred to as the 5.4 hectare, wetland in Table 12.
The objective of the model application was to determine the proper design of a wetland to
optimize wetland NPS pollution control. Other considerations include cost and the size of the
wetland. A smaller wetland would be less expensive to build and would obviously take up less
area. For this reason a smaller area wetland with a higher water depth design was also examined.
This wetland was designated as the Smaller, site in Table 12.
Another design parameter examined was the amount of biomass in the wetland system.
Plants absorb nutrients in the wetland system; therefore if the plant biomass in the system is
doubled, theoretically more nutrients will be assimilated. However, when decomposition of the
plant material occurs there will also be twice as much to decompose. This simulation will allow
plant growth to be examined and is referred to as Plants, in Table 12.
Substrate depth also plays an important role in wetland processes. In the wetland
substrate, many of the transformation processes for the various nutrient cycles occurs. If the
amount of substrate in the bottom does not have a large affect on the total retention of nutrients
in the system, less humic soil would need to be transported to a potential wetland site. The
smaller wetland substrate would also increase the concentration of nutrients in this compartment.
This condition was examined and is referred to as Substrate, in Table 12.
In this analysis, the biomass growth rate was the same for all simulation runs, except for
the "Plant" analysis where the value was doubled. Due to the difference in wetland water depth,
it is unlikely that the same species of vegetation would be able to survive in the differing wetland
constructions, but for these simulations, they were considered to be the same.
3.
Simulation Results
Table 13 lists the results of the model simulations for the five separate events. Listed are
the influent total, effluent total, and reduction efficiency for the NH4+, NO3-, DON, PON, TKN,
TN, TSS, DP and TP for the respective wetland designs for simulation runs of two years. The
influent and effluent totals are based on the entire two-year time frame, with the reduction
efficiency based on the retention of nutrients over this period. In addition, Table 13 lists the
146
TABLE 12: INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL SIMULATION RUNS
FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF NOMINI CREEK
WATERSHED.
P a ra m e te r
BAS E
LE NG TH
W IDTH
HO
HB
HTI
HII
SO
HYD RO L O G Y
P O RP E A T
O UTLE T
A NG V NO T
DIS E F F C
K HCO E F
HO UT
HO V E R
BIO M AS S
B IO INIT
S TA NDIN
P E A TA CRW
P E A TA CRB
P E A TDE NS
P RA TE UP
B IO DE NS
S TDDE NS
P B IO UW
P S TDUW
DE G B IO
DA Y S DE G
(M A XIM UM ) B IO M G RR
BACT ERIA
NITRO S IW
NITRO S IB
HE TE RO IW
HE TE RO IB
NDRA TE W /NDRA TE B
NDO HS A TW /ND O HS A TB
NM A XG RW /NM A XG RB
NNH4HS CW /NNH4HS CB
A E M A XG RW /A E M A XG RB
A NM A XG RW /A NM A XG RB
HTDRW /H TDRB
HTDO HS CW /HTDO HS CB
2.7
He cta re
5.4
He cta re
S m a lle r
P la n ts
S u b stra te
270
100
2.2
0
1
1.4
0.001
540
100
2.2
0
1
1.4
0.001
200
55
2.3
0
1
1.7
0.001
200
55
2.3
0
1
1.7
0.001
200
55
2.3
0.5
1
1.7
0.001
0.32
2
60
0.175
0.0012
1.25
2
0.32
2
60
0.175
0.0012
1.25
2
0.32
2
60
0.175
0.0012
1.6
2
0.32
2
60
0.175
0.0012
1.6
2
0.32
2
60
0.175
0.0012
1.6
2
20000
35000000
300
2000
110
0.3
75
75
0.4
0.4
0.99
20
5
10000
25000000
300
2000
110
0.3
75
75
0.4
0.4
0.99
20
5
4000
15000000
300
2000
110
0.6
75
75
0.4
0.4
0.99
20
5
8000
20000000
300
2000
110
0.6
75
75
0.4
0.4
0.99
20
10
4000
15000000
300
2000
110
0.6
75
75
0.4
0.4
0.99
20
5
10000
50000
15000
1000000
.001/.001
1./1.
.008/.008
1./1.
.08/.01
.2/.04
.0035/.0035
.5/.05
5000
25000
7500
500000
.001/.001
1./1.
.008/.008
1./1.
.08/.01
.2/.04
.0035/.0035
0.5/0.5
2000
10000
3000
200000
.001/.001
1./1.
.008/.008
1./1.
.08/.01
.2/.04
.0035/.0035
.5/.5
2000
10000
3000
200000
.001/.001
1./1.
.008/.008
1./1.
.08/.01
.2/.04
.0035/.0035
.5/.5
2000
10000
3000
100000
.001/.001
1./1.
.008/.008
1./1.
.1/.03
.2/.04
.0035/.0035
.5/.5
147
TABLE 12 (CONT.): INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL
SIMULATION RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF
NOMINI CREEK WATERSHED.
P a ra m e te r
HNO3HS CW /HNO3HS CB
HORGHS CW /HORGHS CB
A NFRA CW /A NFRA CB
HTTE M P FW /HTTE M P FB
2.7
He cta re
.15/.15
50/50
.2-.8
0.0-1.0
5.4
He cta re
0.15/0.15
50/50
.2-.8
0.0-1.0
S m a lle r
.15/.15
50/50
.2-.8
0.0-1.0
P la nts
.15/.15
50/50
.2-.8
0.0-1.0
S ubstra te
.15/.15
50/50
.2-.8
0.0-1.0
CARBON
RE FCINIT
DOCINITW
DOCINITB
P OCINITW
P OCINITB
B IOCCONT
B ODCFRA C
B ODP FRA C
LE A CHR
M ICROB E C
P E A TCC
P OCFA LL
P OCRE S
M A NNC
RE S THC
P OCS IZE
M TCDOC
P OCCOUT
1400000
250000
50000
200000
1000000
0.47
0.8
0.5
0.1
0.53
0.8
0.6
0.001
2
0.005
0.2
0.00005
0.2
700000
125000
25000
100000
500000
0.47
0.8
0.5
0.1
0.53
0.8
0.6
0.001
2
0.005
0.2
0.00005
0.2
340000
50000
10500
40000
2000000
0.47
0.8
0.5
0.1
0.53
0.8
0.6
0.001
2
0.005
0.2
0.00005
0.2
340000
50000
10500
40000
2000000
0.47
0.8
0.5
0.1
0.53
0.8
0.6
0.001
2
0.005
0.2
0.00005
0.2
340000
50000
10500
40000
2000000
0.47
0.8
0.5
0.1
0.53
0.8
0.6
0.001
2
0.005
0.2
0.00005
0.2
NITROGEN
DONINITW
DONINITB
IM M INITW
IM M INITB
NH4INITW
NH4INITB
NO3INITW
NO3INITB
P ONINITW
P ONINITB
RE FNINIT
B IOM CN
B IOM P N
HTNO3Y W /HTNO3Y B
M ICRONC
NS Y IE LDW /NS Y IE LDB
ONP A RTF
P E A TNC
P ONRE S
P ONFA LL
M TCDON
15000
25000
50000
25000000
25000
15000
15000
25000
75000
1500000
50000
23.5
95
.2/1
0.125
.2/.15
0.6
0.025
0.01
0.5
0.00004
7500
12500
25000
12500000
1250
7500
7500
12500
37500
750000
250000
23.5
95
.2/1
0.125
.2/.15
0.6
0.025
0.01
0.5
0.00004
3000
5000
10000
5000000
500
3000
3000
5000
15000
300000
105000
23.5
95
.2/1
0.125
.2/.15
0.6
0.025
0.01
0.5
0.00004
3000
5000
10000
5000000
500
3000
3000
5000
15000
3000000
105000
25
95
.2/1
0.0125
.2/.2
0.6
0.025
0.01
0.5
0.00004
3000
5000
10000
5000000
500
3000
3000
8000
15000
3000000
105000
23.5
95
.5/1.0
0.125
.3/.3
0.6
0.025
0.01
0.5
0.00004
148
TABLE 12 (CONT.): INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL
SIMULATION RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF
NOMINI CREEK WATERSHED.
Parameter
PONSIZE
RESTHN
PONCOUT
2.7
Hectare
0.05
0.01
0.375
5.4
Hectare
0.05
0.01
0.375
Smaller
0.05
0.01
0.375
Plants
0.05
0.01
0.375
Substrate
0.05
0.01
0.375
DISSOLVED OXYGEN
DOINITW
DOINITB
HTDOYW/HTDOYB
NSDOYW/NSDOYB
DOCONCP
MTDOX
MTFWSDOC
DOXYCSAT
BIOOXRB
DOCONCIN
140000
20000
.1/.15
.02/.05
0.001
0.0003
0.00008
8.5
0/.18
5.7
70000
10000
.1/.15
.02/.05
0.001
0.0003
0.00008
8.5
0-.18
5.7
28000
4000
.5/.5
.08/.08
0.001
0.0003
0.00008
8.5
0-.18
5.7
28000
4000
.1/.15
.02/.05
0.001
0.0003
0.00008
8.5
.0-.18
5.7
28000
4000
.4/.4
.05/05
0.001
0.0006
0.00008
8.5
0-.18
5.7
SEDIMENT
SEDCAT
SEDRES
SEDSIZE (SEDCLASS)
SEDFALL (SEDCLASS)
SEDINITW (SEDCLASS)
SEDINITB (SEDCLASS)
SEDSPG (SEDCLASS)
SEDPER (SEDCLASS)
RESTHICK
MANNC
DECOMPR
PSEDDEP
PHOSPHOROUS
DTPHOSIW
DTPHOSIB
BTPHOSI
PPHOSI
PMINPPC
PRMINBPC
ADSORP
MTCPHOS
BIOMPP
PHOSCON
3
3
3
3
0.001
0.001
0.001
0.001
.074-.2
.074/.2
.074-.2
.074-.2
.7/1.2
.7/1.2
.7-1.2
.7-1.2
5000-120000 2500-60000 1000-24000 1000-24000
250000012500005000005000005940000000 2970000000 1210000000 1210000000
2.65
2.65
2.65
2.65
.8/.2
.8/.2
.8/.2
.8/.2
0.0001
0.0001
0.0001
0.0001
0.5
0.5
0.5
0.5
0.1
0.1
0.1
0.1
1
1
1
1
10000
50000
5400000
75000
0.005
0.00005
2
0.00006
300
0
5000
25000
2700000
37500
0.005
0.00005
2
0.00006
300
0
2000
10000
900000
15000
0.005
0.00005
2
0.00006
300
0
2000
10000
900000
15000
0.005
0.00005
2
0.00006
300
0
3
0.001
.074-.2
.7-1.2
1000-24000
500000605000000
2.65
.8/.2
0.0001
0.5
0.1
1
2000
10000
900000
15000
0.005
0.00005
2
0.00006
300
0
149
TABLE 13: INFLUENT, EFFLUENT, AND % REDUCTION OF NUTRIENTS FOR VARIOUS NUTRIENTS FOR 2-YEAR PERIODS OF WETLAND SIMULATIONS
FPOR QN2 SUBWATERSHED DATA.
2.7 hectare
Total In
Total Out
% Reduction
5.4 hectare
Total In
Total Out
% Reduction
Smaller !
Total In
Total Out
% Reduction
Plant *
Total In
Total Out
% Reduction
Substrate #
Total In
Total Out
% Reduction
150
!
*
#
NH4
NO3
DON
PON
TKN
TN
BOD5
TSS
DP
TP
1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06
2.31E+05 3.07E+06 2.00E+06 2.16E+05 2.44E+06 5.52E+06 6.01E+07 1.14E+06 7.88E+05 7.88E+05
-65.79% 18.48% -9.38% 95.97% 66.66% 50.29% 44.62% 99.78% 16.43% 46.33%
NH4
NO3
DON
PON
TKN
TN
BOD5
TSS
DP
TP
1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06
1.96E+05 2.46E+06 2.06E+06 1.22E+05 2.38E+06 4.84E+06 5.91E+07 5.91E+05 6.39E+05 6.39E+05
-41.08% 34.73% -12.95% 97.72% 67.52% 56.38% 45.50% 99.88% 32.24% 56.48%
NH4
NO3
DON
PON
TKN
TN
BOD5
TSS
DP
TP
1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06
6.10E+05 3.38E+06 1.95E+06 5.09E+05 3.07E+06 6.45E+06 6.40E+07 7.61E+07 9.00E+05 9.11E+05
-338.15% 10.32% -6.81% 90.51% 58.14% 41.89% 40.98% 85.09% 4.54% 37.94%
NH4
NO3
DON
PON
TKN
TN
BOD5
TSS
DP
TP
1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06
5.36E+05 3.20E+06 1.95E+06 5.28E+05 3.01E+06 6.21E+06 6.51E+07 7.90E+07 8.39E+05 8.49E+05
-285.62% 15.20% -6.70% 90.16% 58.90% 44.05% 39.94% 84.52% 11.03% 42.12%
NH4
NO3
DON
PON
TKN
TN
BOD5
TSS
DP
TP
1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06
3.05E+05 3.44E+06 1.95E+06 5.63E+05 2.82E+06 6.26E+06 6.42E+07 7.61E+07 9.03E+05 9.14E+05
-119.51% 8.78%
-7.05% 89.49% 61.48% 43.58% 40.83% 85.08% 4.21% 37.73%
HRT
(days)
HLR
(cm/d)
Water
Depth (cm)
4.96
10.92
0.52
HRT
(days)
HLR
(cm/d)
Water
Depth (cm)
9.84
5.49
0.53
HRT
(days)
HLR
(cm/d)
Water
Depth (cm)
3.46
26.72
0.87
HRT
(days)
HLR
(cm/d)
Water
Depth (cm)
3.40
26.72
0.87
HRT
(days)
HLR
(cm/d)
Water
Depth (cm)
3.46
26.72
0.87
average hydraulic residence times (HRT), hydraulic loading rate (HLR), and water depth for the
two-year simulations.
For all simulations, the NH4+ concentration (-41.1% to 338.2%) considerable increased
after traveling through the wetland. This is not an unreasonable result because the loading rates
of NH4+ to the system were minute when compared to the loading rates of the organic N and
NO3- to the system. The wetland acts as a source for NH4+ because a relatively large amount of
organic N goes through the mineralization process, especially during the decomposition stage of
winter, and the NH4+ does not have enough time to be converted to NO3- before leaving the
system.
The largest wetland (5.4 hectares) retained more nutrients than the other wetland designs
because in the "5.4 hectare" wetland there are more bacteria to carry out processes, more area for
transformation processes to occur, and a longer time frame for these processes to occur.
Retention of TSS in the "5.4 hectare" and "2.7 hectare" sites were nearly 100%, as compared to
the 85% retention for the other three designs, which is attributable to the smaller average water
depth in the systems (.52 m to .87 m, respectively). The system with more vegetation (Plants)
retained more nutrients when compared to the "Smaller" wetland; however, the extra amount
retained seem negligible (3% to 5% greater for most parameters) as only a slightly higher
percentage is retained by the wetland system. Another interesting note is that the thickness of
the wetland substrate seemed negligible to the amount of nutrients retained by the wetland
system. In fact, the retention in "Substrate" was greater for NH4+, TKN, and TN. Thus, it seems
that depending on the location of the wetland material, the designer has flexibility with the
amount of substrate that needs to be transported to or from a construction location.
All of the designed wetlands reduced the percentage of BOD5 (39.9%-45.5%), TSS
(84.5%-99.9%), TN (41.9%-56.4%), and total P (37.7%-56.5%) to levels reported by previous
research (Table 14). Depending on the retention requirements of the wetland, an optimal
wetland can be designed for the area by using the information gained from the five simulations.
If an increase in TN retention was needed, it can be seen from Table 13 that the retention
increased with an increase in wetland size, plant amounts, and a decrease in substrate thickness.
These design parameters can be modified until a desirable retention is attained. Conversely, if
the goal is to retain a high percentage of TSS, it is apparent that a larger wetland with a smaller
water depth is more effective for sediment retention.
151
TABLE 14: RANGE OF POLLUTANT REMOVAL EFFICIENCIES REPORTED FOR CONSTRUCTED WETLAND
SYSTEMS
P a r a m e te r
R e m o v a l (% )
BOD5
5 0 -9 0
SS
4 0 -9 4
N it ro g e n
3 0 -9 8
P h o s p h o ro u s
2 0 -9 0
An optimal design is constrained not only by the nutrient retention of the wetland but also
by other constraints, such as area available to construct the wetland and the cost of the system.
For example, it is useless to design a wetland for 5 hectares which costs $50,000 to build, when
there are only 3 hectares to construct on and $35,000 to spend. Working with known constraints
allows a designer to set up the optimal design by accounting for all of the factors that affect
wetland design and construction.
A drawback of modeling the system with a continuously stirred tank reactor design is that
one can not determine the effect of the systems shape, length and width on total retention by the
system. This problem can be addressed with a distributed modeling style, but it is unknown if
the increase in data input requirements make this style feasible since the model would be very
difficult to calibrate, validate and apply.
E.
WET model was evaluated with data collected from a wetland site built for municipal
wastewater treatment located in Benton, Kentucky. Model evaluation included the calibration
and validation of the model by using two statistical analyses, performing a sensitivity analysis
and using SET-WET to demonstrate its application for design of wetlands.
A non-parametric Wilcoxon Rank Sum statistical analysis indicated eight out of nine
examined outflow predictions were not statistically different from the measured observations.
Linear regression analysis showed that six out of nine examined parameters were statistically
similar, and that within the expected operating range, all of the examined outflow parameters
152
were within the 95% confidence intervals of the regression lines. A sensitivity analysis showed
the most significant input parameters to the model were those which directly affect bacterial
growth and oxygen uptake and movement. SET-WET was applied to a hypothetical simulation
for the QN2 subwatershed in the Nomini Creek watershed located in Virginia. Various designs
were applied to the area to determine which design parameters had the largest control upon
sediment and nutrient retention at the site. An optimal site design can be made based on the
SET-WET output and constraints from the site. Selecting one optimal wetland design depends
on many factors, but when considering nutrient retention only, SET-WET may assist in the
optimization of wetland design.
153
V.
problems. Constructed wetlands are an attractive BMP because in addition to controlling NPS
pollutants, they provide other beneficial functions to the environment such as wildlife habitat and
recreation, among others.
One of the more effective ways to enhance the design and implementation of wetlands is
to use computer models. Models provide an ability to make comparisons among alternative
designs and management strategies, thus allowing a wetland to be optimally utilized for its
intended purpose. A model, SET-WET, has been developed which allows these comparisons to
be made.
The SET-WET model is a user-friendly, dynamic, simulation model for design and
evaluation of constructed wetlands in order to optimize NPS pollution control measures.
SET-WET is written in Fortran 77 to facilitate linking with existing NPS models. The model
simulates the hydrologic, N, C, bacteria, DO, vegetative, P and sediment cycles within a wetland
system. The model allows for either free water surface (FWS) or subsurface flow (SSF) wetland
simulations, and is designed in a modular manner; thus it gives the user the flexibility to
concentrate on simulation of specific cycles and processes. The season/time period breakdown
accounts for seasonal variation by allowing the user to change parameter values in the middle of
a simulation run. It allows two forms of data input, one based on measured daily values and
another based on estimates from the SCS curve method in conjunction with runoff concentration
coefficients. Designed as a continuously stirred tank reactor, the model assumes that all
incoming constituents are evenly mixed throughout its entire volume.
The model was calibrated and validated with limited data collected from a constructed
wetland located in Benton, Kentucky. This data was not ideal, as monthly data points needed to
be interpolated to daily values, and the site was designed for wastewater treatment, not NPS
pollution control. Parameter input values were based on previous research and site data. Nonparametric statistical analyses of the validated results indicated that the predicted hydrologic,
NO3-, NH4+, organic N, BOD5, TSS, dissolved and total P concentrations were not statistically
different from the measured observations. The Wilcoxon-Rank statistical analyses indicate that
the confidence in the organic N, TSS, and DP predictions were high as the respective p-values
154
approached 1.0. Predictions for DO were consistently higher than observed values, which are a
cause for concern as the DO concentration directly affects many of the bacteria processes.
Linear regression analysis showed that six out of nine examined parameters were statistically
similar, and that within the expected operating range, all of the nine examined outflow
parameters were within the 95% confidence intervals of the regression lines. Due to the limited
amount of data points used for the statistical analyses, results need to be used with caution.
The sensitivity analysis showed that the model was most sensitive to parameters that
directly or indirectly affect bacterial growth or oxygen concentrations in the system. The model
was very sensitive to parameters such as the heterotrophic DO yield and the anaerobic maximum
growth rate in the soil, as values became incomprehensible. In general, however, the NCOB,
sediment and P cycles are relatively insensitive to changes to any single parameter due to the
complexity of the wetland system.
Hypothetical simulations were conducted for design of a possible wetland located in the
QN2 subwatershed of the Nomini Creek watershed in Virginia. The model made predictions on
the nutrient retention of five wetland designs (2.7 hectare, 5.4 hectare, 1.1 hectare, 2 * Plants,
50% Substrate) which when used in conjunction with other constraints (land, ecological,
monetary), a designer would be able to select an optimum design for the wetland. The designed
wetlands reduced the percentage of BOD5 (39.9%-45.5%), TSS (84.5%-99.9%), total N (41.9%56.4%), and total P (37.7%-56.5%) to levels reported by previous researchers.
SET-WET differs from many existing wetland models in that it uses a systems approach,
and limits the assumptions made concerning the interactions of the various nutrient cycles in a
wetland system. It accounts for C and N interactions, as well as for effect of oxygen levels upon
microbial growth. It also directly links microbial growth and death to the consumption and
transformations of nutrients in the wetland system. Many previous models have accounted for
these interactions with zero and first order rate equations that assume rates are dependent only on
initial concentrations.
SET-WET was developed to be as general as possible, but there were assumptions made
during model development which might not apply to all wetland locations. The model assumes
that the FWS system always has free lying surface water, and that the SSF system never floods.
It does not account for snow or snow melt and assumes that the dissolved and particulate
concentrations in both the surface water and substrate water are evenly mixed through the
155
2)
There is a need for an improved dissolved oxygen model that considers the biomass
oxygenation rate as a function of time. Oxygen is a very sensitive parameter in the SETWET model; therefore the assumption of constant rates over the entire growing season
may be limiting the models functions.
3)
Improve the capabilites of the plant model. There is a need to account for the turnover
ratio of plant life throughout the growing season and not have all death occur only during
a winter period.
4)
156
5)
Examine if the modeling of diffusion in the system is quantified properly. Examine if the
diffusion coefficient needs to be a function of time and concentration, instead of a
uniform rate through each season period.
6)
Better quantify the initial microbial biomass for HT and AT populations, since the model
is sensitive to these values. In addition, quantify growth parameters in situations that are
more applicable to NPS conditions.
7)
Link the model to an existing NPS model, such as ANSWERS, to improve incoming
nutrient flows when the input data is unknown.
8)
Extend the model to a series of continuously stirred tank reactors. This would allow the
analysis of wetland length to width ratios, and should allow more accurate predictions of
hydrologic and nutrient effluent.
The SET-WET model takes a novel approach in its attempt to model the constructed
wetland system. It varies from similar models due to its modularity, extreme generality, time
management approach, and the fact that it models both FWS and SSF wetlands. In many ways
this flexibility is the models strongest asset, because it provides the model user the ability to
attempt a number of designs for constructed wetlands.
In closing, the SET-WET model is a promising start to modeling FWS constructed
wetlands in a manner that incorporates most of the interactions in a wetland system. There is a
need to further test the model's capabilities, but a useful design tool for both FWS and SSF
constructed wetlands has been developed.
157
VI.
Cited Work:
Abtew, W. (1996) Evapotranspiration measurements and modeling for three wetland systems in
south Florida. Water Resources Bulletin 32(3): 465-473
Allen, R. G., M. E. Jensen, J. L. Wright, and R. D. Burman (1989) Operational estimates of
reference evapotranspiration. Agronomy Journal 81: 650-662.
ASTM (1995) Standard practice for evaluating mathematical models for the environmental fate
of chemicals. Standard E 978-92, American Society for Testing and Materials, Philadelphia.
Baker, L. A. (1993) Introduction to nonpoint source pollution and wetland mitigation. In:
Richard K. Olson (editor), Created and natural wetlands for controlling nonpoint source
pollution/Office of Wetlands, Oceans, and Watersheds. U.S. Environmental Protection
Agency, 7-41.
Baker, J. A. (1973) Wetland hydrology. In M. W. Lefor and W. C. Kennard (editors), Proceedings
of the first annual wetlands conference, Storrs, Conn, U Conn, 84-95.
Bastian, R. K., and D. A. Hammer (1993) The use of constructed wetlands for wastewater
treatment and recycling. In: Gerald A. Moshiri (editor), Constructed wetlands for water
quality improvement. Boca Raton: Lewis Publishers, 59-67.
Benefield, L., and F. Molz (1984) Mathematical simulation of a biofilm process. Biotechnology
and Bioengineering 27: 921-931.
Benefield, L. D., J. F. Judkins Jr., and A. D. Parr (1984) Treatment plant hydraulics for
environmental engineers. Prentice-Hall, Inc., Englewood Cliffs, New Jersey.
Blom, C. W., G. M. Bogemann, P. Laan, A. J. M. Candersman, H. M. vandersteeg, and L. A.
Vosenek. (1990) Adaptations to flooding in plants from river areas. Aquatic Botany 38: 2947.
Bowmer, K. H. (1987) Nutrient removal from effluents by and artificial wetland: Influence of
rhizosphere aeration and preferential flow studied using bromide and dye tracers. Water
Resources 21(5): 591-599.
Boyd, C. E. (1978) Chemical composition of wetland plants. In: R. E. Good, D. F. Whigham, and
R. L. Simpson (editors). Freshwater wetlands: ecological processes and management
potential. New York: Academic Press, 155-167.
Brannan, K. (1999) Personal communication: Blacksburg, VA: July, 1999
Brezonik, P. L. (1972) Nitrogen: sources and transformations in natural waters. In: H. E. Allen
and J. R. Kramer (editors), Nutrients in natural waters. Wiley Interface, New York, 1-50.
158
Burns, R. C. and R. W. F. Hardy (1975) Nitrogen fixation in bacteria and higher plants. SpringerVelag, New York, New York.
Chapra, S. C. (1997) Surface water-quality modeling. McGraw-Hill, New York, New York.
Choate, K. D., J. T. Watson, and G. R. Steiner (1990) Demonstration of constructed wetlands for
treatment of municipal wastewaters, monitoring report for the period of March 1998 to
October 1989. TVA/WR/WQ90/11, Tennessee Valley Authority.
Christensen, N., W. J. Mitsch, and S. E. Jorgensen (1994) A first generation ecosystem model of
the Des Plains river experimental wetlands. Ecological Engineering 3: 495-521.
Cowardin, L.M., V. Carter, F. C. Golet, and E. T. Laroe (1979) Classification of wetlands and
deepwater habitats in the United States, U.S. Fish and Wildlife Service pub. FWS/OBS79/31, Washington, DC.
Costanza, R., and F. H. Sklar (1985) Articulation, accuracy and effectiveness of mathematical
models: A review of freshwater wetland applications. Ecological Modelling 27: 45-68.
Cussler, E. L. (1997) Diffusion: mass transfer in fluid systems. Cambridge University Press,
United Kingdom.
Daukas, P., D. Lowry, and W. W. Walker Jr. (1989) Design of wet detention basins and
constructed wetland for treatment of stormwater runoff from a regional shopping mall in
Massachusetts. In: Donald A. Hammer (editor), Constructed wetlands for wastewater
treatment: municipal, industrial, and agricultural, Lewis Publishers, Inc., Chelsea, Michigan,
686-694.
Dennison, M. S., and J. F. Berry (1993) Wetlands: guide to science, law, and technology. Noyes
Publications. Park Ridge, New Jersey.
Department of Conservation and Recreation (1996) NPS: nonpoint source pollution and you
(brochure). Virginia Department of Conservation and Recreation, Richmond, VA.
Detenback, N. E., and P. L. Brezonik (1991) Phosphorous sorption by sediments from a softwater seepage lake. 1. An evaluation of kinetic and equilibrium models. Environmental
Science and Technology, 25: 395-403.
Dorge, J. (1994) Modeling nitrogen transformations in freshwater wetlands. Estimating nitrogen
retention and removal in natural wetlands in relation to their hydrology and nutrient loadings.
Ecological Modeling, 75/76:409-420.
Dorge, J. (1998) Personal communication: Blacksburg, VA: September, 1998.
159
Dierberg, F. E., and P. L. Brezonik (1984) Nitrogen and phosphorous mass balances in a cypress
dome receiving wastewater. In: K. C. Ewel and H. T. Odum (editors), Cypress swamps.
Gainesville, University of Florida Press, 112-118.
Dillaha, T. A. (1990) Role of best management practices in restoring the health of the
Chesapeake Bay. In: Perspectives on the Chesapeake Bay, 1990: Advances in Estuarine
Sciences. Chesapeake Bay Program, USEPA, Washington, DC CBP/TRS41/90.
Duever, M. J. (1988) Hydrologic processes for models of freshwater wetlands. In : W.J. Mitsch,
M. Straskraba and S. E. Jorgensen (editors), Wetland Modelling, Elsevier, Amsterdam, 9-39.
Environmental Law Institute, Wetlands deskbook. (Washington, D.C.: author) pg. 18.
EPA (1988) Constructed wetlands and aquatic plant systems for municipal wastewater treatment:
Design manual EPA/625/1-88/022. Washington, DC: Office of research and development.
Fennessy, M. S., C. C. Brueske, and W. J. Mitsch (1994) Sediment deposition patterns in restored
freshwater wetlands using sediment traps. Ecological Engineering 3:409-428.
Fetter, C. W. (1994). Applied hydrogeology. Prentice Hall, Upper Saddle River, New Jersey.
Fields, S. (1993) Regulations and policies relating to the use of wetlands for nonpoint source
control. In: Richard K. Olson (editor), Created and natural wetlands for controlling nonpoint
source pollution/Office of Wetlands, Oceans, and Watersheds. U.S. Environmental Protection
Agency, 151-158.
French, R. H. (1984) Lake modeling: state of the art. Critical Reviews in Environmental
Chemistry. 13(4): 311-357.
Fyock, O. L. (1977) Nitrification requirements of water reuse systems for rainbow trout.
Colorado Division of Wildlife, Fisheries Research Section, No. 41; February 1977.
Gambrell, R. P. and W. H. Patrick Jr. (1978) Chemical and microbiological properties of
anaerobic soils and sediments. In: D.D. Hook and R. M. M. Crawford (editors), Plant Life in
Aerobic Environments, Ann Arbor Sci. Pub. Inc., Ann Arbor Michigan, 375-423.
Gidley, T. M. (1995) Development of a constructed subsurface flow wetland simulation model.
MS Thesis, North Carolina State, pp. 139.
Good, R. E., D. F. Whigham, and R. L. Simpson. (1978) Freshwater wetlands: ecological
processes and management potential. Academic Press, New York, New York.
Grady Jr., C. P. L. and H. C. Lim. (1980) Biological wastewater treatment theory and
applications. New York: Marcel Dekker Inc.
160
161
162
Mitsch, W. J., and J. G. Gosselink (1986) Wetlands. Van Nostrand Reinhold. New York, New
York.
Mitsch, W. J., and J. G. Gosselink (1993) Wetlands. Van Nostrand Reinhold. New York, New
York.
Mitsch, W. J., J. K. Cronk, X. Wu, R. W. Nairn, and D. L. Hey (1995) Phosphorous retention in
constructed freshwater riparian marshes. Ecological Applications, 5(3):830-845.
Mitsch, W.J., and B. C. Reeder. (1991) Modelling nutrient retention of a freshwater coastal
wetland: estimating the roles of primary productivity, sedimentation, resuspension, and
hydrology. Ecological modelling 54:151-187.
Mostaghimi, S., S. Shukla, and P. W. McClellan (1998) BMP Impacts on nitrate and pesticide
transport to groundwater in the Nomini creek watershed: Report No. NC-0298.
Niswander S. F., and W. J. Mitsch (1995) Functional analysis of a two-year old created in-stream
wetland: Hydrology, phosphorus retention, and vegetation survival and growth. Wetlands,
15(3):212-225.
Novotny, V. and G. Chesters (1981) Handbook of nonpoint source pollution sources and
management. Van Nostrand Reinhold Company. New York, New York.
Novotny, V., and H. Olem (1994) Water quality: prevention, identification and management of
diffuse pollution. Van Nostrand Reinhold. New York, New York.
Ott, R. L. (1993) An introduction to statistical methods and data analysis. Duxbury Press.
Belmont, CA.
Perrich, J. R. (1981) Activated carbon adsorption for wastewater treatment. CRC Press, Inc.
Boca Raton, FL.
Raisin, G. W. and D. S. Mitchell (1995) The use of wetlands for the control of non-point source
pollution. Water, Source and Technology, 32(3): 177-186.
Reddy, K. R., Khaleel, R., Overcash, M. R., and P. W. Westerman (1979) A nonpoint source
model for land areas receiving animal wastes. I: Mineralization of organic nitrogen.
Transactions of the ASAE 22: 863.
Reddy, K. R. (1982) Mineralization of nitrogen in organic soils. Soil Sci. Soc. Am. J., 46(3):
561-566.
Reddy, K. R. and W. H. Patrick (1984) Nitrogen transformations and loss in flooded soils and
sediments. CRC Critical Reviews in Environmental Control, 13(4): 273-309.
163
Reed, S. C. (1994) Design of subsurface flow constructed wetland for wastewater treatment. In:
S. C. Reed, E. J. Middlebrooks, and R. W. Crites (editors). Natural systems for waste
management and treatment, 2nd Edition. New York, McGraw and Hill.
Ruffner, J. A. (1980) Climates of the states, Volume 2. Gale Research, Detroit, Michigan.
Salvesen, D. (1994) Wetlands: mitigating and regulating development impacts. Second edition.
Washington D.C.: ULI the Urban Land Institute.
Shaw, S. P., and C. G. Fredine (1956) Wetlands of the United States, their extent, and their value
for waterfowl and other wildlife. U.S. Department of Interior, Fish and Wildlife Service,
Circular 39, Washington D.C.
Silverman, G. S. (1989) Treatment of nonpoint source pollutants urban runoff and agricultural
wastes: Development of urban runoff treatment wetlands in Fremont, California. In: Donald
A. Hammer (editor), Constructed wetlands for wastewater treatment: municipal, industrial,
and agricultural, Lewis Publishers, Inc., Chelsea, Michigan, 669-676.
Soil Conservation Service (1968) Hydrology, supplement A to sec.4, Engineering Handbook,
USDA-SCS, Washington D.C.
Snoeyink, V. L., and D. Jenkins (1980) Water chemistry. John Wiley & Sons, New York, New
York.
Tanji K. K. (1982) Modeling of the soil nitrogen cycle. In: F. J. Stevenson (editor). Nitrogen in
agricultural soils. Madison, WI: American Society of Agronomy, Inc., Crop Science Society
of America, Inc., and Soil Science Society of America Inc., 721-772.
Tanner, C. C. (1996) Plants for constructed wetland treatment systems A comparison of the
growth and nutrient uptake of eight emergent species., Ecological Engineering (7) 59-83.
Tchobangolous, G. and F. L. Burton (1991) Wastewater engineering: treatment, disposal and
reuse. Third edition. New York: Metcalf and Eddy, Inc. McGraw-Hill, Inc.
Teague L. L., A. L. Kenimer, B. J. Lesker, and F. L. Mitchell (1997) Constructed wetlands for the
control of agricultural nonpoint source pollution. ASAE Paper No. 972037
Tiner Jr., R. W. (1984) Wetlands of the United States: current status and recent trends, National
Wetlands Inventory Project. U.S. Fish and Wildlife Service.
USDA, Soil Conservation Service (1988) National handbook of conservation practices.
Washington, D. C.
USEPA (1993) Created and natural wetlands for controlling nonpoint source pollution. Boca
Raton: Smoley C.K.
164
USSCS (1975) Soil taxonomy: A basic system of soil classification for making and interpreting
soil surveys. U. S. Soil Conservation Service Agricultural Handbook 436. Washington, D.C.
Urban, N. R., and S. J. Eisenreich (1988) Nitrogen cycling in a forested Minnesota bog.
Canadian Journal of Botany, 66: 435-449.
Walton, R., R. S. Chapman, and J. E. Davis (1996) Development and application of the wetlands
dynamic water budget model. Wetlands, 16(3): 347-357.
Watson, J. T., Sherwood, C. R., Kadlec, R. H., Knight, R. L., and A. E. Whitehouse (1989)
Performance expectation and loading rates for constructed wetlands. In: Donald A. Hammer
(editor), Constructed wetlands for wastewater treatment: municipal, industrial, and
agricultural, Lewis Publishers, Inc., Chelsea, Michigan, 319-351.
Wheaton, F., J. Hochheimer, and G.E. Kaiser (1991) Fixed film nitrification filters for
aquaculture. In: Brune, D.E. and J. R. Tomasso (editors), Aquaculture and Water Quality.
Baton Rouge, LA: The World Aquaculture Society, 272-303.
Whitaker, G., and C. R. Terrell (1993) Federal programs for wetland restoration and use of
wetlands for nonpoint source pollution control. In: Richard K. Olson (editor), Created and
natural wetlands for controlling nonpoint source pollution/Office of Wetlands, Oceans, and
Watersheds. U.S. Environmental Protection Agency, 201-215.
Widener, A. S. (1995) A mathematical model of the nitrogen cycle in a constructed wetland. MS
Thesis, Virginia Polytechnic Institute and State University. pp. 145.
Wood. R. A. (1996) The weather almanac. Gale research, Detroit, MI.
Zak, D. R. and D. F. Grigal (1991) Nitrogen mineralization, nitrification, and denitrification in
upland and wetland ecosystems. Oegologia, 88: 189-196.
165
VII. Appendices
166
DESCRIPTION
UNITS
Amount of standing dead biological mass removal
g biomass/day
Amount of living biological mass removal
g biomass/day
Type of adsorption model used
Aerobic heterotroph maximum growth rate at optimal
conditions in peat water
day -1
Aerobic heterotroph maximum growth rate at optimal
conditions in surface water
day -1
growth of bottom aerobic heterotrophs
g microbes/day
growth of surface aerobic heterotrophs
g microbes/day
Ambient Air Temperature
degrees C
Kadlec equation (slope exponent)
plant ammonium uptake from wetland bottom
g N/ day
plant ammonium uptake from wetland surface
g N/ day
fraction of heterotrophs that are anaerobes - function g anaerobic HT/ g
of DO conc. in bottom
total HT
fraction of heterotrophs that are anaerobes - function g anaerobic HT/ g
of DO conc. in surface
total HT
Angle of weir notch (outlet=2)
degrees
growth of bottom anaerobic heterotrophs
g microbes/day
growth of surface anaerobic heterotrophs
g microbes/day
Anaerobic heterotroph maximum growth rate at
optimal conditions in peat water
day -1
Anaerobic heterotroph maximum growth rate at
optimal conditions in surface water
day -1
amount of point flow contribution (point=1)
m 3 /day
Area of outflow pump (outlet=3)
Atmospheric Deposition (0=not include, 1=include)
Kadlec equation (depth exponent)
Fraction of carbon in plants
standing dead degradation rate - set for 99% over the
course of a year - converted to POC
Living biomass density
oxygen flux from rootzone aeration by plants
Initial total plant mass
total plant carbon mass
total biomass in the wetland system
volume of biomass
Ratio of carbon to nitrogen in plants
biomass death
biomass growth
Rate at which plants grow
amount of biomass removal for external reasons
m2
g C/ g biomass
day -1
kg/m 3
g O2/day
g Plant
gC
g biomass
g biomass
g C/g N
g biomass/day
g biomass/day
g biomass/
m 2 -day
g biomass/day
167
MODEL
COMPONENT
Biompn
Biompp
Biooxrb
DESCRIPTION
Ratio of plant mass to nitrogen in plants
Biomass plant/phosphorous ratio
rate at which oxygen is added to the rootzone by
plants
Bocconc
Bocdinfco
Bodcfrac
Bodpfrac
Bodrc
Bioremd
Bioreml
UNITS
g plant mass/g N
g plant/g
phosphorous
gO 2 /m 2 *day
mg BOD5/L
mg BOD5/L
g C/g BOD5
g. part. BOD5/ g
total BOD5
mg BOD5/L
g DON/m 2
g NH4 /m 2
Dadno3
g NO3 /m 2
Dadpon
Daylen
g PON/m 2
hours daylight/12
day of season
period
day of season
period
days
day of season
period
day of season
period
btphos
Btphosi
BTPHOST
check (1-7)
Coefk
Contc
Daddon
Dadnh4
Dayremd
Dayreml
Daysdeg
Dayw ins
Dayw ine
deadphos
deadphot
deathb
deathw
Decomp
Degbio
g PP
g P-phos
g PP
g PP/day
g PP/day
g N/day
g N /day
g deg. SS/
g SS
% degradation
168
MODEL
COMPONENT
deltaht
denitb
denitw
Diseffc
disphosc
Disphosc
disphosi
DOCB
doccb
doccw
Docininc
Docinitb
Docinitw
docleach
docminib
docminiw
docmt
Doconcin
Doconcp
docout
docperc
DOCW
doinf
Doinitb
Doinitw
donammb
donammw
donat
DONB
doncb
Donconc
doncw
donec
donimmb
donimmw
donin
Donininc
DESCRIPTION
change in wetland bottom height due to peat
accumulation, sediment resuspension/settling
conversion of bottom nitate to N2 gas by bottom
anaerobic heterotrophs
conversion of surface nitate to N2 gas by surface
anaerobic heterotrophs
Effective discharge coefficient
DP watershed runoff inflow concentration
Dissolved phosphorous inflow concentrations from
catchment
total amount of DP entering wetland system
amount of dissolved organic carbon in bottom
dissolved organic carbon concentration in bottom
dissolved organic carbon concentration in surface
DOC concentration for percolating sources
Initial amount of dissolved organic carbon in peat
water
Initial amount of dissolved organic carbon in surface
water
leaching of DOC from standing dead to DOC
conversion of bottom DOC to microbial biomass and
carbon dioxide
conversion of surface DOC to microbial biomass and
carbon dioxide
DOC mass transfer
Dissolved oxygen concentration in water input
Oxygen concentration in precipitation
effluent DOC flux
DOC percolation/infiltration to/from wetland bottom
amount of dissolved organic carbon in surface
dissolved oxygen additions from watershed runoff
Total mass of oxygen in the peat water
Total mass of oxygen in the surface water
conversion of excess DON to ammonium by bottom
heterotrophic bacteria during organics degradation
conversion of excess DON to ammonium by surface
heterotrophic bacteria during organics degradation
DON contributed by atmospheric deposition
amount of dissolved organic nitrogen in bottom
dissolved organic nitrogen concentration in bottom
DON concentration in point flow
dissolved organic nitrogen concentration in surface
DON effleunt concentration
incorporation of DON by bottom heterotrophic
microbes during organics decomposition
incorporation of DON by surface heterotrophic
microbes during organics decomposition
incoming DON from watershed runoff additions
DON concentration from percolating source
UNITS
m
g N/ day
g N/ day
m
mg DP/L
mg/l
g DP/day
gC
mg DOC/L
mg DOC/L
mg C/L
gC
gC
g C/ day
g C/day
g C/day
g C/day
mg O2/L
mg O2/L
g C/day
g C/day
gC
g O2/day
g O2
g O2
g N/ day
g N/ day
g DON/ day
g DON
mg DON/L
mg DON/L
mg DON/L
g N/l
g N /day
g N /day
g N/day
mg DON/L
169
MODEL
COMPONENT
Doninitb
Doninitw
donminb
donminw
donmt
donout
donperc
DONW
doout
doperc
DOXYB
Doxycsat
DOXYW
dphosec
dphosout
DTPHOSB
dtphoscb
dtphoscw
Dtphosib
Dtphosiw
DTPHOSW
Evap
evapcoef
Evapt
evapt
evaptt
fixnit
Flowout
flowrate
Freundk
Freundn
Hb
heatij
heatindx
HETEROB
DESCRIPTION
Initial dissolved organic nitrogen in peat water
Initial dissolved organic nitrogen in surface water
conversion of excess bottom DON to ammonium by
bottom heterotrophic bacteria during organics
degradation
conversion of excess surface DON to ammonium by
surface heterotrophic bacteria during organics
degradation
mass transfer of DON between surface and bottom
effluent DON flux
DON percolation/infiltration from bottom
amount of dissolved organic nitrogen in surface
effluent disolved oxygen flux
percolation/infiltration of DO from wetland bottom
amount of dissolved oxygen in wetland bottom
DO saturation constant
amount of dissolved oxygen in wetland surface
Effluent dissolved phosphorous concentration
effluent dissolved phosphorous flux
total dissolved phosphorous in wetland bottom
Dissolved phosphorous concentration for bottom
Dissolved phosphorous concentration for surface
Initial amount of dissolved phosphorous in wetland
bottom water pool
Initial amount of dissolved phosphorous in wetland
surface water pool
total dissolved phosphorous in wetland surface
Evaporation model (0=Thornthwaite, 1=pan)
calculated climate parameter for Thornthwaite's ET
method
Amount of evaporation from wetland cell
water loss due to evapotranspiration
amount of water loss due to evapotranspiration for
hourly time step
amount of atmospheric nitrogen converted to DON
by microbes
Amount of water removed (outlet=4)
calculated parameter for SCS curve method amount of surface runoff
Freundlich isotherm constant
Freundlich isotherm constant
Bottom of peat component in wetland cell
calculated climate parameter for Thornthwaite's ET
method
calculated climate parameter for Thornthwaite's ET
method
amount of heterotrophs in bottom
UNITS
gN
gN
g N/day
g N/day
g DON/day
g N/day
g DON/day
g DON/day
g O2/day
g O2/day
g O2
mg O2/L
g O2
mg DP/L
g DP/day
g DP
mg DP/L
mg DP/L
g DP
g DP
g DP
m3/day
m3
m3/hour
g ON/day
m3/day
mm
m
g microbes
170
MODEL
COMPONENT
Heteroib
Heteroiw
HETEROW
Hii
hit
hnh4immb
hnh4immw
Hno3hscb
Hno3hscw
Ho
Horghscb
Horghscw
Hout
Hover
hrt
htdeathb
htdeathw
Htdohscb
Htdohscw
Htdoyb
Htdoyw
Htdrb
Htdrw
htgrow b
htgrow w
Hti
Htno3y
htrespb
htrespw
httempfb
httempfw
DESCRIPTION
Initial total mass of heterotrophic bacteria in peat
water
Initial total mass of heterotrophic bacteria in surface
water
amount of heterotrophs in surface
Initial height of water in wetland cell relative to (z=0)
wetland surface water height for hourly time step
bottom ammonium nitrogen utilized by bottom
heterotrophic bacteria during organics degradation
surface ammonium nitrogen utilized by surface
heterotrophic bacteria during organics degradation
Anaerobic heterotroph nitrate half saturation
constant for peat water
Anaerobic heterotroph nitrate half saturation
constant for surface water
Top of wetland cell
Heterotroph organics half saturation constant for
peatwater
Heterotroph organics half saturation constant for
surface water
Height of overflow for outlet option (outlet=1, 2 , 3 ,4,
or 6)
Height for water overflow of weir (outlet=1)
hydraulic residence time
death of bottom heterotrophs
death of surface heterotrophs
Aerobic heterotroph dissolved oxygen half saturation
constant in surface water
Aerobic heterotroph dissolved oxygen half saturation
constant in peat water
Oxygen yield of aerobic heterotrophs in peat water
Oxygen yield of aerobic heterotrophs in surface
water
Heterotroph death rate in peat water
Heterotroph death rate in surface water
growth of bottom heterotrophs
growth of surface heterotrophs
Top of peat component in wetland cell
Nitrate yield of anaerobic heterotrophs
bottom oxygen loss by bottom heterotroph growth
surface oxygen loss by surface heterotroph growth
Heterotroph bottom temperature factor - reduces
growth rate at nonoptimal conditions
Heterotroph surface temperature factor - reduces
growth rate at nonoptimal conditions
htyieldb
htyieldw
UNITS
g microbes
g microbes
g microbes
m
m
g NH4+-N/day
g NH4+-N/day
mg NO3--/L
mg NO3--/L
m
mg C/L
mg C/L
m
m
days
g microbes/day
g microbes/day
mg O2/L
mg O2/L
g microbes/g O2
g microbes/g O2
g microbes/day
g microbes/day
g microbes/day
g microbes/day
m
g microbes/
g NO3--N
g O2/day
g O2/day
g microbes/ g C
degraded
g microbes/ g C
degraded
171
MODEL
COMPONENT
Hydconp
Hydtype
Imminitb
Imminitw
IMMNB
immnout
immnoutd
IMMNW
j
k
Khcoef
Leachr
Length
Linpartc
m
Mannc
Mannc
Microbec
microdb
microdw
Micronc
mictcnb
mictcnw
Mtcdoc
Mtcdon
Mtchphos
Mtcnh4
Mtcno3
Mtdox
mtdoxy
mtfws
Mtfwsdoc
ndeathb
ndeathw
Ndohsatb
Ndohsatw
Ndrateb
Ndratew
nh4at
NH4B
nh4cb
Nh4conc
DESCRIPTION
Hydraulic conductivity of wetland gravel bed
Hydrologic input (0 = NPS pollution runoff model,
1=SCS runoff curve method)
Nitrogen in plants and microbes in peat water
Nitrogen in plants and microbes in surface water
nitrogen in bottom plants and microbes
Removal of immoblized nitrogen in living biomass
due to external removals
Removal of immoblized nitrogen in standing dead
due to external removals
nitrogen in surface plants and microbes
equivalent to TIMPER
equivalent to number of time periods
head correction factor (outlet=2)
Rate at which biomass leaches DOC
Average length of wetland cell
Linear isotherm partition coefficient
equivalent to STRPER
Manning's coefficient
Manning's coefficient
Fraction of cell mass that is carbon
sloughing of dead bottom microbial cells to POC
sloughing of dead surface microbial cells to POC
Fraction of nitrogen in microbial cells
ratio of carbon to nitrogen in bottom microbial cells
ratio of carbon to nitrogen in surface microbial cells
Mass transfer coefficient for DOC
Mass Transfer coefficient for DON
Mass transfer coefficient for DP
Mass transfer coefficient for NH4
Mass transfer coefficient for NO3
Mass transfer coefficient for oxygen
mass transfer of DO between surface and bottom
Re-aeration of DO between water surface and
atmosphere
Re-aeration mass transfer coefficient
death of Nitrosomonas in bottom
death of Nitrosomonas in surface
Nitrosomonas dissolved oxygen half saturation
constant in peat water
Nitrosomonas dissolved oxygen half saturation
constant in surface water
Nitrosomonas death rate in peat water
Nitrosomonas death rate in surface water
NH4 additions due to atmospheric deposition
amount of ammonium in wetland bottom
ammonium concentration in wetland bottom
NH4 concentration in point flow
UNITS
m/day
gN
gN
gN
g N/day
g N/day
gN
# time period
# of time periods
m
g C/g N
m
l/g
# stress period
s/m1/3
s/m1/3
g C/g microbes
g C/day
g C/day
g N/g microbes
g C/g N
g C/g N
cm/s
cm/s
cm/s
cm/s
cm/s
cm/s
g O2/day
g O2/day
cm/s
g microbes/day
g microbes/day
mg O2/L
mg O2/L
g microbes/day
g microbes/day
g N/day
gN
mg NH4+-/L
mg NH4+-/L
172
MODEL
COMPONENT
nh4cw
nh4e c
nh4in
Nh4inc
Nh4ininc
DESCRIPTION
ammonium concentration in wetland surface
UNITS
mg NH4+-/L
mg NH4+-/L
mg NH4+-/L
Nh4initw
nh4mt
nh4out
nh4pe rc
Nh4rc
NH4W
Nitcycle
Nitfix
Nitfra te
Nh4initb
nitrifb
nitrifw
Nitrosib
Nitrosiw
NITROSOB
NITROSOW
nitupb
nitupw
nle a ch
Nma x grb
Nma x grw
Nnh4hscb
Nnh4hscw
no3a t
NO3B
no3cb
No3conc
no3cw
no3e c
no3in
No3inc
No3ininc
No3initb
mg NH4+-/L
g NH4
g NH4
g N/ day
g N/day
g N/day
mg NH4+-/L
gN
-
mg NO3--/L
mg NO3--/L
mg/l
g N/day
mg NO3--/L
mg NO3--/L
g NO3
173
MODEL
COMPONENT
No3initw
DESCRIPTION
Initial oxidized nitrogen in surface water
no3mt
no3out
no3perc
No3rc
NO3W
Nsdoyw
nsgrowb
nsgroww
nsrespb
nsrespw
Nsdoyb
nstempfw
Nsyieldb
Nsyieldw
nstempfb
Numstper
Numtmper
Onpartf
Orgininc
Orgnrc
Outflow
effluent flowrate
outflow
outflowt
Outlet
Outwidth
Oxyconc
Oxyrc
Panevap
Pbiouw
pbodin
Pcycle
Peatacrw
peatcacb
peatcacw
Peatcc
Peatdens
peatnacb
Peatacrb
UNITS
g NO3
g N/day
mg NO3--/L
g N/day
mg NO3--/L
gN
g microbes/g O2
g microbes/g O2
g microbes/day
g microbes/day
g O2/day
g O2/day
g microbes/ g
NH4+-N
g microbes/ g
NH4+-N
# of periods
# time steps
g PON/g TON
mg ON/L
mg ON/L
m 3 /day
m 3 /day
m 3 /hour
m
mg O2/L
mg O2/L
m/day
% mass
underwater
g C/day
g solids/day
g solids/day
g C/day
g C/day
g C/g solids
kg/m3
g N /day
174
MODEL
COMPONENT
peatnacw
Peatnc
Percinf
Percinfa
percinft
percitt
Ph
phdep
Phoscon
phosmt
phosper
phosperb
phosperb
phosperw
phosupb
phosupw
physdeg
physdegc
Pininc
pminpp
Pminppc
pminppt
POCB
poccb
poccw
Pocfall
Pocinitb
Pocinitw
pocminib
pocminiw
pocout
pocre
Pocres
pocset
Pocsize
DESCRIPTION
nitrogen in accumulating surface refractory solids
Fraction of nitrogen in accumulated solids
Percolation or Infiltration additions (0=no,1=yes)
Percolation/Infiltration rate for stress period
percolation/infiltration gains/losses to/from wetland
bottom
amount of percolation/infiltration changes for hourly
time step
pH of water in wetland system
pH factor for volatilization
Incoming amount of dissolved phosphorous from
point source additions
mass transfer of DP from surface and bottom
attachment ratio for incoming phosphorous to
respective sediment categories
ratio of phosphorous attachment upon bottom
sediment
attachment ratio for bottom phosphorous to
respective sediment categories
attachment ratio for surface phosphorous to
respective sediment categories
plant uptake of DP from bottom
plant uptake of DP from surface
degradation of standing dead to refractory material
conversion of standing dead to plants to POC
Dissolved phosphorous concentration for percolating
water source
phosphorous remineralization from surface
particulate pool for individual sediment category
Mineralization rate from surface particulate pool
total amount of remineralization from surface
particulate pool
amount of POC in bottom
particulate organic carbon concentration in bottom
particulate organic carbonconcentration in surface
POC Fallling rate
Initial amount of particulate organic carbon in surface
water
Initial amount of particulate organic carbon in peat
water
conversion of bottom POC to microbial biomass and
carbon dioxide
conversion of surface POC to microbial biomass and
carbon dioxide
effluent POC flux
POC resuspension form wetland bottom
UNITS
g N /day
g N/g solids
m/day
m3
m 3 /hour
pH
g DP/day
g DP/day
g DP/day
g DP/day
g biomass/day
g C/day
mg DP/L
g DP/day
day -1
g PP/day
g C/day
mg POC/L
mg POC/L
m/day
gC
gC
g C/day
g C/day
g C/day
g C/day
% (g POC
resuspended/g
POC in peat)
g C/day
mm
175
MODEL
COMPONENT
POCW
Point
ponammb
ponammw
ponat
PONB
poncb
Ponconc
poncw
ponec
Ponfall
ponimmb
ponimmw
ponin
Poninitb
Poninitw
ponminb
ponminw
ponout
ponre
DESCRIPTION
amount of POC in surface
Point flow addition (0=no, 1=yes)
conversion of excess bottom PON to ammonium by
bottom heterotrophic bacteria during organics
degradation
conversion of excess surface PON to ammonium by
surface heterotrophic bacteria during organics
degradation
PON additions due to atmospheric deposition
amount of PON in wetland bottom
particulate organic nitrogen concentration in bottom
PON concentration in point flow
particulate organic nitrogen concentration in surface
effluent PON concentration
PON fall rate
incorporation of bottom PON by bottom heterotrophic
bacteria during organics degradation
incorporation of surface PON by surface
heterotrophic bacteria during organics degradation
particulate organic nitrogen flux from watershed
Initial particulate organic nitrogen in peat water
Initial particulate organic nitrogen in surface water
conversion of excess bottom PON to ammonium by
bottom heterotrophic bacteria during organics
degradation
conversion of excess surface PON to ammonium by
surface heterotrophic bacteria during organics
degradation
effluent PON flux
resuspension of PON from wetland bottom
Ponres
ponset
Ponsize
PONW
Porosity
Porpeat
PPHOS
Pphoscal
pphoscw
pphosin
pphosinc
pphosinp
pphosint
pphosout
pphosres
pphosset
UNITS
g C/day
-
g N/day
g N/day
g N/day
g N/day
mg PON/L
mg PON/L
mg PON/L
mg PON/L
m/d
g N/day
g N/day
g N/day
gN
gN
g N/day
g N/day
g N/day
g N/day
g PON
resuspended/ g
PON in peat
g N/day
mm
g N/day
m 3 /m 3
m 3 /m 3
g PP
g P-Phos
mg PP/L
g PP/day
mg PP/L
g PP/day
g PP/day
g PP/day
g PP/day
g PP/day
176
M ODEL
CO M P ONENT
P P HOS T
P pphosi
P ra te up
pre cip
P re cipr
prm inbp
prm inbpt
P se dde p
P stduw
P rm inbpc
re 1
re 1b
re 2
REFC
Re fcinit
REFN
Re fninit
Re sthc
Re sthick
Re sthn
re susp
re suspv
re suspv
S ccurve
sde a dout
sde com p
se dbsa
se dbv
S e dca t
S e dcons
S e dcycle
se dde lta
se dde p
S e dfa ll
UNITS
g PP
m g P P /L
% tak en from
bottom
m3
m /day
g DP /day
day -1
g P P /day
s edim ent
depos ition/total
s urfac e s edim ent
Ratio m as s
underwater
-
g
g
g
g
C
C
N
N
m
m
m
g s ed./day
m /day
m /d
g biom as s /day
g S S /day
mm2
mm3
# c ategories
m g s ed./L
g s edim ent/day
g s ed./day
m /d
177
MODEL
COMPONENT
sedincv
Sedinitb
Sedinitw
sedint
Sedint
sedinw
sedout
sedpart
DESCRIPTION
incoming total volume of sediment for particle
category
Inititial amount of sediment weight in peat water
Initial amount of sediment weight in surface water
total amount of sediment runoff from watershed
Total incoming sediment from watershed area
amount of incoming sediment from all sources for
particle category
effluent sediment flux for particle category
amount of sediment particles for each particle class
sedtsap
sedtsatb
sedtsatp
sedtsatw
Sedper
SEDQTYB
SEDQTYW
Sedrc
Sedres
Sedsize
Sedspg
sedtotal
sedtsab
sedtsaw
So
sobodin
STANDDC
STANDDT
standdv
Standin
Stddens
storparm
strper
time
timper
Tj
TOCB
toccb
toccw
TOCW
TONB
toncb
UNITS
mm 3
g sediment
g sediment
g sed./day
g sed./day
g sed./day
g sed./day
# of particles
mass sed. cat./
mass sed. Total
g sed.
g sed.
mg sediment/l
% sediment
mm
g sediment/day
mm 2
mm 2
mm 2
mm 2
mm 2
mm 2
m/m
g C/ day
gC
g biomass
g biomass
g Plant
kg/m 3
# season period
# time period
degrees Celsius
g TOC
mg TOC/L
mg TOC/L
g TOC
g ON
mg TON/L
178
MODEL
COMPONENT
toncw
TONW
Toppump
tphosec
Volat
volatiz
DESCRIPTION
total organic nitrogen concentration in surface
Total organic nitrogen in wetland surface
Height of outflow top (outlet=4)
Total phosphorous effluent concentration
Volatilization (0=not included, 1=included)
conversion of ammonium ion to ammonium gas
Volatr
Waddon
Wadnh4
Volatilization rate
Wet atmospheric deposition rate for DON
Wet atmospheric deposition rate for NH4
Wadpon
Wasno3
WATERB
watervel
WATERVOL
Watinput
watint
watintt
Wattemp
Wettype
Width
winkco
winter
Wtrshdar
UNITS
mg TON/L
g ON
m
mg P/L
g N/day
day -1
mg DON/L
mg NH4+-/L
mg PON/L
mg NO3--/L
m3
m/day
m3
m3/day
m3/day
m3/hour
degrees C
m
m2
179
180
181
APFLOW (J)
End if
Else if HYDTYPE =1 then
If EVAP = 0 then
PRECIPR, AIRTEMP, DAYLEN
Else if EVAP = 1 then
PRECIPR, PANEVAP
End if
End if
If POINT = 1 then
APFLOW (J)
End if
BIOMASS submodel:
-VEGST (biomass season period)
-This is only called if NITCYCLE = 1 or SEDCYCLE =1
If M = 1 and WETTYPE = 0 then
BIOINIT, STANDIN, PEATACRW, PEATACRB, PEATDENS,
PRATEUP,BIODENS, STDDENS, PBIOUW, PSTDUW
Else if M = 1 and WETTYPE = 1 then
BIOINIT, STANDIN, PEATACRB, PEATDENS, PRATEUP
End if
If M =1 then
BIOMREML, DAYREML, ABIOREML, BIOMREMD, DAYREMD,
ABIOREMD, DAYWINS, DAYWINE, DEGBIO, DAYSDEG
If M>1 then
CHECK 1, CHECK2
If CHECK1 = 0 then
If WETTYPE = 0 then
PEATACRW, PEATACRB, PRATEUP, BIODENS,
STDDENS, PBIOUW, PSTDUW
Else if WETTYPE = 1 then
PEATACRB, PRATEUP
End if
If CHECK2 = 0 then
182
183
184
If CHECK2 = 0 then
If POINT = 1 then
BODCONC
If PERCINF =1 then
DOCININC
If HYDTYPE = 1 then
BODRC
-CARTM (carbon time period)
If HYDTYPE = 0 then
BODINFCO
NITROGEN submodel:
-NITSTR (nitrogen season cycle)
If M = 1 and WETTYPE = 0 then
ATDEP, NITFIX, VOLAT,
DONINITW, DONINITB, IMMINITW, IMMINITB, NH4INITW, NH4INITB,
NO3INITW, NO3INITB, PONINITW, PONINITB, REFNINIT, BIOMCN,
BIOMPN, HTNO3YW, HTNO3YB, MICRONC, NSYIELDW, NSYIELDB,
ONPARTF, PEATNC, PONRES, PONFALL, MTCDON, MTCNH4,
MTCNO3, PONSIZE, RESTHN, PONCOUT,
Else if M = 1 and WETTYPE = 1 then
ATDEP, NITFIX, VOLAT,
DONINITB, IMMINITB, NH4INITB, NO3INITB, PONINITB, REFNINIT,
BIOMCN, BIOMPN, HTNO3YB, MICRONC, NSYIELDB, ONPARTF, PEATNC
End if
If NITFIX = 1 then
NITFRATE
If VOLAT =1 then
VOLATR
If ATDEP =1 then
DADDON, WADDON, DADPON, WADPON,
DADNH4, WADNH4, DADNO3, WADNO3
If POINT = 1 then
DONCONC, PONCONC, NO3CONC, NH4CONC
If PERCINF = 1 then
DONININC, NH4ININC, NO3ININC
If HYDTYPE = 1 then
ORGNRC, NH4RC, NO3RC
185
If M>1 then
CHECK1, CHECK2, CHECK3, CHECK4, CHECK5, CHECK6, CHECK7
If CHECK1 = 0 then
If WETTYPE = 0 then
BIOMCN, BIOMPN, HTNO3YW,HTNO3YB, MICRONC, NSYIELDW,
NSYIELDB, ONPARTF, PEATNC, PONRES, PONFALL, MTCDON,
MTCNH4, MTCNO3, MANNC, PONSIZE, RESTHN, PONCOUT
Else if WETTYPE = 1 then
BIOMCN, BIOMPN, HTNO3YB, MICRONC,
NSYIELDB, ONPARTF, PEATNC
End if
If CHECK2 = 0 then
NITFRATE
End if
If CHECK3 = 0 then
VOLATR
End if
If CHECK4 = 0 then
DADDON, WADDON, DADPON, WADPON,
DADNH4, WADNH4, DADNO3, WADNO3
End if
If CHECK5 = 0 then
DONCONC, PONCONC, NO3CONC, NH4CONC
End if
If CHECK6 = 0 then
DONININC, NH4ININC, NO3ININC
End if
IF CHECK7 = 0 then
ORGNRC, NH4RC, NO3RC
End if
-NITTIME (nitrogen time period)
If VOLAT = 0 and HYDTYPE = 0 then
ORGNINC, NH4INC, NO3INC
Else if VOLAT = 1 and HYDTYPE = 1 then
ORGNINC, NH4INC, NO3INC, pH
Else if VOLAT = 1 and HYDTYPE =1 then
pH
186
End if
DISSOLVED OXYGEN submodel:
-OXYSTR (dissolved oxygen season period)
If M = 1 and WETTYPE = 0 then
DOINITW, DOINITB, HTDOYW, HTDOYB, NSDOYW, NSDOYB,
DOCONCP, MTDOX, MTFWSDOC, DOXYCSAT
Else if M = 1 and WETTYPE = 1 then
DOINITB, HTDOYB, NSDOYB, DOCONCP
End if
If POINT = 1 then
OXYCONC
If PERCINF = 1 then
DOININC
If HYDTYPE = 1 then
OXDRC
If M>1 then
CHECK1, CHECK2
If CHECK1 = 0 then
If WETTYPE = 0 then
HTDOYW, HTDOYB, NSDOYW, NSDOYB, DOCONCP,
MTDOX, MTFWSDOC, DOXYCSAT
Else if WETTYPE = 1 then
HTDOYB, NSDOYB, DOCONCP
End if
If CHECK2 = 0 then
If POINT = 1 then
OXYCONC
If PERCINF = 1 then
DOININC
If HYDTYPE =1 then
OXDRC
-OXYTIME (dissolved oxygen time period)
If HYDTYPE = 0 then
BIOOXRB, DOCONCIN
Else if HYDTYPE =1 then
BIOOXRB
187
End if
SEDIMENT submodel:
-SEDSTR (sediment season period)
-Can only be simulated when HYDTYPE = 0
If M = 1 then
SEDCAT, SEDRES
Now depending on the number of sediment categories (SEDCAT)
DO SEDCLASS=1,SEDCAT
SEDSIZE (SEDCLASS), SEDFALL (SEDCLASS), SEDINITW (SEDCLASS),
SEDINITB (SEDCLASS), SEDSPG (SEDCLASS), SEDPER (SEDCLASS)
CONTINUE
If M = 1 then
RESTHICK, MANNC, DECOMPR, PSEDDEP
If POINT =1 then
SEDCONC
If M>1 then
CHECK
IF CHECK = 0 then
RESTHICK, MANNC, DECOMPR
-SEDTIME (sediment time period)
SEDINT
PHOSPHOROUS submodel:
-PHOSSTR (phosphorous season period)
-Can only be called if SEDCYCLE =1
M = 1 then
DTPHOSIW, DTPHOSIB, BTPHOSI, PPHOSI, PMINPPC,
PRMINBPC, ADSORP, MTCPHOS, BIOMPP
If PERCINF = 1 then
PININC
If ADSORP = 0 then
FREUNDK, FREUNDN
188
189
C
C
C
C
C
C
C
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
190
/
/
/
REAL NITROSIW,NITROSIB,DOINITW,DOINITB,POCINITW,POCINITB,
REFCINIT,STANDIN,IMMINITW,IMMINITB,NH4INITW,NH4INITB,HRT,
NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,HETEROIW,
HETEROIB,LENGTH,WIDTH,HO,HB,SO,BIOINIT,PEATACRB,DECOMPR
HYDTYPE=0
POINT=0
NITCYCLE=0
PCYCLE=0
SEDCYCLE=0
WTRSHDAR=0
SCCURVE=0
ATDEP=0
NITFIX=0
VOLAT=0
/
/
6
/
/
/
191
7
/
/
/
/
/
/
/
DO 400 STRPER=1,NUMSTPER
M=STRPER
CALL HYDSTR(NUMTMPER,POINT,PORPEAT,M,WETTYPE,HYDCONP,
POROSITY,PERCINF,PERCINFA)
IF (NITCYCLE.EQ.1. OR. SEDCYCLE.EQ.1) THEN
CALL VEGST(BIOINIT,BIOMREML,DAYREML,ABIOREML,BIOMREMD,
DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,STANDIN,
PEATACRW,PEATACRB,M,PEATDENS,WETTYPE,PRATEUP,PBIOUW,
BIODENS,STDDENS,PSTDUW,DAYWINE)
ENDIF
/
/
/
IF (NITCYCLE.EQ.1) THEN
CALL CARSTR (DOCINITW,DOCINITB,POCINITW,POCINITB,
POCFALL,POCRES,BIOCCONT,BODCFRAC,MTCDOC,POINT,PERCINF,
BODPFRAC,LEACHR,MICROBEC,PEATCC,BODCONC,HYDTYPE,BODRC,
MANNC,POCSIZE,RESTHC,M,REFCINIT,WETTYPE,POCCOUT,DOCININC)
/
/
/
CALL BACSTR(NITROSIW,NITROSIB,HETEROIW,HETEROIB,M,NDRATEW,
NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,
AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,
HTDOHSCB,HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,WETTYPE)
/
/
/
/
/
/
/
/
/
/
/
/
/
C
192
275 FORMAT
/
/
ENDIF
(T12,'NH4EC',T23,'NO3EC',T33,'DONEC',T43,'PONEC',
T52,'DOXYCW', t61 'BOD5EC', t71, 'sedoutc', T81, 'DPHOSEC',
T91, 'TPHOSEC')
FROM VEGCYCLE
BIOMASST(0)=BIOINIT
STANDDT(0)=STANDIN
BIOMASSV(0)=BIOMASST(0)/1000*(1/BIODENS)
STANDDV(0)=STANDDT(0)/1000*(1/STDDENS)
ENDIF
/
/
/
/
/
/
/
/
CALL VEGTM(BIOMASST,BIOMREML,DAYREML,ABIOREML,BIOMREMD,
DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,PHYSDEG,
BIOMGROW,STANDDT,BIOMOUT,BIOMDTH,M,J,K,WETTYPE,
BIODENS,STDDENS,BIOMASSV,STANDDV,DAYWINE)
ENDIF
CALL HYDROT(HYDTYPE,APFLOW,WATERVOL,WATERB,POROSITY,
HEATINDX,EVAPCOEF,HI,HT,WATINPUT,PRECIP,PRECIPR,EVAPT,
OUTFLOW,WATERVEL,HTI,HII,WTRSHDAR,SCCURVE,J,K,M,
PORPEAT,NITCYCLE,SEDCYCLE,WATINT,WETTYPE,EVAP,HYDCONP,
PERCINF,PERCINFA,PERCINFT,BIODENS,STDDENS,PBIOUW,
BIOMASSV,STANDDV,PSTDUW,HRT)
IF(NITCYCLE.EQ.1) THEN
IF (M.EQ.1 .AND.J.EQ.1) THEN
OXYGEN
IF (WETTYPE.EQ.0) THEN
DOXYW(0)=DOINITW
DOXYCW(0)=DOXYW(0)/WATERVOL(0)
ENDIF
DOXYB(0)=DOINITB
DOXYCB(0)=DOXYB(0)/WATERB(0)
CARBON
IF (WETTYPE.EQ.0) THEN
DOCW(0)=DOCINITW
DOCCW(0)=DOCW(0)/WATERVOL(0)
POCW(0)=POCINITW
POCCW(0)=POCW(0)/WATERVOL(0)
TOCW(0)=DOCW(0)+POCW(0)
TOCCW(0)=TOCW(0)/WATERVOL(0)
ENDIF
DOCB(0)=DOCINITB
DOCCB(0)=DOCB(0)/WATERB(0)
POCB(0)=POCINITB
POCCB(0)=POCCB(0)/WATERB(0)
REFC(0)=REFCINIT
TOCB(0)=DOCB(0)+POCB(0)
TOCCB(0)=TOCB(0)/WATERB(0)
BIOMASS(0)=BIOMASST(0)*BIOCCONT
STANDDC(0)=STANDDT(0)*BIOCCONT
NITROGEN
IF (WETTYPE.EQ.0)THEN
DONW(0)=DONINITW
DONCW(0)=DONW(0)/WATERVOL(0)
IMMNW(0)=IMMINITW
NH4W(0)=NH4INITW
NH4CW(0)=NH4W(0)/WATERVOL(0)
NO3W(0)=NO3INITW
NO3CW(0)=NO3W(0)/WATERVOL(0)
PONW(0)=PONINITW
PONCW(0)=PONW(0)/WATERVOL(0)
TONW(0)=DONW(0)+PONW(0)
193
TONCW(0)=TONW(0)/WATERVOL(0)
ENDIF
DONB(0)=DONINITB
DONCB(0)=DONB(0)/WATERB(0)
IMMNB(0)=IMMINITB
NH4B(0)=NH4INITB
NH4CB(0)=NH4B(0)/WATERB(0)
NO3B(0)=NO3INITB
NO3CB(0)=NO3B(0)/WATERB(0)
PONB(0)=PONINITB
PONCB(0)=PONB(0)/WATERB(0)
REFN(0)=REFNINIT
TONB(0)=DONB(0)+PONB(0)
TONCB(0)=TONB(0)/WATERB(0)
C
BACTERIA
IF (WETTYPE.EQ.0) THEN
NITROSOW(0)=NITROSIW
HETEROW(0)=HETEROIW
ENDIF
NITROSOB(0)=NITROSIB
HETEROB(0)=HETEROIB
ENDIF
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
CALL CARTIME(HI,HT,HYDTYPE,BIOMASST,BIOCCONT,BODCFRAC,
BODPFRAC,LEACHR,MICROBEC,PEATCC,REFCINIT,PHYSDEGC,PERCINFT,
POCRES,POCFALL,WATINPUT,HTGROWW,HTGROWB,HTYIELDW,WETTYPE,
HTYIELDB,DONW,DONB,TONW,TONB,DOXYCW,DOXYCB,BIOMASS,DOCININC,
BODRC,DAYSDEG,OUTFLOW,BODCONC,STANDDT,DOCLEACH,MICTCNB,
DOCW,DOCB,POCW,POCB,DOCCW,DOCCB,WATERVOL,WATERB,J,K,M,
POCCW,POCCB,TOCW,TOCB,TOCCW,TOCCB,REFC,STANDDC,MICTCNW,
PEATCACW,PEATCACB,MANNC,RESTHC,POCSIZE,APFLOW,NDEATHW,POCCOUT,
NDEATHB,HTDEATHW,HTDEATHB,PEATACRW,PEATACRB,PONW,PONB,MICDTHW,
MICDTHB,MTCDOC,WATERVEL,DOCINITW,DOCINITB,POCINITW,POCINITB,
PHYSDEG,BIOMDTH,SDEADOUT,POCOUT,DOCOUT)
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
194
/
/
/
/
C
C
C
IF (SEDCYCLE.EQ.1) THEN
CALL SEDTIME (OUTFLOW,WATERVEL,HT,HI,PHOSPERB,SEDOUTT,
RESTHICK,SEDSIZE,SEDFALL,SEDPER,SEDQTYW,SEDQTYB,SEDRES,
MANNC,PHOSPER,J,K,M,SEDRC,WATINPUT,SEDCAT,APFLOW,SEDCONC,
SEDINITW,SEDINITB,SEDBSA,SEDBV,SEDSPG,WATERVOL,DECOMPR,
SEDPART,PHOSPERW,SEDDEP,RESUSP,SEDOUT,PCYCLE,PHYSDEG,PSEDDEP)
ENDIF
INITIAL DETERMINATION OF WATER VOLUME IF NOT AFFECTED BY NUTRIENT
OR SEDIMENT PROCESSES
IF (NITCYCLE.EQ.0 .AND. SEDCYCLE.EQ.0) THEN
WATERVOL(J)=WATERVOL(J-1)+WATINT(J)-EVAPT(J)-OUTFLOW(J)-PERCINFT(J)
ENDIF
/
/
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
IF (NITCYCLE.EQ.1. .OR. SEDCYCLE.EQ.1) THEN
CALL DELTAH(DELTAHT,NITCYCLE,SEDCYCLE,SEDTOTAL,SEDDELTA,
SEDCAT,SEDSPG,PEATACRB,PEATDENS,HT,NUMTMPER,J,SEDQTYB,WATERB,
PORPEAT)
WATERVOL(J)=WATERVOL(J-1)+WATINT(J)-EVAPT(J)
-OUTFLOW(J)-PERCINFT(J)+(BIOMASSV(J-1)*PBIOUW)
+(STANDDV(J-1)*PSTDUW)-(BIOMASSV(J)*PBIOUW)
-(STANDDV(J)*PSTDUW)-(LENGTH*WIDTH*(HT(J)-HT(J-1)))
ENDIF
ELSEIF (WETTYPE.EQ.1) THEN
WATERB(J)=WATERB(J-1)+WATINT(J)-EVAPT(J)-OUTFLOW(J)-PERCINFT(J)
ENDIF
DETERMINATION OF HYDRAULIC RETENTION TIME
IF (WETTYPE.EQ.0) THEN
HRT=WATERVOL(J)/OUTFLOW(J)
ELSEIF (WETTYPE.EQ.1) THEN
HRT=WATERB(J)/OUTFLOW(J)
ENDIF
IF (PCYCLE.EQ.1) THEN
IF (M.EQ.1 .AND. J.EQ.1) THEN
DTPHOSCW(0)=DTPHOSIW/WATERVOL(0)
DTPHOSCB(0)=DTPHOSIB/WATERB(0)
ENDIF
/
/
/
/
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
WRITE(15,75) M,J,WATERVOL(J),WATERB(J),WATINPUT(J),PRECIP(J),
/
WATINT(J),OUTFLOW(J),EVAPT(J),HRT, hi(j), ht(j)
75 FORMAT (T1, (I2), T4, (I3), T7, 6(F9.2, 2X),T73,(F5.1,1X),T81,3(F5.2,2x))
ELSEIF(WETTYPE.EQ.1) THEN
WRITE(15,100)M,J,WATERB(J),WATINPUT(J),PRECIP(J),WATINT(J),
/
OUTFLOW(J),EVAPT(J),hrt,hi(j),ht(j)
100 FORMAT (T1, (I1), T3, (I3), T7, 6(F9.2, 2X),T73,3(F5.2,1X))
ENDIF
IF (HI(J) .GT. HO) THEN
WRITE(*,101) M,J
101 FORMAT ('THE WETLAND SYSTEM HAS OVERFLOWN THE BANKS AT SEASON'
/
' PERIOD ', I2,', TIME PERIOD ', I3)
ENDIF
C
DETERMINATION OF NUTIENT CONCENTRATIONS
195
C
C
IF (NITCYCLE.EQ.1) THEN
CARBON
IF(WETTYPE.EQ.0) THEN
DOCCW(J)=DOCW(J)/WATERVOL(J)
POCCW(J)=POCW(J)/WATERVOL(J)
TOCW(J)=DOCW(J)+POCW(J)
TOCCW(J)=TOCW(J)/WATERVOL(J)
BOD5CW(J)=((DOCOUT(J)+POCOUT(J))/(1.4*BODCFRAC*OUTFLOW(J)))
ENDIF
DOCCB(J)=DOCB(J)/WATERB(J)
POCCB(J)=POCB(J)/WATERB(J)
TOCB(J)=DOCB(J)+POCB(J)
TOCCB(J)=TOCB(J)/WATERB(J)
BOD5CB(J)=DOCCB(J)/(1.4*BODCFRAC*OUTFLOW(J))
NITROGEN
IF (WETTYPE.EQ.0) THEN
DONCW(J)=DONW(J)/WATERVOL(J)
NH4CW(J)=NH4W(J)/WATERVOL(J)
NO3CW(J)=NO3W(J)/WATERVOL(J)
PONCW(J)=PONW(J)/WATERVOL(J)
TONCW(J)=TONW(J)/WATERVOL(J)
ENDIF
DONCB(J)=DONB(J)/WATERB(J)
NH4CB(J)=NH4B(J)/WATERB(J)
NO3CB(J)=NO3B(J)/WATERB(J)
PONCB(J)=PONB(J)/WATERB(J)
TONCB(J)=TONB(J)/WATERB(J)
OXYGEN/NO BACTERIA NECESSARY
IF (WETTYPE.EQ.0) THEN
DOXYCW(J)=DOXYW(J)/WATERVOL(J)
ENDIF
DOXYCB(J)=DOXYB(J)/WATERB(J)
ENDIF
SEDIMENT
IF (SEDCYCLE.EQ.1) THEN
SEDOUTC(J)=SEDOUTT(J)/OUTFLOW(J)
ENDIF
PHOSPHOROUS
IF (PCYCLE.EQ.1) THEN
DTPHOSCW(J)=DTPHOSW(J)/WATERVOL(J)
DTPHOSCB(J)=DTPHOSB(J)/WATERB(J)
DPHOSEC(J)=DPHOSOUT(J)/OUTFLOW(J)
TPHOSEC(J)=(DPHOSOUT(J)+PPHOSOUT(J))/OUTFLOW(J)
ENDIF
C
C
196
SET NEW SEASON PERIOD VALUES EQUAL TO END OF PREVIOUS SEASON PERIOD
IF (WETTYPE.EQ.0) THEN
WATERVOL(0)=WATERVOL(J)
IF (NITCYCLE.EQ.0 .AND. SEDCYCLE.EQ.0) THEN
WATERB(0)=WATERB(J)
ENDIF
HI(0)=HI(J)
ELSEIF (WETTYPE.EQ.1) THEN
WATERB(0)=WATERB(J)
HI(0)=HI(J)
ENDIF
197
DOXYCB(0)=DOXYCB(J)
NITROGEN
IF (WETTYPE.EQ.0) THEN
DONW(0)=DONW(J)
DONCW(0)=DONCW(J)
IMMNW(0)=IMMNW(J)
NH4W(0)=NH4W(J)
NH4CW(0)=NH4CW(J)
NO3W(0)=NO3W(J)
NO3CW(0)=NO3CW(J)
PONW(0)=PONW(J)
PONCW(0)=PONCW(J)
TONW(0)=TONW(J)
TONCW(0)=TONCW(J)
ENDIF
DONB(0)=DONB(J)
DONCB(0)=DONCB(J)
IMMNB(0)=IMMNB(J)
NH4B(0)=NH4B(J)
NH4CB(0)=NH4CB(J)
NO3B(0)=NO3B(J)
NO3CB(0)=NO3CB(J)
PONB(0)=PONB(J)
PONCB(0)=PONCB(J)
REFN(0)=REFN(J)
TONB(0)=TONB(J)
TONCB(0)=TONCB(J)
ENDIF
C
PHOSPHORUS
IF (PCYCLE.EQ.1) THEN
DTPHOSCW(0)=DTPHOSCW(J)
DTPHOSCB(0)=DTPHOSCB(J)
ENDIF
400 CONTINUE
C
C
C
SCREEN HELPERS
IF (NITCYCLE.EQ.1) THEN
WRITE(*,425)
425 FORMAT ('Nitrogen cycle is included in the simulation.')
ELSEIF (NITCYCLE.EQ.0) THEN
WRITE(*,450)
450 FORMAT ('Nitrogen cycle is not included in the simulation.')
ENDIF
IF (SEDCYCLE.EQ.1) THEN
WRITE(*,500)
500 FORMAT (/,'Sediment cycle is included in the simulation.')
ELSEIF (SEDCYCLE.EQ.0) THEN
WRITE(*,600)
600 FORMAT (/,'Sediment cycle is not included in the simulation.')
ENDIF
IF (PCYCLE.EQ.1) THEN
WRITE(*,700)
700 FORMAT (/,'Phosphorous cycle is included in the simulation.',/)
ELSEIF (PCYCLE.EQ.0) THEN
WRITE(*,800)
800 FORMAT (/,'Phosphorous cycle is not included in the '
/
'simulation.',/)
ENDIF
C
END OF SIMULATION
WRITE (*,900)
900 FORMAT ('Normal termination of simulation.',/)
C
1000 END PROGRAM WETLAND
198
BASE SUBMODEL
/
/
/
C
C
C
C
REAL LENGTH,WIDTH,HO,HB,HTI,HII,SO,WTRSHDAR,
SCCURVE,TJ,HEATIJ,HEATINDX,EVAPCOEF
HEATINDX=0
READ(2,*) WETTYPE,HYDTYPE,POINT,NITCYCLE,SEDCYCLE,PCYCLE,
EVAP,PERCINF
DESCRIPTION OF WETLAND SIZE AND INITIAL WATER LEVELS
IF (HYDTYPE.EQ.0) THEN
READ (2,*) NUMSTPER,LENGTH,WIDTH,HO,HB,HTI,HII,SO
ENDIF
IF (HYDTYPE.EQ.1) THEN
READ (2,*) NUMSTPER,LENGTH,WIDTH,HO,HB,HTI,HII,SO,WTRSHDAR,SCCURVE
ENDIF
IF (EVAP.EQ.0) THEN
HEATINDX=0
DO 75 KEEP=1,12
HEATIJ=0
READ(2,*) TJ
HEATIJ=(TJ/5)**1.514
HEATINDX=HEATINDX+HEATIJ
75 CONTINUE
EVAPCOEF=6.75E-7*(HEATINDX**3)-7.71E-5*(HEATINDX**2)
/
+1.792E-2*HEATINDX+0.49239
ELSE IF (EVAP.EQ.1) THEN
HEATINDX=0
EVAPCOEF=0
ENDIF
C
END
199
HYDROLOGIC SUBMODELS
/
/
/
/
/
C
C
SUBROUTINE HYDSTR(NUMTMPER,POINT,PORPEAT,M,WETTYPE,HYDCONP,
POROSITY,PERCINF,PERCINFA)
COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,
KHCOEF
INTEGER POINT,NUMTMPER,CHECK1,CHECK2,OUTLET,WETTYPE,PERCINF
REAL HOUT,OUTWIDTH,HOVER,ANGVNOT,HYDCONP,
FLOWOUT,TOPPUMP,AREAPIPE,CONTC,DISEFFC,PERCINFA
ALPHA,BETA,COEFK,PORPEAT,KHCOEF,POROSITY
NUMBER OF TIME PERIODS IN SEASON PERIOD
IF (M.EQ.1) THEN
IF (WETTYPE.EQ.0) THEN
READ(3,*) NUMTMPER,PORPEAT
HYDCONP=0.0
POROSITY=0
ELSEIF (WETTYPE.EQ.1)THEN
READ (3,*) NUMTMPER,POROSITY,HYDCONP
PORPEAT=0
ENDIF
IF (PERCINF.EQ.1) THEN
READ (3,*) PERCINFA
ELSEIF (PERCINF.EQ.0) THEN
PERCINFA=0.
ENDIF
HOUT=0.
OUTWIDTH=0.
HOVER=0.
ANGVNOT=0.
FLOWOUT=0.
TOPPUMP=0.
AREAPIPE=0.
CONTC=0.
DISEFFC=0.
ALPHA=0.
BEATA=0.
COEFK=0.
KHCOEF=0.
READ (3,*) OUTLET
IF (OUTLET.EQ.1) THEN
READ (3,*) HOUT,OUTWIDTH,HOVER
ELSEIF (OUTLET.EQ.2)THEN
READ(3,*) ANGVNOT,DISEFFC,KHCOEF,HOUT,HOVER
ELSEIF (OUTLET.EQ.3) THEN
READ (3,*) HOUT, AREAPIPE, CONTC
ELSEIF (OUTLET.EQ.4) THEN
READ (3,*) FLOWOUT,TOPPUMP
ELSEIF (OUTLET.EQ.5) THEN
READ (3,*) ALPHA,BETA,COEFK,HOUT
ELSEIF (OUTLET.EQ.6)THEN
READ(3,*) HOUT
END IF
ENDIF
C
C
200
ENDIF
IF (CHECK2.EQ.0) THEN
IF (OUTLET.EQ.1) THEN
READ(3,*) HOUT,OUTWIDTH,HOVER
ELSEIF (OUTLET.EQ.2) THEN
READ(3,*) ANGVNOT,DISEFFC,KHCOEF,HOUT,HOVER
ELSEIF (OUTLET.EQ.3) THEN
READ(3,*) HOUT,AREAPIPE,CONTC
ELSEIF (OUTLET.EQ.4) THEN
READ(3,*)FLOWOUT,TOPPUMP
ELSEIF (OUTLET.EQ.5) THEN
READ(3,*)ALPHA,BETA,COEFK,HOUT
ELSEIF (OUTLET.EQ.6) THEN
READ(3,*)HOUT
ENDIF
ENDIF
ENDIF
END
C
C
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
C
C
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,
KHCOEF
REAL WATERVOL(0:500),WATERB(0:500),WATINPUT(0:500),
WATINT(0:500),PRECIP(0:500),EVAPT(0:500),OUTFLOW(0:500),
HI(0:500),HT(0:500),HYDCONP,PANEVAP,POROSITY,PERCINFA,
EVAPTT(24),WATINTT(24),OUTFLOWT(24),HIT(0:24),APFLOW(0:500),
PERCINFT(0:500), PERCITT(24),BIOMASSV(0:500),STANDDV(0:500)
INTEGER OUTLET,HYDTYPE,NITCYCLE,SEDCYCLE,WETTYPE,EVAP,TIME,
PERCINF
REAL FT3TOM3,ALPHA,BETA,COEFK,AIRTEMP,DISEFFC
PORPEAT,WTRSHDAR,SCCURVE,HOUT,OUTWIDTH,HOVER,ANGVNOT,
LENGTH,WIDTH,HO,SO,HB,FLOWOUT,TOPPUMP,AREAPIPE,CONTC,
PRECIPR,KHCOEF,M3TOFT3,MTOFT,FTTOM,DAYLEN,HTI,HII,
BIODENS,STDDENS,PBIOUW,PSTDUW,HRT
INITIAL CONDITIONS OF WATER IN SYSTEM
C
C
SUBROUTINE HYDROT(HYDTYPE,APFLOW,WATERVOL,WATERB,POROSITY,
HEATINDX,EVAPCOEF,HI,HT,WATINPUT,PRECIP,PRECIPR,EVAPT,
OUTFLOW,WATERVEL,HTI,HII,WTRSHDAR,SCCURVE,J,K,M,
PORPEAT,NITCYCLE,SEDCYCLE,WATINT,WETTYPE,EVAP,HYDCONP,
PERCINF,PERCINFA,PERCINFT,BIODENS,STDDENS,PBIOUW,
BIOMASSV,STANDDV,PSTDUW,HRT)
201
WATERB(0)=LENGTH*WIDTH*(HI(0)-HB)*POROSITY
WRITE (15,*) 'OUTPUT DATA FOR HYDROLOGIC COUNTS IN WETLAND'
WRITE(15,125)
125 FORMAT
(T10, 'WATERB',T19,'WATINPUT',T32'PRECIP',T43,'WATINT',
/
T53,'OUTFLOW',T66, 'EVAPT',T76, 'HRT')
WRITE(15,150)
M,WATERB(0)
150 FORMAT (T1, (I1), T4,(' 0 '),T7,(F9.2,2X))
ENDIF
ENDIF
IF (NITCYCLE.EQ.0 .AND.SEDCYCLE.EQ.0) THEN
IF (WETTYPE.EQ.0)THEN
HT(J)=HTI
WATERB(J)=LENGTH*WIDTH*(HT(J)-HB)*PORPEAT
ENDIF
ENDIF
IF (WETTYPE.EQ.1) THEN
HT(J)=HTI
ENDIF
C
C
READ IN INPUT
IF (HYDTYPE.EQ.0) THEN
IF (EVAP.EQ.0) THEN
READ (3,*) WATINPUT(J),PRECIPR,AIRTEMP,DAYLEN
ELSEIF (EVAP.EQ.1) THEN
READ(3,*)WATINPUT(J),PRECIPR,PANEVAP
ENDIF
IF (POINT.EQ.1) THEN
READ(3,*) APFLOW(J)
ELSEIF (POINT.EQ.0) THEN
APFLOW(J)=0
ENDIF
PRECIP(J)=PRECIPR*WIDTH*LENGTH
WATINT(J)=WATINPUT(J)+PRECIP(J)+APFLOW(J)
ENDIF
IF (HYDTYPE.EQ.1) THEN
IF (EVAP.EQ.0) THEN
READ (3,*)PRECIPR,AIRTEMP,DAYLEN
ELSEIF (EVAP.EQ.1) THEN
READ(3,*)PRECIPR,PANEVAP
ENDIF
IF (POINT.EQ.1) THEN
READ(3,*) APFLOW(J)
ELSEIF (POINT.EQ.0) THEN
APFLOW(J)=0
ENDIF
C
C
C
C
PRECIP(J)=PRECIPR*WIDTH*LENGTH
STORPARM=25400./SCCURVE-254.
FLOWRATE=(((PRECIPR*1000.)-0.2*STORPARM)**2.)/((PRECIPR*1000.)
+0.8*STORPARM)
WATINPUT(J)=(FLOWRATE/1000.)*(WTRSHDAR-LENGTH*WIDTH)
WATINT(J)=WATINPUT(J)+PRECIP(J)+APFLOW(J)
ENDIF
EVAPORATION CALCULATIONS
IF (EVAP.EQ.0) THEN
IF(AIRTEMP.LE.0)THEN
EVAPT(J)=0.
ELSEIF(AIRTEMP.GT.0) THEN
EVAPT(J)=(1.6*DAYLEN)*((10.*AIRTEMP/HEATINDX)**EVAPCOEF)*
WIDTH*LENGTH/(30.*100.)
202
ENDIF
ELSEIF (EVAP.EQ.1) THEN
EVAPT(J)=PANEVAP*LENGTH*WIDTH
ENDIF
C
C
PERCOLATION/INFILTRATION CALCULATIONS
IF (WETTYPE.EQ.0) THEN
IF (PERCINF.EQ.0) THEN
PERCINFT(J)=0
ELSEIF (PERCINF.EQ.1) THEN
PERCINFT(J)=PERCINFA*LENGTH*WIDTH
ENDIF
ELSEIF (WETTYPE.EQ.1) THEN
IF (PERCINF.EQ.0) THEN
PERCINFT(J)=0
ELSEIF (PERCINF.EQ.1) THEN
PERCINFT(J)=PERCINFA*LENGTH*WIDTH
ENDIF
ENDIF
C
C
/
/
/
/
/
IF (OUTLET.EQ.1) THEN
IF (HIT(TIME-1).LE.HOUT) THEN
OUTFLOWT(TIME)=0
ELSEIF (HIT(TIME-1).GT.HOUT .AND. HIT(TIME-1).LT.HOVER) THEN
OUTFLOWT(TIME)=1.84*(OUTWIDTH-(0.2*(HIT(TIME-1)-HOUT)))*
((HIT(TIME-1)-HOUT)**1.5)*3600
ELSEIF (HIT(TIME-1).GT.HOVER) THEN
OUTFLOWT(TIME)=((1.843*(OUTWIDTH-(0.2*(HOVER-HOUT))*
((HOVER-HOUT)**1.5)))+((HIT(TIME-1)-HOVER)*WIDTH))*3600
END IF
ELSE IF (OUTLET.EQ.2) THEN
IF (HIT(TIME-1).LE.HOUT) THEN
OUTFLOWT(TIME)=0
ELSEIF (HIT(TIME-1).GT.HOUT) THEN
OUTFLOWT(TIME)=2.363*DISEFFC*(TAN(ANGVNOT/360*22/7))*
(((HIT(TIME-1)-HOUT)+KHCOEF)**(5./2.))*3600
ELSEIF (HIT(TIME-1).GT.HOVER) THEN
OUTFLOWT(TIME)=(2.363*DISEFFC*(TAN(ANGVNOT/360*22/7))*
(((HOVER-HOUT)+KHCOEF)**(5./2.))+(HIT(TIME-1)-HOVER*WIDTH))
*3600
ENDIF
ELSE IF (OUTLET.EQ.3) THEN
IF (HIT(TIME-1).LE.HOUT) THEN
OUTFLOWT(TIME)=0
ELSE IF (HIT(TIME-1).GT.HOUT) THEN
OUTFLOWT(TIME)=3600*CONTC*AREAPIPE*(SQRT(2.*9.81*
(HIT(TIME-1)-HOUT)))
ENDIF
ELSE IF (OUTLET.EQ.4) THEN
IF (HIT(TIME-1).LE.TOPPUMP) THEN
OUTFLOWT(TIME)=0
ELSEIF (HIT(TIME-1).GT.TOPPUMP) THEN
OUTFLOWT(TIME)=FLOWOUT/24
ENDIF
ELSEIF (OUTLET.EQ.5) THEN
IF (HIT(TIME-1).LE.HOUT) THEN
WATERVEL=0
OUTFLOWT(TIME)=0
203
/
/
ELSEIF(HIT(TIME-1).GT.HOUT) THEN
WATERVEL=COEFK*(((HIT(TIME-1)-HT(J-1))/2)**BETA)*
(SO**ALPHA)/((HIT(TIME-1)-HT(J-1))/2)
OUTFLOWT(TIME)=WATERVEL*WIDTH*(HIT(TIME-1)-HT(J-1))/24
ENDIF
ELSE IF (OUTLET.EQ.6) THEN
IF (HIT(TIME-1).GE.HOUT)THEN
OUTFLOWT(TIME)=MAX(WIDTH*(HIT(TIME-1)-HOUT)*HYDCONP*SO/24,
(WATERB(J-1)*HYDCONP/(LENGTH**2*POROSITY)*(WATERB(J-1)/
(LENGTH*WIDTH*POROSITY)-HOUT)))/24
ELSEIF(HIT(TIME-1).LT.HOUT) THEN
OUTFLOWT(TIME)=0.0
ENDIF
ENDIF
IF(WETTYPE.EQ.0) THEN
HIT(TIME)=HIT(TIME-1)+((WATINTT(TIME)-EVAPTT(TIME)-OUTFLOWT(TIME)
/
-PERCITT(TIME))/(LENGTH*WIDTH))
ELSEIF (WETTYPE.EQ.1) THEN
HIT(TIME)=HIT(TIME-1)+((WATINTT(TIME)-EVAPTT(TIME)-OUTFLOWT(TIME)
/
-PERCITT(TIME))/(LENGTH*WIDTH*POROSITY))
ENDIF
OUTFLOW(J)=OUTFLOW(J)+OUTFLOWT(TIME)
200 CONTINUE
/
204
VEGETATION SUBMODELS
/
/
/
/
/
C
C
SUBROUTINE VEGST(BIOINIT,BIOMREML,DAYREML,ABIOREML,BIOMREMD,
DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,STANDIN,
PEATACRW,PEATACRB,M,PEATDENS,WETTYPE,PRATEUP,PBIOUW,
BIODENS,STDDENS,PSTDUW,DAYWINE)
REAL BIOINIT,ABIOREML,DEGBIO,DAYSDEG,ABIOREMD,REFCINIT,PSTDUW,
STANDIN,PEATACRW,PEATACRB,PEATDENS,STDDENS,BIODENS,PBIOUW
INTEGER BIOMREML,DAYREML,BIOMREMD,DAYREMD,CHECK1,CHECK2,
WETTYPE,DAYWINS,DAYWINE
READ INITIAL INPUT VALUES
/
C
C
C
IF (M.EQ.1) THEN
IF (WETTYPE.EQ.0) THEN
READ (24,*) BIOINIT,STANDIN,PEATACRW,PEATACRB,PEATDENS,
PRATEUP,BIODENS,STDDENS,PBIOUW,PSTDUW
ELSEIF (WETTYPE.EQ.1) THEN
READ (24,*) BIOINIT,STANDIN,PEATACRB,PEATDENS,PRATEUP
ENDIF
READ (24,*) BIOMREML,DAYREML,ABIOREML,BIOMREMD,
DAYREMD,ABIOREMD,DAYWINS,DAYWINE,DEGBIO,DAYSDEG
ENDIF
CHECK FOR CHANGES IN PARAMETERS BETWEEN SEASON PERIODS
IF (M.GT.1) THEN
READ(24,*) c, CHECK 1,CHECK2
IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THEN
READ (24,*) PEATACRW,PEATACRB,PRATEUP,BIODENS,
STDDENS,PBIOUW,PSTDUW
ELSEIF(WETTYPE.EQ.1)THEN
READ(24,*) PEATACRB,PRATEUP
ENDIF
ENDIF
IF (CHECK2.EQ.0) THEN
READ(24,*) BIOMREML,DAYREML,ABIOREML,BIOMREMD,DAYREMD,
ABIOREMD,DAYWINS,DAYWINE,DEGBIO,DAYSDEG
ENDIF
ENDIF
END
C
C
/
/
/
/
/
/
/
SUBROUTINE VEGTM(BIOMASST,BIOMREML,DAYREML,ABIOREML,BIOMREMD,
DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,PHYSDEG,
BIOMGROW,STANDDT,BIOMOUT,BIOMDTH,M,J,K,WETTYPE,
BIODENS,STDDENS,BIOMASSV,STANDDV,DAYWINE)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
REAL BIOMASST(0:500),BIOMGROW(0:500),BIOMDTH(0:500),
SDEADOUT(0:500),PHYSDEG(0:500),STANDDT(0:500),BIOMOUT(0:500),
BIOMASSV(0:500),STANDDV(0:500)
REAL ABIOREML,DEGBIO,DAYSDEG,ABIOREMD,WINKCO,
LENGTH,WIDTH,HO,HB,SO,PRATEUP,BIODENS,STDDENS
INTEGER BIOMREML,DAYREML,BIOMREMD,DAYREMD,WINTER,WETTYPE,
DAYWINS,DAYWINE
PARAMETER (BIODEGR=.012616904)
READ(24,*) BIOMGRR
IF (DAYWINS.EQ.0) THEN
WINTER=0
GO TO 10
ENDIF
IF (J .LT. DAYWINS) THEN
WINTER=0
ELSE IF ((J .GE. DAYWINS) .AND. (J .LE. DAYWINE)) THEN
205
WINTER=1
ELSE IF (J .GT. DAYWINE) THEN
WINTER=0
ENDIF
10
IF (WINTER.EQ.0) THEN
BIOMGROW(J)=LENGTH*WIDTH*BIOMGRR
BIOMDTH(J)=0.0
ENDIF
IF (WINTER.EQ.1) THEN
C
C
C
C
IF (BIOMREML.EQ.0) THEN
BIOMOUT(J)=0.0
ELSE IF (BIOMREML.EQ.1) THEN
IF (J.EQ.DAYREML) THEN
BIOMOUT(J)=ABIOREML
ELSE
BIOMOUT(J)=0.0
ENDIF
ENDIF
IF (BIOMREMD.EQ.0) THEN
SDEADOUT(J)=0.0
ELSE IF (BIOMREMD.EQ.1) THEN
IF (J.EQ.DAYREMD) THEN
SDEADOUT(J)=ABIOREMD
ELSE
SDEADOUT(J)=0.0
ENDIF
ENDIF
PHYSDEG(J)=BIODEGR*STANDDT(J-1)
C
C
C
206
CARBON SUBMODELS
/
/
/
/
/
/
C
C
SUBROUTINE CARSTR(DOCINITW,DOCINITB,POCINITW,POCINITB,
POCFALL,POCRES,BIOCCONT,BODCFRAC,MTCDOC,POINT,PERCINF,
BODPFRAC,LEACHR,MICROBEC,PEATCC,BODCONC,HYDTYPE,BODRC,
MANNC,POCSIZE,RESTHC,M,REFCINIT,WETTYPE,POCCOUT,DOCININC)
REAL DOCINITW,DOCINITB,POCINITW,POCINITB,REFCINIT,
BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,MTCDOC,
PEATCC,POCFALL,POCRES,POCCOUT,BODRC
BODCONC,MANNC,RESTHC,POCSIZE,DOCININC
INTEGER CHECK1,CHECK2,HYDTYPE,CHECK3,WETTYPE,POINT,PERCINF
READ IN INITIAL DATA
/
/
/
IF (M.EQ.1) THEN
IF (WETTYPE.EQ.0) THEN
READ (12,*) REFCINIT,DOCINITW,DOCINITB,POCINITW,POCINITB,
BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,PEATCC,
POCFALL,POCRES,MANNC,RESTHC,POCSIZE,MTCDOC,POCCOUT
ELSEIF (WETTYPE.EQ.1) THEN
READ (12,*) REFCINIT,DOCINITB,POCINITB,BIOCCONT,
BODCFRAC,BODPFRAC,LEACHR,MICROBEC,PEATCC
ENDIF
IF (POINT.EQ.0) THEN
BODCONC=0.
ELSEIF (POINT.EQ.1) THEN
READ (12,*) BODCONC
ENDIF
IF (PERCINF.EQ.0) THEN
DOCININC=0.0
ELSEIF (PERCINF.EQ.1) THEN
READ (12,*) DOCININC
ENDIF
IF (HYDTYPE.EQ.0) THEN
BODRC=0.0
ELSEIF (HYDTYPE.EQ.1) THEN
READ(12,*) BODRC
ENDIF
ENDIF
C
C
/
/
/
IF (M.GT.1) THEN
READ (12,*) C, CHECK1,CHECK2
IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0)THEN
READ (12,*) BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,
PEATCC,POCFALL,POCRES,MANNC,RESTHC,POCSIZE,
MTCDOC,POCCOUT
ELSEIF(WETTYPE.EQ.1) THEN
READ (12,*) BIOCCONT,BODCFRAC,BODPFRAC,
LEACHR,MICROBEC,PEATCC
ENDIF
ENDIF
IF (CHECK2.EQ.0) THEN
IF (POINT.EQ.1) THEN
READ(12,*) BODCONC
ENDIF
IF (PERCINF.EQ.1) THEN
READ(12,*) DOCININC
ENDIF
IF(HYDTYPE.EQ.1) THEN
READ(12,*) BODRC
ENDIF
ENDIF
207
ENDIF
END
C
/
/
/
/
/
/
/
/
/
/
SUBROUTINE CARTIME(HI,HT,HYDTYPE,BIOMASST,BIOCCONT,BODCFRAC,
BODPFRAC,LEACHR,MICROBEC,PEATCC,REFCINIT,PHYSDEGC,PERCINFT,
POCRES,POCFALL,WATINPUT,HTGROWW,HTGROWB,HTYIELDW,WETTYPE,
HTYIELDB,DONW,DONB,TONW,TONB,DOXYCW,DOXYCB,BIOMASS,DOCININC,
BODRC,DAYSDEG,OUTFLOW,BODCONC,STANDDT,DOCLEACH,MICTCNB,
DOCW,DOCB,POCW,POCB,DOCCW,DOCCB,WATERVOL,WATERB,J,K,M,
POCCW,POCCB,TOCW,TOCB,TOCCW,TOCCB,REFC,STANDDC,MICTCNW,
PEATCACW,PEATCACB,MANNC,RESTHC,POCSIZE,APFLOW,NDEATHW,POCCOUT,
NDEATHB,HTDEATHW,HTDEATHB,PEATACRW,PEATACRB,PONW,PONB,MICDTHW,
MICDTHB,MTCDOC,WATERVEL,DOCINITW,DOCINITB,POCINITW,POCINITB,
PHYSDEG,BIOMDTH,SDEADOUT,POCOUT,DOCOUT)
/
/
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,
KHCOEF
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
C /
/
/
/
/
REAL HTYIELDB(0:500),HTYIELDW(0:500),WATERVOL(0:500),
WATERB(0:500),DOCMINIB(0:500),TOCCB(0:500),DOCCB(0:500),
POCMINIB(0:500),PEATCACB(0:500),PEATCACW(0:500),
DOCMINIW(0:500),BIOMASST(0:500),BIOMASS(0:500),STANDDT(0:500),
POCMINIW(0:500),MICTCNW(0:500),MICTCNB(0:500),DOXYCB(0:500),
POCOUT(0:500),DOCOUT(0:500),POCRE(0:500),DOXYCW(0:500),
DOCW(0:500),DOCCW(0:500),DOCB(0:500),PHYSDEGC(0:500),
POCW(0:500),POCCW(0:500),POCB(0:500),POCCB(0:500),REFC(0:500),
STANDDC(0:500),TOCW(0:500),TOCCW(0:500),TOCB(0:500),
BIOMGROW(0:500),BIOMDTH(0:500), BIOMOUT(0:500),PONB(0:500),
SOBODIN(0:500),PBODIN(0:500),MICRODW(0:500),PERCINFT(0:500),
MICRODB(0:500),DOCMT(0:500),POCSET(0:500),DOCLEACH(0:500),
HI(0:500),HT(0:500),SDEADOUT(0:500),PONW(0:500),
WATINPUT(0:500),OUTFLOW(0:500),TONW(0:500),TONB(0:500),
HTGROWB(0:500),HTGROWW(0:500),DONW(0:500),DONB(0:500),
NDEATHW(0:500),NDEATHB(0:500),HTDEATHW(0:500),HTDEATHB(0:500),
DOCPERC(0:500),APFLOW(0:500), PHYSDEG(0:500)
INTEGER HYDTYPE,WETTYPE,OUTLET
BIOMREML,DAYREML,WINTER,BIOMREMD,DAYREMD
REAL DOCINITW,DOCINITB,POCINITW,POCINITB,REFCINIT,POCRES,
BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,WATERVEL,
PEATCC,RESTHC,POCSIZE,LENGTH,WIDTH,HO,HB,SO,MANNC,
ABIOREML,DAYWIN,DEGBIO,POCFALL,BODCONC,ABIOREMD,RESUSPV,
DAYSDEG,MTCDOC,BODRC,PEATACRB,PEATACRW,POCCOUT
C
C
C
C
208
C
C
DOC RELATIONS
/
/
IF (HYDTYPE.EQ.0) THEN
SOBODIN(J)=1.4*(1-BODPFRAC)*BODCFRAC*(BODINFCO*(WATINPUT(J)+
APFLOW(J)*(BODCONC)))
ELSEIF (HYDTYPE.EQ.1) THEN
SOBODIN(J)=1.4*(1-BODPFRAC)*BODCFRAC*(BODRC*WATINPUT(J)+
APFLOW(J)*BODCONC)
ENDIF
DOCLEACH(J)=LENGTH*WIDTH*LEACHR
IF(WETTYPE.EQ.0) THEN
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
DOCOUT(J)=DOCW(J-1)*OUTFLOW(J)/(WATERVOL(J-1))
ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THEN
DOCOUT(J)=.75*DOCW(J-1)*OUTFLOW(J)/WATERVOL(J-1)
ENDIF
ELSEIF (WETTYPE.EQ.1) THEN
DOCOUT(J)=DOCB(J-1)*OUTFLOW(J)/(WATERB(J-1))
ENDIF
IF (WETTYPE.EQ.0) THEN
IF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
DOCMINIW(J)=DONW(J-1)/TONW(J-1)*HTGROWW(J)/HTYIELDW(J)
ELSE
DOCMINIW(J)=DOCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)
ENDIF
ENDIF
IF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
DOCMINIB(J)=DONB(J-1)/TONB(J-1)*HTGROWB(J)/HTYIELDB(J)
ELSE
DOCMINIB(J)=DOCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
ENDIF
IF (WETTYPE.EQ.0) THEN
DOCMT(J)=864*LENGTH*WIDTH*MTCDOC*(DOCCB(J-1)-DOCCW(J-1))
ENDIF
IF (PERCINFT(J).GT.0.) THEN
DOCPERC(J)=DOCB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
209
DOCPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THEN
DOCPERC(J)=PERCINFT(J)*DOCININC
ENDIF
C
C
/
C
C
IF (WETTYPE.EQ.0) THEN
DOCW(J)=DOCW(J-1)+SOBODIN(J)-DOCOUT(J)-DOCMINIW(J)+DOCMT(J)
DOCB(J)=DOCB(J-1)-DOCMINIB(J)-DOCMT(J)+DOCLEACH(J)-DOCPERC(J)
ELSEIF (WETTYPE.EQ.1) THEN
DOCB(J)=DOCB(J-1)+SOBODIN(J)+DOCLEACH(J)-DOCOUT(J)
-DOCMINIB(J)-DOCPERC(J)
ENDIF
POC RELATIONS
/
/
IF (HYDTYPE.EQ.0) THEN
PBODIN(J)=1.4*(BODPFRAC)*BODCFRAC*(BODINFCO*WATINPUT(J)+
APFLOW(J)*BODCONC)
ELSEIF (HYDTYPE.EQ.1) THEN
PBODIN(J)=1.4*(BODPFRAC)*BODCFRAC*(BODRC*WATINPUT(J)+
APFLOW(J)*BODCONC)
ENDIF
PHYSDEGC(J)=BIODEGR*STANDDC(J-1)
IF(WETTYPE.EQ.0) THEN
MICRODW(J)=(NDEATHW(J)+HTDEATHW(J))*MICROBEC
PEATCACW(J)=PEATACRW*PEATCC
ENDIF
MICRODB(J)=(NDEATHB(J)+HTDEATHB(J))*MICROBEC
PEATCACB(J)=PEATACRB*PEATCC
IF (WETTYPE.EQ.0) THEN
IF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
POCMINIW(J)=PONW(J-1)/TONW(J-1)*HTGROWW(J)/HTYIELDW(J)
ELSE
POCMINIW(J)=POCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)
ENDIF
ENDIF
IF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
POCMINIB(J)=PONB(J-1)/TONB(J-1)*HTGROWB(J)/HTYIELDB(J)
ELSE
POCMINIB(J)=POCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
ENDIF
IF (WETTYPE.EQ.0) THEN
IF (POCFALL.LT.(HI(J-1)-HT(J-1))) THEN
POCSET(J)=POCW(J-1)*(POCFALL)/(HI(J-1)-HT(J-1))
ELSE IF (POCFALL.GE.(HI(J-1)-HT(J-1)))THEN
POCSET(J)=POCW(J-1)
ENDIF
ENDIF
IF(WETTYPE.EQ.0) THEN
RESUSPV=7.2*(((POCFALL/86400)**1/3)*
((HI(J-1)-HT(J-1))**(1./6.))/(MANNC*((POCSIZE**(2./3.)))))
IF (WATERVEL.GE.RESUSPV) THEN
POCRE(J)=POCB(J-1)*POCRES*RESTHC/(HT(J-1)-HB)
ELSE
POCRE(J)=0
ENDIF
ENDIF
IF (WETTYPE.EQ.0) THEN
IF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
OUTLET.EQ.5) THEN
IF (POCFALL .GT. (HI(J-1)-HT(J-1))) THEN
POCOUT(J)=0.0
210
/
/
/
C
C
C
IF (WETTYPE.EQ.0) THEN
POCW(J)=POCW(J-1)+PBODIN(J)+MICRODW(J)-PEATCACW(J)
-POCMINIW(J)-POCSET(J)+POCRE(J)-POCOUT(J)+PHYSDEGC(J)
POCB(J)=POCB(J-1)+MICRODB(J)-PEATCACB(J)-POCMINIB(J)
+POCSET(J)-POCRE(J)
ELSEIF (WETTYPE.EQ.1) THEN
POCB(J)=POCB(J-1)+PHYSDEGC(J)+PBODIN(J)+MICRODB(J)-PEATCACB(J)
-POCMINIB(J)
ENDIF
CARBON BIOMASS/REFRACTORY CARBON/STANDING DEAD CARBON
MASS BALANCE
BIOMASS(J)=BIOMASS(J-1)+((BIOMASST(J)-BIOMASST(J-1))*BIOCCONT)
IF (WETTYPE.EQ.0) THEN
REFC(J)=REFC(J-1)+PEATCACW(J)+PEATCACB(J)
ELSEIF (WETTYPE.EQ.1) THEN
REFC(J)=REFC(J-1)+PEATCACB(J)
ENDIF
STANDDC(J)=STANDDC(J-1)+((BIOMDTH(J)-PHYSDEG(J)-SDEADOUT(J))
*BIOCCONT)-DOCLEACH(J)
211
BACTERIA SUBMODELS
/
/
/
/
/
/
C
C
SUBROUTINE BACSTR(NITROSIW,NITROSIB,HETEROIW,HETEROIB,M,NDRATEW,
NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,
AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,
HTDOHSCB,HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,WETTYPE)
REAL NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,
NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,HETEROIW,HETEROIB,
AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,
HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,AEMAXGRW
INTEGER CHECK1,CHECK2,WETTYPE,M
READ IN INITIAL VALUES
/
/
/
C
C
IF (M.EQ.1) THEN
IF (WETTYPE.EQ.0) THEN
READ (6,*) NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,
NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB
READ(6,*)HETEROIW,HETEROIB,AEMAXGRW,AEMAXGRB,ANMAXGRW,
ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW,HNO3HSCB,
HORGHSCW,HORGHSCB
ELSEIF (WETTYPE.EQ.1) THEN
READ(6,*) NITROSIB,NDRATEB,NDOHSATB,NMAXGRB,NNH4HSCB
READ(6,*) HETEROIB,AEMAXGRB,ANMAXGRB,HTDRB,
HTDOHSCB,HNO3HSCB,HORGHSCB
ENDIF
ENDIF
CHECK TO SEE VALUE CHANGES BETWEEN SEASON PERIODS
IF (M.GT.1) THEN
READ (6,*) C, CHECK1,CHECK2
IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THEN
READ(6,*)NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,
NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB
ELSEIF(WETTYPE.EQ.1) THEN
READ(6,*)NDRATEB,NDOHSATB,NMAXGRB,NNH4HSCB
ENDIF
ENDIF
IF (CHECK2.EQ.0) THEN
IF (WETTYPE.EQ.0) THEN
READ (6,*) AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,
HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,
HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB
ELSEIF (WETTYPE.EQ.1) THEN
READ (6,*) AEMAXGRB,ANMAXGRB,HTDRB,
HTDOHSCB,HNO3HSCB,HORGHSCB
ENDIF
ENDIF
/
/
/
ENDIF
END
C
C
/
/
/
/
/
/
/
/
/
/
REAL HETEROW(0:500),NITROSOW(0:500),NITROSOB(0:500),
ANFRACW(0:500),ANFRACB(0:500),HTDEATHB(0:500),DOXYCW(0:500),
NSGROWW(0:500),NSGROWB(0:500),AEROHTGW(0:500),ANHTGW(0:500),
HTGROWW(0:500),HTDEATHW(0:500),NDEATHW(0:500),NDEATHB(0:500),
212
C
C
/
/
/
/
HETEROB(0:500),AEROHTGB(0:500),ANHTGB(0:500),HTGROWB(0:500),
DOXYCB(0:500),DOXYW(0:500),DOXYB(0:500),NH4CW(0:500),
NH4CB(0:500),TOCCW(0:500),TOCCB(0:500),NO3CW(0:500),
NO3W(0:500),NO3B(0:500),NO3CB(0:500)
/
/
/
/
REAL NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW,
NMAXGRB,NNH4HSCW,NNH4HSCB,HETEROIW,HETEROIB,AEMAXGRW,AEMAXGRB,
ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW,
HNO3HSCB,HORGHSCW,HORGHSCB,WATTEMPW,WATTEMPB,NSTEMPFW,
HTTEMPFW,NSTEMPFB,HTTEMPFB
INTEGER WETTYPE,J,K,M
READ IN DAILY INPUT VALUES
IF (WETTYPE.EQ.0) THEN
READ(6,*) WATTEMPW,WATTEMPB
ELSEIF (WETTYPE.EQ.1) THEN
READ(6,*) WATTEMPB
ENDIF
C
C
C
C
213
/
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
NSGROWW(J)=(NMAXGRW*((NH4CW(J-1))/(NNH4HSCW+
(NH4CW(J-1))))*((DOXYCW(J-1))/(NDOHSATW+(DOXYCW(J-1))))
*NITROSOW(J-1)*NSTEMPFW)
NDEATHW(J)=NDRATEW*NITROSOW(J-1)
ENDIF
NSGROWB(J)=(NMAXGRB*((NH4CB(J-1))/
(NNH4HSCB+(NH4CB(J-1))))*((DOXYCB(J-1))/(NDOHSATB+
(DOXYCB(J-1))))*NITROSOB(J-1)*NSTEMPFB)
NDEATHB(J)=NDRATEB*NITROSOB(J-1)
NITROSOMONAS MASS BALANCE
IF (WETTYPE.EQ.0) THEN
NITROSOW(J)=NITROSOW(J-1)+NSGROWW(J)-NDEATHW(J)
ENDIF
NITROSOB(J)=NITROSOB(J-1)+NSGROWB(J)-NDEATHB(J)
C
C
/
/
IF (WETTYPE.EQ.0) THEN
AEROHTGW(J)=(AEMAXGRW*(TOCCW(J-1))/
((TOCCW(J-1))+HORGHSCW)*HTTEMPFW*(DOXYCW(J-1))
/((DOXYCW(J-1))+HTDOHSCW)*(1-ANFRACW(J))*HETEROW(J-1))
/
/
/
ANHTGW(J)=((ANMAXGRW*TOCCW(J-1))/
((TOCCW(J-1))+HORGHSCW)*HTTEMPFW*HTDOHSCW/(HTDOHSCW+
(DOXYCW(J-1)))*(NO3CW(J-1))/((NO3CW(J-1))
+HNO3HSCW)*ANFRACW(J)*HETEROW(J-1))
HTGROWW(J)=AEROHTGW(J)+ANHTGW(J)
HTDEATHW(J)=HTDRW*HETEROW(J-1)
ENDIF
/
/
AEROHTGB(J)=(AEMAXGRB*(TOCCB(J-1))/
((TOCCB(J-1))+HORGHSCB)*HTTEMPFB*(DOXYCB(J-1))/
((DOXYCB(J-1))+HTDOHSCB)*(1-ANFRACB(J))*HETEROB(J-1))
/
/
/
ANHTGB(J)=((ANMAXGRB*TOCCB(J-1))/
((TOCCB(J-1))+HORGHSCB)*HTTEMPFB*HTDOHSCB/(HTDOHSCB+
(DOXYCB(J-1)))*(NO3CB(J-1))/((NO3CB(J-1))
+HNO3HSCB)*ANFRACB(J)*HETEROB(J-1))
HTGROWB(J)=AEROHTGB(J)+ANHTGB(J)
HTDEATHB(J)=HTDRB*HETEROB(J-1)
C
C
C
C
OUTPUT DATA
214
NITROGEN SUBMODELS
/
/
/
/
/
/
/
/
/
/
/
/
/
/
C
C
/
/
/
/
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
READ (8,*) ATDEP,NITFIX,VOLAT,
DONINITW,DONINITB,IMMINITW,IMMINITB,NH4INITW,NH4INITB,
NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,BIOMCN,
BIOMPN,HTNO3YW,HTNO3YB,MICRONC,NSYIELDW,NSYIELDB,
ONPARTF,PEATNC,PONRES,PONFALL,MTCDON,MTCNH4,
MTCNO3,PONSIZE,RESTHN,PONCOUT
ELSEIF(WETTYPE.EQ.1) THEN
READ(8,*) ATDEP,NITFIX,VOLAT,
DONINITB,IMMINITB,NH4INITB,NO3INITB,PONINITB,REFNINIT,
BIOMCN,BIOMPN,HTNO3YB,MICRONC,NSYIELDB,ONPARTF,PEATNC
ENDIF
NITROGEN FIXATION
IF (NITFIX.EQ.0) THEN
NITFRATE=0.0
ELSEIF (NITFIX.EQ.1) THEN
READ(8,*) NITFRATE
ENDIF
C
C
VOLATILIZATION
IF (VOLAT.EQ.0) THEN
VOLATR=0.0
ELSE
READ(8,*) VOLATR
ENDIF
C
C
ATMOSPHERIC DEPOSITION
IF (ATDEP.EQ.0) THEN
DADDON=0.0
WADDON=0.0
DADPON=0.0
WADPON=0.0
DADNH4=0.0
WADNH4=0.0
DADNO3=0.0
WADN03=0.0
ELSE
READ(8,*)DADDON,WADDON,DADPON,WADPON,
DADNH4,WADNH4,DADNO3,WADNO3
ENDIF
215
C
C
POINT FLOW
IF (POINT.EQ.1) THEN
READ (8,*) DONCONC,PONCONC,NO3CONC,NH4CONC
ELSE
DONCONC=0.0
PONCONC=0.0
NO3CONC=0.0
NH4CONC=0.0
ENDIF
C
C
PERCOLATION/INFILITRATION
IF (PERCINF.EQ.0) THEN
DONININC=0.0
NH4ININC=0.0
NO3ININC=0.0
ELSEIF (PERCINF.EQ.1) THEN
READ (8,*) DONININC,NH4ININC,NO3ININC
ENDIF
C
C
INPUT TYPE
IF (HYDTYPE.EQ.0) THEN
ORGNRC=0.0
NH4RC=0.0
NO3RC=0.0
ELSEIF (HYDTYPE.EQ.1) THEN
READ (8,*) ORGNRC,NH4RC,NO3RC
ENDIF
ENDIF
C
C
/
/
/
IF (M.GT.1) THEN
READ (8,*)c,CHECK1,CHECK2,CHECK3,CHECK4,CHECK5,CHECK6,CHECK7
IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THEN
READ(8,*)BIOMCN,BIOMPN,HTNO3YW,HTNO3YB,MICRONC,NSYIELDW,
NSYIELDB,ONPARTF,PEATNC,PONRES,PONFALL,MTCDON,
MTCNH4,MTCNO3,PONSIZE,RESTHN,PONCOUT
ELSEIF(WETTYPE.EQ.1)THEN
READ(8,*)BIOMCN,BIOMPN,HTNO3YB,MICRONC,
NSYIELDB,ONPARTF,PEATNC
ENDIF
ENDIF
IF (CHECK2.EQ.0) THEN
READ(8,*) NITFRATE
ENDIF
IF (CHECK3.EQ.0) THEN
READ(8,*) VOLATR
ENDIF
IF (CHECK4.EQ.0) THEN
READ (8,*) DADDON,WADDON,DADPON,WADPON,
DADNH4,WADNH4,DADNO3,WADNO3
ENDIF
IF (CHECK5.EQ.0) THEN
READ (8,*) DONCONC,PONCONC,NO3CONC,NH4CONC
ENDIF
IF (CHECK6.EQ.0) THEN
READ(8,*) DONININC,NH4ININC,NO3ININC
ENDIF
IF (CHECK7.EQ.0) THEN
READ(8,*) ORGNRC,NH4RC,NO3RC
ENDIF
ENDIF
END
216
C
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
REAL OUTFLOW(0:500),HTGROWW(0:500),HTGROWB(0:500),
IMMNB(0:500),NH4CW(0:500),PONCB(0:500),WATERVOL(0:500),
WATERB(0:500),HI(0:500),HT(0:500),DONW(0:500),DONB(0:500),
IMMNW(0:500),NH4B(0:500),NO3W(0:500),NO3B(0:500),PONW(0:500),
PONB(0:500),TONCW(0:500),TONCB(0:500),DONOUT(0:500),
REFN(0:500),TONW(0:500),TONB(0:500),DONIN(0:500),
HNH4IMMW(0:500),HNH4IMMB(0:500),DONIMMW(0:500),DONIMMB(0:500),
DONMINW(0:500),DONMINB(0:500),FIXNIT(0:500),DONAT(0:500),
PRECIP(0:500),WATINPUT(0:500),DONMT(0:500),NH4W(0:500),
NLEACH(0:500),NITUPB(0:500),AMMUPB(0:500),DONCW(0:500),
PONIMMW(0:500),PONIMMB(0:500),DEATHB(0:500),IMMNOUT(0:500),
IMMNOUTD(0:500),NH4IN(0:500),NH4OUT(0:500),NITRIFW(0:500),
NITRIFB(0:500),NH4MT(0:500),NH4AT(0:500),VOLATIZ(0:500),
NO3IN(0:500),NO3OUT(0:500),DENITW(0:500),DENITB(0:500),
NO3AT(0:500),NO3MT(0:500),DEATHW(0:500),PONIN(0:500),
PEATNACW(0:500),PEATNACB(0:500),PONMINW(0:500),PONMINB(0:500),
PONAT(0:500),PONSET(0:500),PONRE(0:500),DONCB(0:500),
NH4CB(0:500),NO3CW(0:500),NO3CB(0:500),PONCW(0:500),
TOCW(0:500),MICTCNW(0:500),DOCW(0:500),NITUPW(0:500),
TOCB(0:500),HTDEATHW(0:500),NO3(0:500),ANHTGB(0:500),
MICTCNB(0:500),DOCB(0:500),HTYIELDW(0:500),HTYIELDB(0:500),
BIOMGROW(0:500),NSGROWB(0:500),POCW(0:500),POCB(0:500),
NSDEATHW(0:500),HTDEATHB(0:500),DONIMM(0:500),PON(0:500),
NSGROWW(0:500),NH4ATDEP(0:500),ANHTGW(0:500),DOCLEACH(0:500),
PEATACW(0:500),PEATACB(0:500),PHYSDEGC(0:500),
NSDEATHB(0:500),PONOUT(0:500),NH4EC(0:500),NO3EC(0:500),
DONEC(0:500),PONEC(0:500),AMMUPW(0:500),APFLOW(0:500),
PERCINFT(0:500),DONPERC(0:500),NH4PERC(0:500),NO3PERC(0:500)
INTEGER NITFIX,ATDEP,VOLAT,DAYREMD,DAYREML,BIOMREML,BIOMREMD,
HYDTYPE,WETTYPE,OUTLET
REAL MTCNH4,NSYIELDW,MTCNO3,NSYIELDB,WATERVEL,NH4RC,
BIOMCN,HTNO3YW,HTNO3YB,MICRONC,ONPARTF,PEATNC,BIOMPN
NITFRATE,VOLATR,PONRES,NO3RC,ORGNRC,DONINITW,DONINITB,
PONFALL,DONCONC,PONCONC,NO3CONC,NH4CONC,REFNINIT,
ABIOREMD,ABIOREML,MTCDON,NO3INITW,NO3INITB,PONINITW,PONINITB,
RESUSPV,MANNC,PONSIZE,RESTHN,DADNH4,WADNH4,DADNO3,WADNO3,
DADDON,WADDON,DADPON,WADPON,NH4INITB,NH4INITW,
LENGTH,WIDTH,HO,HB,SO,ORGNINC,NH4INC,NO3INC,
HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,KHCOEF,PONCOUT,
TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,PRATEUP
/
/
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,S0
COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,
KHCOEF
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
C
C
217
READ(8,*)ORGNINC,NH4INC,NO3INC,PH
ELSEIF (VOLAT.EQ.1 .AND. HYDTYPE.EQ.1) THEN
READ (8,*) PH
ENDIF
NLEACH(J)=DOCLEACH(J)/BIOMCN
C
C
DON PROCESSES
IF (WETTYPE.EQ.0) THEN
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
DONOUT(J)=DONW(J-1)*OUTFLOW(J)/(WATERVOL(J-1))
ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THEN
DONOUT(J)=DONW(J-1)*.75
ENDIF
ELSEIF (WETTYPE.EQ.1) THEN
DONOUT(J)=DONB(J-1)*OUTFLOW(J)/WATERB(J-1)
ENDIF
DONEC(J)=DONOUT(J)/OUTFLOW(J)
IF (HYDTYPE.EQ.0) THEN
DONIN(J)=(1-ONPARTF)*(WATINPUT(J)*ORGNINC+APFLOW(J)*DONCONC)
ELSEIF (HYDTYPE.EQ.1) THEN
DONIN(J)=(1-ONPARTF)*(WATINPUT(J)*ORGNRC+APFLOW(J)*DONCONC)
ENDIF
IF (WETTYPE.EQ.0)THEN
HNH4IMMW(J)=MICRONC*HTGROWW(J)*NH4W(J-1)/(TONW(J-1)+NH4W(J-1))
ENDIF
HNH4IMMB(J)=MICRONC*HTGROWB(J)*NH4B(J-1)/(TONB(J-1)+NH4B(J-1))
IF (WETTYPE.EQ.0) THEN
IF ((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
DONIMMW(J)=(DONW(J-1)/(TONW(J-1)+NH4W(J-1))*MICRONC*HTGROWW(J))
ELSE
DONIMMW(J)=(DOCW(J-1)/TOCW(J-1))*(MICRONC*HTGROWW(J)-HNH4IMMW(J))
ENDIF
ENDIF
IF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
DONIMMB(J)=(DONB(J-1)/(TONB(J-1)+NH4B(J-1))*MICRONC*HTGROWB(J))
ELSE
DONIMMB(J)=(DOCB(J-1)/TOCB(J-1))*(MICRONC*HTGROWB(J)-HNH4IMMB(J))
ENDIF
IF (WETTYPE.EQ.0) THEN
IF (((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) .OR. (DOCW(J-1).LT.0.1)) THEN
DONMINW(J)=0.0
ELSE
DONMINW(J)=(DOCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)
*DONW(J-1)/DOCW(J-1)-DONIMMW(J))
IF (DONMINW(J).GE.0) THEN
DONMINW(J)=DONMINW(J)
ELSE
DONMINW(J)=0
ENDIF
ENDIF
ENDIF
IF (((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) .OR. (DOCB(J-1).LT.0.1)) THEN
DONMINB(J)=0.0
ELSE
DONMINB(J)=(DOCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
*DONB(J-1)/DOCB(J-1)-DONIMMB(J))
IF(DONMINB(J).LE. 0) THEN
DONMINB(J)=0.0
ENDIF
ENDIF
IF (NITFIX.EQ.1) THEN
FIXNIT(J)=NITFRATE*LENGTH*WIDTH
218
ENDIF
IF (ATDEP.EQ.1) THEN
IF (PRECIP(J).EQ.0.) THEN
DONAT(J)=DADDON*LENGTH*WIDTH
ELSE
DONAT(J)=PRECIP(J)*WADDON
ENDIF
ENDIF
IF (WETTYPE.EQ.0) THEN
DONMT(J)=864*LENGTH*WIDTH*MTCDON*(DONCB(J-1)-DONCW(J-1))
ENDIF
IF (PERCINFT(J).GT.0.) THEN
DONPERC(J)=DONB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
DONPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THEN
DONPERC(J)=PERCINFT(J)*DONININC
ENDIF
C
C
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
DONW(J)=DONW(J-1)+DONIN(J)-DONOUT(J)-DONIMMW(J)
-DONMINW(J)+FIXNIT(J)+DONAT(J)+DONMT(J)
DONB(J)=DONB(J-1)-DONIMMB(J)-DONMINB(J)-DONMT(J)-DONPERC(J)
+NLEACH(J)
ELSEIF (WETTYPE.EQ.1) THEN
DONB(J)=DONB(J-1)+DONIN(J)+NLEACH(J)-DONOUT(J)-DONIMMB(J)DONMINB(J)+FIXNIT(J)+DONAT(J)-DONPERC(J)
ENDIF
NO3 PROCESSES
IF (WETTYPE.EQ.0) THEN
IF (NO3B(J-1).GT.(PRATEUP*BIOMGROW(J)/BIOMPN)) THEN
NITUPB(J)=PRATEUP*BIOMGROW(J)/BIOMPN
ELSE
NITUPB(J)=NO3B(J-1)
ENDIF
IF (NO3W(J-1) .GT. ((1-PRATEUP)*BIOMGROW(J)/BIOMPN)) THEN
NITUPW(J)=(BIOMGROW(J)*(1-PRATEUP))/BIOMPN
ELSE
NITUPW(J)=NO3W(J-1)
ENDIF
ENDIF
IF (WETTYPE.EQ.1) THEN
IF (NO3B(J-1).GT.(BIOMGROW(J)/BIOMPN)) THEN
NITUPB(J)=BIOMGROW(J)/BIOMPN
ELSE
NITUPB(J)=NO3B(J-1)
ENDIF
ENDIF
C
C
AMMONIUM PROCESSES
/
/
/
IF (WETTYPE.EQ.0) THEN
AMMUPW(J)=NSGROWW(J)*MICRONC+(BIOMGROW(J)/BIOMPN*(1-PRATEUP)
-(NITUPW(J)))+HNH4IMMW(J)
AMMUPB(J)=NSGROWB(J)*MICRONC+((BIOMGROW(J)*PRATEUP/BIOMPN)
-(NITUPB(J)))+HNH4IMMB(J)
ELSEIF (WETTYPE.EQ.1) THEN
AMMUPB(J)=NSGROWB(J)*MICRONC+(BIOMGROW(J)/BIOMPN
-(NITUPB(J)))+HNH4IMMB(J)
ENDIF
219
IF (WETTYPE.EQ.0) THEN
IF ((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
PONIMMW(J)=PONW(J-1)/(TONW(J-1)+NH4W(J-1))*MICRONC*HTGROWW(J)
ELSE
PONIMMW(J)=POCW(J-1)/TOCW(J-1)*(MICRONC*HTGROWW(J)(HNH4IMMW(J)))
ENDIF
DEATHW(J)=.2*PHYSDEGC(J)/BIOMCN+MICRONC*(HTDEATHW(J)+NSDEATHW(J))
ENDIF
/
C
C
C
C
C
C
C
IF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
PONIMMB(J)=PONB(J-1)/(TONB(J-1)+NH4B(J-1))*MICRONC*HTGROWB(J)
ELSE
PONIMMB(J)=POCB(J-1)/(TOCB(J-1))*(MICRONC*HTGROWB(J)(HNH4IMMB(J)))
ENDIF
DEATHB(J)=.8*PHYSDEGC(J)/BIOMCN+MICRONC*(HTDEATHB(J)+NSDEATHB(J))
HERE I AM GOING TO PUT INFORMATION ON THE REMOVAL OF THE
BIOMASS FROM THE SYSTEM WHICH WILL THEN REMOVE A CERTAIN
PERCENTAGE OF IMMOBILIZED N THE SYSTEM. WHAT I NEED
TO FIGURE OUT IS THE AMOUNT THAT NEEDS TO BE REMOVED BASED
ON THE SEPARATION RATIO BETWEEN THE PLANTS AND MICROBES
BUT WHAT IS THIS SEPARATION
IF (BIOMREML.EQ.0) THEN
IMMNOUT(J)=0.0
ELSE IF (BIOMREML.EQ.1) THEN
IF (J.EQ.DAYREML) THEN
IMMNOUT(J)=ABIOREML/BIOMPN
ELSE
IMMNOUT(J)=0.0
ENDIF
ENDIF
C
/
/
/
C
C
C
C
C
IF (BIOMREMD.EQ.0) THEN
IMMNOUTD(J)=0.0
ELSE IF (BIOMREMD.EQ.1) THEN
IF (J.EQ.DAYREMD) THEN
IMMNOUTD(J)=ABIOREMD/BIOMPN
ELSE
IMMNOUTD(J)=0.0
ENDIF
ENDIF
PUT THE RK FOR IMMOBILIZED N HERE
IF (WETTYPE.EQ.0) THEN
IMMNW(J)=IMMNW(J-1)+DONIMMW(J)+PONIMMW(J)-DEATHW(J)+NITUPW(J)+
AMMUPW(J)
IMMNB(J)=IMMNB(J-1)+NITUPB(J)+DONIMMB(J)+AMMUPB(J)
+PONIMMB(J)-DEATHB(J)-IMMNOUT(J)-IMMNOUTD(J)
ELSEIF (WETTYPE.EQ.1) THEN
IMMNB(J)=IMMNB(J-1)+NITUPB(J)+DONIMMB(J)+AMMUPB(J)+PONIMMB(J)
-DEATHB(J)-IMMNOUT(J)-IMMNOUTD(J)
ENDIF
NH4IN=NH4 INFLUENT
PONMINW=PON MINERALIZATION //PONMINB
NH4OUT=NH4 OUTFLOW
AMMUP=ALREADY CALCULATED
NITRIFW=NITRFICATION RATE //NITRIFB
IF (HYDTYPE.EQ.0) THEN
NH4IN(J)=(NH4INC*WATINPUT(J))+NH4CONC*APFLOW(J)
ELSEIF (HYDTYPE.EQ.1) THEN
NH4IN(J)=NH4RC*WATINPUT(J)+NH4CONC*APFLOW(J)
ENDIF
C
IF (WETTYPE.EQ.0) THEN
IF (OUTFLOW(J).LT.WATERVOL(J-1)) THEN
NH4OUT(J)=NH4W(J-1)/(WATERVOL(J-1))*OUTFLOW(J)
ELSEIF (OUTFLOW(J).GE.WATERVOL(J-1)) THEN
NH4OUT(J)=NH4W(J-1)*.75
220
ENDIF
ELSEIF (WETTYPE.EQ.1) THEN
NH4OUT(J)=NH4B(J-1)/WATERB(J-1)*OUTFLOW(J)
ENDIF
NH4EC(J)=NH4OUT(J)/OUTFLOW(J)
IF (WETTYPE.EQ.0) THEN
NITRIFW(J)=NSGROWW(J)/NSYIELDW
ENDIF
NITRIFB(J)=NSGROWB(J)/NSYIELDB
C
C
C
C
C
c
C
C
C
C
C
C
C
C
C
C
IF (PERCINFT(J).GT.0.) THEN
NH4PERC(J)=NH4B(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
NH4PERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THEN
NH4PERC(J)=PERCINFT(J)*NH4ININC
ENDIF
C
/
/
/
IF (WETTYPE.EQ.0) THEN
IF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1) .OR.
(POCW(J-1).LT.0.1)) THEN
PONMINW(J)=0.0
ELSE
PONMINW(J)=POCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)
*PONW(J-1)/POCW(J-1)(POCW(J-1)/TOCW(J-1)*(MICRONC*HTGROWW(J)-HNH4IMMW(J)))
ENDIF
ENDIF
IF((TOCB(J-1)/TONB(J-1)) .GT. MICTCNB(J-1) .OR.
(POCB(J-1) .LT. 0.1)) THEN
PONMINB(J)=0.0
ELSE
PONMINB(J)=POCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
*PONB(J-1)/POCB(J-1)-
221
C
C
(POCB(J-1)/TOCB(J-1)*(MICRONC*HTGROWB(J)-HNH4IMMB(J)))
IF (PONMINB(J).LE.0) THEN
PONMINB(J)=0.0
ENDIF
ENDIF
AMMONIUM MASS BALANCE
IF (WETTYPE.EQ.0) THEN
NH4W(J)=NH4W(J-1)+NH4IN(J)+DONMINW(J)+PONMINW(J)-NH4OUT(J)
-NITRIFW(J)+NH4ATDEP(J)-VOLATIZ(J)+NH4MT(J)-AMMUPW(J)
NH4B(J)=NH4B(J-1)+DONMINB(J)+PONMINB(J)-AMMUPB(J)-NITRIFB(J)-NH4MT(J)
/
C
C
C
C
IF(WETTYPE.EQ.0) THEN
NO3W(J)=NO3W(J-1)+NO3IN(J)+NITRIFW(J)-NO3OUT(J)-DENITW(J)
+NO3AT(J)+NO3MT(J)-NITUPW(J)
NO3B(J)=NO3B(J-1)+NITRIFB(J)-NITUPB(J)-DENITB(J)-NO3MT(J)--NO3PERC(J)
222
/
C
C
/
/
RESUSPV=7.2*(((PONFALL/86400)**1/3)*
((HI(J-1)-HT(J-1))**(1./6.))/(MANNC*
((PONSIZE**(2./3.)))))
IF (WATERVEL.GE.RESUSPV) THEN
PONRE(J)=PONB(J-1)*PONRES*RESTHN/(HT(J-1)-HB)
ELSE
PONRE(J)=0
ENDIF
ENDIF
IF (WETTYPE.EQ.0) THEN
IF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
OUTLET.EQ.5) THEN
IF (PONFALL .GT. (HI(J-1)-HT(J-1))) THEN
PONOUT(J)=0.0
ELSEIF (PONFALL .LT. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J).LT.WATERVOL(J-1)) THEN
PONOUT(J)=PONCOUT*((PONW(J-1))*OUTFLOW(J)/WATERVOL(J-1))
ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THEN
PONOUT(J)=.5*PONCOUT*((PONW(j-1))*OUTFLOW(J)/WATERVOL(J-1))
ENDIF
ENDIF
ELSEIF (OUTLET.EQ.4)THEN
IF (PONFALL .GT. (HI(J-1)-HT(J-1))) THEN
PONOUT(J)=0.
ELSEIF (PONFALL .LT. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J).LE.WATERVOL(J-1)) THEN
PONOUT(J)=((PONW(J-1))*OUTFLOW(J)/WATERVOL(J-1)*PONCOUT)
ELSEIF (OUTFLOW(J).GE.WATERVOL(J-1)) THEN
PONOUT(J)=((PONW(J-1))*.5*OUTFLOW(J-1)/WATERVOL(J-1)*PONCOUT)
ENDIF
ENDIF
ENDIF
ENDIF
PONEC(J)=PONOUT(J)/OUTFLOW(J)
MASS BALANCE FOR PON
223
C
/
/
/
C
C
IF (WETTYPE.EQ.0) THEN
PONW(J)=PONW(J-1)+DEATHW(J)+PONIN(J)-PEATNACW(J)-PONIMMW(J)
-PONMINW(J)-PONSET(J)+PONRE(J)-PONOUT(J)+PONAT(J)
PONB(J)=PONB(J-1)+DEATHB(J)-PEATNACB(J)-PONIMMB(J)
-PONMINB(J)+PONSET(J)-PONRE(J)
ELSEIF (WETTYPE.EQ.1) THEN
PONB(J)=PONB(J-1)+DEATHB(J)+PONIN(J)-PEATNACB(J)-PONIMMB(J)PONMINB(J)+PONAT(J)
ENDIF
MASS BALANCE FOR REFRACTORY N
IF (WETTYPE.EQ.0) THEN
REFN(J)=REFN(J-1)+PEATNACW(J)+PEATNACB(J)
TONW(J)=PONW(J)+DONW(J)
TONB(J)=PONB(J)+DONB(J)
ELSEIF (WETTYPE.EQ.1) THEN
REFN(J)=REFN(J-1)+PEATNACB(J)
TONB(J)=PONB(J)+DONB(J)
ENDIF
C
C
224
OXYGEN SUBMODELS
/
/
/
C
C
IF (M.EQ.1) THEN
IF (WETTYPE.EQ.0) THEN
READ (10,*) DOINITW,DOINITB,HTDOYW,HTDOYB,NSDOYW,NSDOYB,
DOCONCP,MTDOX,MTFWSDOC,DOXYCSAT
ELSEIF(WETTYPE.EQ.1) THEN
READ(10,*) DOINITB,HTDOYB,NSDOYB,DOCONCP
ENDIF
IF (POINT.EQ.0) THEN
OXYCONC=0.
ELSEIF (POINT.EQ.1) THEN
READ (10,*) OXYCONC
ENDIF
IF (PERCINF.EQ.0) THEN
DOININC=0.0
ELSEIF (PERCINF.EQ.1) THEN
READ (12,*) DOININC
ENDIF
IF (HYDTYPE.EQ.0 ) THEN
OXDRC=0.0
ELSEIF (HYDTTYPE.EQ.1) THEN
READ(10,*) OXDRC
ENDIF
ENDIF
C
C
C
IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THEN
READ(10,*)HTDOYW,HTDOYB,NSDOYW,NSDOYB,DOCONCP,
MTDOX,MTFWSDOC,DOXYCSAT
ELSEIF (WETTYPE.EQ.1) THEN
READ(10,*) HTDOYB,NSDOYB,DOCONCP
ENDIF
ENDIF
IF (CHECK2.EQ.0) THEN
IF(POINT.EQ.1) THEN
READ(10,*) OXYCONC
ENDIF
IF (PERCINF.EQ.1) THEN
READ (10,*) DOININC
ENDIF
IF (HYDTYPE.EQ.1) THEN
READ(10,*) OXDRC
ENDIF
ENDIF
ENDIF
C
END
C
C
225
/
/
/
/
/
/
/
/
/
/
REAL DOXYW(0:500),DOXYB(0:500),WATERVOL(0:500),WATERB(0:500),
OUTFLOW(0:500),DOINF(0:500),DOOUT(0:500),HI(0:500),HT(0:500),
NSGROWW(0:500),AEROHTGB(0:500),DOXYCW(0:500),DOXYCB(0:500),
NSGROWB(0:500),AEROHTGW(0:500),WATINPUT(0:500),PRECIP(0:500),
NSRESPW(0:500),NSRESPB(0:500),MTDOXY(0:500),MTFWS(0:500),
HTRESPW(0:500),HTRESPB(0:500),DOPERC(0:500),PERCINFT(0:500),
APFLOW(0:500),BIOMASST(0:500)
/
/
REAL DOINITW,DOINITB,HTDOYW,NSDOYW,NSDOYB,HTDOYB,OXYCONC,OXDRC,
DOCONCP,MTFWSDOC,MTDOX,LENGTH,WIDTH,HO,SO,HB
WATERVEL,BIOOXRB,BIOFLUXB,DOXYCSAT,DOININC
INTEGER HYDTYPE,WETTYPE,J,K,M
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
C
C
C
C
MICROBIAL RESPIRATION
IF (WETTYPE.EQ.0) THEN
HTRESPW(J)=AEROHTGW(J)/HTDOYW
NSRESPW(J)=NSGROWW(J)/NSDOYW
ENDIF
HTRESPB(J)=AEROHTGB(J)/HTDOYB
NSRESPB(J)=NSGROWB(J)/NSDOYB
IF (WETTYPE.EQ.0) THEN
MTDOXY(J)=864*LENGTH*WIDTH*MTDOX*(DOXYCB(J-1)-DOXYCW(J-1))
MTFWS(J)=864*LENGTH*WIDTH*MTFWSDOC*(DOXYCSAT-DOXYCW(J-1))
ENDIF
IF (PERCINFT(J).GT.0.) THEN
DOPERC(J)=DOXYB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
DOPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THEN
DOPERC(J)=PERCINFT(J)*DOININC
ENDIF
C
C
226
C
/
/
/
C
C
C
IF (WETTYPE.EQ.0) THEN
DOXYW(J)=DOXYW(J-1)+DOINF(J)+MTDOXY(J)+MTFWS(J)-DOOUT(J)
-HTRESPW(J)-NSRESPW(J)
DOXYB(J)=DOXYB(J-1)+BIOFLUXB-HTRESPB(J)-NSRESPB(J)-MTDOXY(J)
-DOPERC(J)
ELSEIF (WETTYPE.EQ.1) THEN
DOXYB(J)=DOXYB(J-1)+DOINF(J)-DOOUT(J)+BIOFLUXB-HTRESPB(J)
-NSRESPB(J)-DOPERC(J)
ENDIF
WRITE TO OUTPUT FILES
227
PHOSPHOROUS SUBMODELS
/
/
SUBROUTINE PHOSSTR(DTPHOSIW,PPHOSI,BTPHOSI,M,MTCPHOS,
PMINPPC,LINPARTC,FREUNDK,FREUNDN,ADSORP,BIOMPP,
PRMINBPC,POINT,PHOSCON,PHOSRC,DTPHOSIB,PININC,PERCINF)
INTEGER ADSORP,POINT,CHECK1,CHECK2,PERCINF
/
/
C
C
C
REAL DTPHOSI,PPHOSI,BTPHOSI,PMINPPC,PRMINBPC,PHOSCON,
LINPARTC,FREUNDK,FREUNDN,BIOMPP,MTCPHOS,PININC
PHOSRC
INITIAL INPUT VALUES
IF (M.EQ.1) THEN
READ(14,*) DTPHOSIW,DTPHOSIB,BTPHOSI,PPHOSI,PMINPPC,
PRMINBPC,ADSORP,MTCPHOS,BIOMPP
IF (PERCINF.EQ.0) THEN
PININC=0.0
ELSEIF (PERCINF.EQ.1) THEN
READ (10,*) PININC
ENDIF
IF (ADSORP.EQ.0) THEN
READ (14,*) FREUNDK,FREUNDN
LINPARTC=0.
ELSEIF (ADSORP.EQ.1) THEN
READ(14,*) LINPARTC
FREUNDK=0.
FREUNDN=0.
ELSEIF (ADSORP.EQ.2) THEN
FREUNDK=0
FREUNDN=0
READ(14,*) LINPARTC
ENDIF
IF (POINT.EQ.1 .OR. ADSORP.EQ.2) THEN
READ(14,*) PHOSCON
ELSE
PHOSCON=0.0
ENDIF
ENDIF
C
C
C
C
C
/
228
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
SEDBSA,HI,HT,PPHOST,BTPHOST,SEDINW,APFLOW,PHOSCON,
PHOSPER,BIOMPP,BIOMASST,PHYSDEG,DTPHOSIW,PPHOSI,
WATINPUT,SEDDEP,RESUSP,PRMINBPC,MTCPHOS,BTPHOSI,WATERVOL,
J,K,M,LINPARTC,POINT,PHOSRC,PHOSPERB,DTPHOSIB,DTPHOSB,
WATERB,SEDSPG,SEDBV,PHOSPERW,BIOMGROW,SEDOUT,PRATEUP,
PERCINFT,PININC,DTPHOSCW,DTPHOSCB,DPHOSOUT,PPHOSOUT)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
REAL DTPHOSW(0:500),SEDINITB(5),PPHOSIN(5,0:500),SEDINITW(5),
SEDQTYB(5,0:500),SEDQTYW(5,0:500),SEDBSA(5),PPHOS(5,0:500),
BTPHOS(5,0:500),PPHOSOUT(5,0:500),PPHOSSET(5,0:500),
PPHOSRES(5,0:500),HI(0:500),HT(0:500),DISPHOSI(0:500),
SEDINW(5,0:500),PHOSPER(5,0:500),WATERB(0:500),
PRMINBPT(0:500),PHYSDEG(0:500),BIOMASST(0:500),OUTFLOW(0:500),
PPHOST(0:500),BTPHOST(0:500),PPHOSINT(0:500),PMINPP(5,0:500),
PRMINBP(5,0:500),DPHOSOUT(0:500),WATERVOL(0:500),
PHOSMT(0:500),DEADPHOT(0:500),PHOSUPB(0:500),DTPHOSB(0:500),
DEADPHOS(5,0:500),PPHOSUP(5,0:500),WATINPUT(0:500),
PMINPPT(0:500) ,PPHOSCW(0:500),SEDDEP(5,0:500),
SEDINCV(5,0:500),SEDPART(5,0:500),SEDBV(5),BIOMGROW(0:500),
RESUSP(5,0:500),PHOSPERB(5,0:500),SEDSPG(5),
SEDTSAW(5,0:500),SEDTSAB(5,0:500),PHOSPERW(5,0:500),
SEDOUT(5,0:500),PHOSUPW(0:500),APFLOW(0:500),PERCINFT(0:500),
PHOSPERC(0:500),DTPHOSCW(0:500),DTPHOSCB(0:500)
INTEGER ADSORP,POINT,J,K,M,SEDCAT,SEDCLASS
REAL DISPHOSC,DTPHOSIB,PPHOSI,FREUNDN,FREUNDK,LINPARTC,MTCPHOS,
LENGTH,WIDTH,HO,HB,SO,PPHOSINC,BTPHOSI,PRMINBPC,
BIOMPP,PHOSCON,PPHOSINP,PHOSRC,PPHOSCAL,DTPHOSIW,PININC
PPHOSTOT,SEDTSATP,SEDTSATU,SEDTSATC,PRATEUP
IF (M.EQ.1 .AND. J.EQ.1) THEN
DTPHOSW(0)=DTPHOSIW
DTPHOSB(0)=DTPHOSIB
PPHOST(0)=PPHOSI
BTPHOST(0)=BTPHOSI
PPHOSCW(0)=PPHOST(0)/WATERVOL(0)
DO 25 SEDCLASS=1,SEDCAT
PPHOS(SEDCLASS,0)=PPHOSI*PHOSPERW(SEDCLASS,0)
25 CONTINUE
DO 50 SEDCLASS=1,SEDCAT
BTPHOS(SEDCLASS,0)=BTPHOSI*PHOSPERB(SEDCLASS,0)
50 CONTINUE
ENDIF
IF (J.EQ.1) THEN
DO 60 CLEAR=1,100
PMINPPT(CLEAR)=0.0
PRMINBPT(CLEAR)=0.0
BTPHOST(CLEAR)=0.0
PPHOST(CLEAR)=0.0
60 CONTINUE
ENDIF
IF (HYDTYPE.EQ.0 .AND. (ADSORP.EQ.0 .OR. ADSORP.EQ.1)) THEN
READ (14,*) DISPHOSC
ELSEIF (HYDTYPE.EQ.0 .AND. ADSORP.EQ.2) THEN
READ (14,*) DISPHOSC, PPHOSCAL
ENDIF
IF (HYDTYPE.EQ.1) THEN
DISPHOSC=PHOSRC
ENDIF
IF (ADSORP.EQ.0) THEN
PPHOSINC=FREUNDK*(DISPHOSC**(1./FREUNDN))/1000
IF (POINT.EQ.1) THEN
PPHOSINP=FREUNDK*((PHOSCON)**(1./FREUNDN))/1000
ELSE IF (POINT.EQ.0) THEN
PPHOSINP=0.0
229
ENDIF
ELSEIF(ADSORP.EQ.1) THEN
PPHOSINC=DISPHOSC*LINPARTC/1000
IF (POINT.EQ.1) THEN
PPHOSINP=(PHOSCON)*LINPARTC/1000
ELSEIF (POINT.EQ.0) THEN
PPHOSINP=0.0
ENDIF
ELSEIF (ADSORP.EQ.2) THEN
IF (POINT.EQ.1) THEN
PPHOSINP=(PHOSCON)*LINPARTC/1000
ELSEIF (POINT.EQ.0) THEN
PPHOSINP=0.0
ENDIF
ENDIF
IF (ADSORP.EQ.0 .OR. ADSORP.EQ.1) THEN
PPHOSINT(J)=WATINPUT(J)*PPHOSINC+APFLOW(J)*PPHOSINP
ELSEIF (ADSORP.EQ.2) THEN
PPHOSINT(J)=PPHOSCAL+APFLOW(J)*PPHOSINP
ENDIF
IF (HYDTYPE.EQ.0) THEN
DISPHOSI(J)=WATINPUT(J)*DISPHOSC+APFLOW(J)*PHOSCON
ELSEIF (HYDTYPE.EQ.1) THEN
DISPHOSI(J)=WATINPUT(J)*PHOSRC+APFLOW(J)*PHOSCON
ENDIF
DO 75 SEDCLASS=1,SEDCAT
PMINPP(SEDCLASS,J)=PMINPPC*PPHOS(SEDCLASS,J-1)
PRMINBP(SEDCLASS,J)=PRMINBPC*BTPHOS(SEDCLASS,J-1)
PMINPPT(J)=PMINPPT(J)+PMINPP(SEDCLASS,J)
PRMINBPT(J)=PRMINBPT(J)+PRMINBP(SEDCLASS,J)
75 CONTINUE
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
DPHOSOUT(J)=DTPHOSW(J-1)*OUTFLOW(J)/WATERVOL(J-1)
ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THEN
DPHOSOUT(J)=.75*DTPHOSW(J-1)* OUTFLOW(J)/WATERVOL(J-1)
ENDIF
PHOSMT(J)=864*LENGTH*WIDTH*MTCPHOS*(DTPHOSCB(J-1)-DTPHOSCW(J-1))
IF (DTPHOSB(J-1).GT.(PRATEUP*BIOMGROW(J)/BIOMPP)) THEN
PHOSUPB(J)=BIOMGROW(J)/BIOMPP*PRATEUP
ELSE
PHOSUPB(J)=DTPHOSB(J-1)
ENDIF
IF (DTPHOSW(J-1).GT.((1-PRATEUP)*BIOMGROW(J)/BIOMPP)) THEN
PHOSUPW(J)=BIOMGROW(J)/BIOMPP*(1-PRATEUP)
ELSE
PHOSUPW(J)=DTPHOSW(J-1)
ENDIF
IF (PERCINFT(J).GT.0.) THEN
PHOSPERC(J)=DTPHOSB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
PHOSPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THEN
PHOSPERC(J)=PERCINFT(J)*PININC
ENDIF
C
C
C
DTPHOSW(J)=DTPHOSW(J-1)+DISPHOSI(J)+PMINPPT(J)DPHOSOUT(J)+PHOSMT(J)-PHOSUPW(J)
DTPHOSB(J)=DTPHOSB(J-1)-PHOSMT(J)+PRMINBPT(J)-PHOSUPB(J)+PHOSPERC(J)
230
DO 125 SEDCLASS=1,SEDCAT
PPHOSIN(SEDCLASS,J)=PPHOSINT(J)*PHOSPER(SEDCLASS,J)
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
PPHOSOUT(SEDCLASS,J)=SEDOUT(SEDCLASS,J)/
/
SEDQTYW(SEDCLASS,J)*PPHOS(SEDCLASS,J-1)
ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THEN
PPHOSOUT(SEDCLASS,J)=.5*SEDOUT(SEDCLASS,J)/
/
SEDQTYW(SEDCLASS,J)*PPHOS(SEDCLASS,J-1)
ENDIF
125 CONTINUE
DO 150 SEDCLASS=1,SEDCAT
PPHOSSET(SEDCLASS,J)=SEDDEP(SEDCLASS,J)/
/
SEDQTYW(SEDCLASS,J-1)*PPHOS(SEDCLASS,J-1)
150 CONTINUE
DO 175 SEDCLASS=1,SEDCAT
PPHOSRES(SEDCLASS,J)=RESUSP(SEDCLASS,J)
/
/SEDQTYB(SEDCLASS,J-1)*BTPHOS(SEDCLASS,J-1)
175 CONTINUE
DO 200 SEDCLASS=1,SEDCAT
PPHOS(SEDCLASS,J)=PPHOS(SEDCLASS,J-1)+PPHOSIN(SEDCLASS,J)/
PMINPP(SEDCLASS,J)-PPHOSSET(SEDCLASS,J)+PPHOSRES(SEDCLASS,J)/
PPHOSOUT(SEDCLASS,J)
200 CONTINUE
DO 225 SEDCLASS=1,SEDCAT
PPHOST(J)=PPHOST(J)+PPHOS(SEDCLASS,J)
225 CONTINUE
DEADPHOT(J)=PHYSDEG(J)/BIOMPP
DO 300 SEDCLASS=1,SEDCAT
DEADPHOS(SEDCLASS,J)=DEADPHOT(J)*PHOSPERB(SEDCLASS,J-1)
300 CONTINUE
DO 350 SEDCLASS=1,SEDCAT
BTPHOS(SEDCLASS,J)=BTPHOS(SEDCLASS,J-1)+PPHOSSET(SEDCLASS,J)
/
+DEADPHOS(SEDCLASS,J)-PRMINBP(SEDCLASS,J)-PPHOSRES(SEDCLASS,J)
BTPHOST(J)=BTPHOST(J)+BTPHOS(SEDCLASS,J)
350 CONTINUE
SEDTSATC=0
DO 360 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,J)=SEDQTYB(SEDCLASS,J)*(1./
/
(SEDSPG(SEDCLASS)/1000.))
SEDPART(SEDCLASS,J)=SEDINCV(SEDCLASS,J)/SEDBV(SEDCLASS)
SEDTSAB(SEDCLASS,J)=SEDPART(SEDCLASS,J)*SEDBSA(SEDCLASS)
SEDTSATC=SEDTSATC+SEDTSAB(SEDCLASS,J)
360 CONTINUE
DO 375 SEDCLASS=1,SEDCAT
PHOSPERB(SEDCLASS,J)=SEDTSAB(SEDCLASS,J)/SEDTSATC
375 CONTINUE
SEDTSATU=0
DO 380 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,J)=SEDQTYW(SEDCLASS,J)*(1./
/
(SEDSPG(SEDCLASS)/1000.))
SEDPART(SEDCLASS,J)=SEDINCV(SEDCLASS,J)/SEDBV(SEDCLASS)
SEDTSAW(SEDCLASS,J)=SEDPART(SEDCLASS,J)*SEDBSA(SEDCLASS)
SEDTSATU=SEDTSATU+SEDTSAW(SEDCLASS,J)
380 CONTINUE
DO 390 SEDCLASS=1,SEDCAT
PHOSPERW(SEDCLASS,J)=SEDTSAW(SEDCLASS,J)/SEDTSATU
390 CONTINUE
C
C
WRITE TO OUTPUT FILES
C
231
232
SEDIMENT SUBMODELS
/
/
/
DO 100 SEDCLASS=1,SEDCAT
READ(9,*) SEDSIZE(SEDCLASS),SEDFALL(SEDCLASS),
/
SEDINITW(SEDCLASS),SEDINITB(SEDCLASS),
/
SEDSPG(SEDCLASS),SEDPER(SEDCLASS)
100 CONTINUE
READ (9,*) RESTHICK,MANNC,DECOMPR, PSEDDEP
DO 200 SEDCLASS=1,SEDCAT
SEDBV(SEDCLASS)=(88./21.*(((SEDSIZE(SEDCLASS))/2)**3))
SEDBSA(SEDCLASS)=(88./7.*((SEDSIZE(SEDCLASS)/2)**2))
200 CONTINUE
IF (POINT.EQ.1) THEN
READ (9,*) SEDCONC
ELSE
SEDCONC=1.0
ENDIF
ENDIF
C
C
C
IF (M.GT.1) THEN
READ (9,*) c, CHECK
IF (CHECK.EQ.0) THEN
READ (9,*) RESTHICK,MANNC,SEDCONC,
DECOMPR,SEDRES,PSEDDEP
ENDIF
ENDIF
END
C
C
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
233
234
DO 195 SEDCLASS=1,SEDCAT
PHOSPERB(SEDCLASS,0)=SEDTSAB(SEDCLASS,0)/SEDTSATB
195 CONTINUE
ENDIF
ENDIF
250
/
/
300
DO 400 SEDCLASS=1,SEDCAT
RESUSPV(SEDCLASS,J)=7.2*(((SEDFALL(SEDCLASS)/86400.)**(1./3.))
/
*((HI(J-1)-HT(J-1))**(1./6.))/(MANNC*(((SEDSIZE(SEDCLASS)/1000.)**(2./3.)))))
400 CONTINUE
C
C
CHECK IF RESUSPENSION THEORY IS MET
C
RE1B=(2.*((HI(J-1)-HT(J-1))**(1./6.)))/((MANNC*(9.81**(0.5))))
DO 500 SEDCLASS=1,SEDCAT
RE1(SEDCLASS,J)=(SEDSIZE(SEDCLASS)/1000.)*1000*
/
(RESUSPV(SEDCLASS,J)/86400.)/0.001
451
/
452
/
453
/
454
/
IF (RE1(SEDCLASS,J).GT.RE1B) THEN
IF (SEDCLASS.EQ.1) THEN
IF (C1.EQ.0) THEN
WRITE (*,451)
FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR'
' PARTICLE CATEGORY 1')
ENDIF
C1=1
ENDIF
IF (SEDCLASS.EQ.2) THEN
IF (C2.EQ.0) THEN
WRITE (*,452)
FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR'
' PARTICLE CATEGORY 2')
ENDIF
C2=1
ENDIF
IF (SEDCLASS.EQ.3) THEN
IF (C3.EQ.0) THEN
WRITE (*,453)
FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR'
' PARTICLE CATEGORY 3')
ENDIF
C3=1
ENDIF
IF (SEDCLASS.EQ.4) THEN
IF (C4.EQ.0) THEN
WRITE (*,454)
FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR'
' PARTICLE CATEGORY 4')
ENDIF
C4=1
ENDIF
IF (SEDCLASS.EQ.5) THEN
235
IF (C5.EQ.0) THEN
WRITE (*,455)
455
FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR'
/
' PARTICLE CATEGORY 5')
ENDIF
C5=1
ENDIF
ENDIF
500 CONTINUE
DO 550 SEDCLASS=1,SEDCAT
IF (RESUSPV(SEDCLASS,J).LE.WATERVEL) THEN
RESUSP(SEDCLASS,J)=SEDQTYB(SEDCLASS,J-1)*SEDRES*
/
RESTHICK/(HT(J-1)-HB)
ELSEIF (RESUSPV(SEDCLASS,J).GT.WATERVEL) THEN
RESUSP(SEDCLASS,J)=0.0
ENDIF
550 CONTINUE
DO 600 SEDCLASS=1,SEDCAT
IF (SEDFALL(SEDCLASS) .LT. (HI(J-1)-HT(J-1))) THEN
SEDDEP(SEDCLASS,J)=(SEDQTYW(SEDCLASS,J-1))*
/
(SEDFALL(SEDCLASS))/(HI(J-1)-HT(J-1))*PSEDDEP
ELSE IF (SEDFALL(SEDCLASS) .GE. (HI(J-1)-HT(J-1)))THEN
SEDDEP(SEDCLASS,J)=SEDQTYW(SEDCLASS,J-1)*PSEDDEP
ENDIF
600 CONTINUE
DO 700 SEDCLASS=1,SEDCAT
IF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
/
OUTLET.EQ.5) THEN
IF (SEDFALL(SEDCLASS) .GT.
/
(HI(J-1)-HT(J-1))) THEN
SEDOUT(SEDCLASS,J)=0.
ELSEIF (SEDFALL(SEDCLASS) .LE. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
SEDOUT(SEDCLASS,J)=((SEDQTYW(SEDCLASS,J-1)))
/
*OUTFLOW(J)/WATERVOL(J-1)
ELSEIF (OUTFLOW (J) .GE. WATERVOL(J-1)) THEN
SEDOUT(SEDCLASS,J)=(.5*(SEDQTYW(SEDCLASS,J-1)))
/
*OUTFLOW(J)/WATERVOL(J-1)
ENDIF
ENDIF
ELSEIF (OUTLET.EQ.4) THEN
IF (SEDFALL(SEDCLASS) .GT.(HI(J-1)-HT(J-1))) THEN
SEDOUT(SEDCLASS,J)=0.
ELSEIF (SEDFALL(SEDCLASS) .LE. (HI(J-1)-TOPPUMP))THEN
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
SEDOUT(SEDCLASS,J)=(SEDQTYW(SEDCLASS,J-1))
/
*OUTFLOW(J)/WATERVOL(J-1)
ELSEIF (OUTFLOW (J) .GE. WATERVOL(J-1)) THEN
SEDOUT(SEDCLASS,J)=(.5*SEDQTYW(SEDCLASS,J-1))
/
*OUTFLOW(J)/WATERVOL(J-1)
ENDIF
ENDIF
ENDIF
SEDOUTT(J)=SEDOUTT(J)+SEDOUT(SEDCLASS,J)
700 CONTINUE
C
C
AMOUNT OF DECOMPOSITION FROM PLANT LIFE
C
SDECOMP(SEDCAT,J)=DECOMPR*SEDQTYB(SEDCAT,J-1)
C
C
MASS BALANCE FOR SEDIMENT
C
DO 800 SEDCLASS=1,SEDCAT
SEDQTYW(SEDCLASS,J)=SEDQTYW(SEDCLASS,J-1)+RESUSP(SEDCLASS,J)/
-SEDOUT(SEDCLASS,J)-SEDDEP(SEDCLASS,J)+SEDINW(SEDCLASS,J)
800 CONTINUE
236
DO 900 SEDCLASS=1,SEDCAT
SEDQTYB(SEDCLASS,J)=SEDQTYB(SEDCLASS,J-1)-RESUSP(SEDCLASS,J)
/
+SEDDEP(SEDCLASS,J)-SDECOMP(SEDCLASS,J)
900 CONTINUE
1000
IF (J.EQ.K) THEN
DO 1000 SEDCLASS=1,SEDCAT
SEDQTYW(SEDCLASS,0)=SEDQTYW(SEDCLASS,J)
SEDQTYB(SEDCLASS,0)=SEDQTYB(SEDCLASS,J)
CONTINUE
ENDIF
C
END
237
DELTAHT SUBMODEL
/
/
SUBROUTINE DELTAH(DELTAHT,NITCYCLE,SEDCYCLE,SEDTOTAL,SEDDELTA,
SEDCAT,SEDSPG,PEATACRB,PEATDENS,HT,NUMTMPER,J,SEDQTYB,WATERB,
PORPEAT)
COMMON/DESCRIBE/LENGTH,WIDTH,HO,HB,SO
REAL SEDDELTA(5),SEDQTYB(5,0:500),HT(0:500),PEATACRB,WATERB(0:500)
REAL DELTAHT,PEATDENS,SEDSPG,PORPEAT,LENGTH,WIDTH,HO,HB,SO
INTEGER NITCYCLE,SEDCYCLE,SEDCAT,J
IF (NITCYCLE.EQ.1 .AND. SEDCYCLE.EQ.0) THEN
DELTAHT=(PEATACRB/(PEATDENS*1000.))/(LENGTH*WIDTH)
ENDIF
75
85
ENDIF
HT(J)=HT(J-1)+DELTAHT
WATERB(J)=LENGTH*WIDTH*(HT(J)-HB)*PORPEAT
IF (J.EQ.NUMTMPER) THEN
HT(0)=HT(J)
WATERB(0)=WATERB(J)
ENDIF
END
238
239
FIGURE E.1.: SIMULATED AND OBSERVED VALUES FOR OUTFLOW, PLOTTED BESIDE THE DETERMINED LINEAR
REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.2.: SIMULATED AND OBSERVED VALUES FOR AMMONIUM CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
240
FIGURE E.3.: SIMULATED AND OBSERVED VALUES FOR NITRATE CONCENTRATION, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.4.: SIMULATED AND OBSERVED VALUES FOR ORGANIC NITROGEN CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
241
FIGURE E.5.: SIMULATED AND OBSERVED VALUES FOR DISSOLVED OXYGEN CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.6.: SIMULATED AND OBSERVED VALUES FOR BOD5 CONCENTRATIONS, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
242
FIGURE E.7.: SIMULATED AND OBSERVED VALUES FOR TOTAL SUSPENDED SOLID CONCENTRATIONS, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.8.: SIMULATED AND OBSERVED VALUES FOR TOTAL PHOSPHOROUS CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
243
B ase Value
N H4
NO3
DON
PO N
DO
BOD5
T SS
DP
TP
0.01
0.05
0.01
0.05
240000
25000
0.15
0.15
50
50
0.15
0.5
0.00125
0.001
1
1
0.002
0.002
2400
1000
0.005
0.005
1
1
(NC)/(+ E )
(+ D)/(+ D)
(+ D)/(+ D)
(+ B )/(+ B )
(NC)/(+ E )
(+ D)/(+ D)
(-C)/(-C)
(-B )/(-B )
(-D)/(-D)
(-D)/(-D)
(-D)/(-E )
(-C)/(-C)
(-D)/(-D)
(-C)/(-C)
(-C)/(-C)
(-B )/(-B )
(-C)/(-C)
(-B )/(-B )
(+ D)/(+ D)
(+ C)/(+ C)
(+ D)/(+ D)
(+ C)/(+ C)
(-C)/(-C)
(-B )/(-B )
(NC)/(-E )
(-C)/(-C)
(-D)/(-D)
(-C)/(-C)
(NC)/(-E )
(-C)/(-C)
(+ D)/(+ D)
(+ B )/(+ B )
(+ C)/(+ C)
(+ C)/(+ C)
(-C)/(-C)
(-C)/(-C)
(+ D)/(+ D)
(+ B )/(+ B )
(-C)/(-C)
(-C)/(-C)
(-C)/(-C)
(-C)/(-C)
(+ D)/(+ D)
(+ D)/(+ D)
(+ D)/(+ D)
(+ D)/(+ D)
(-C)/(-C)
(-C)/(-C)
(NC)/(-D)
(-D)/(-C)
(-B )/(-B )
(-B )/(-B )
(NC)/(-C)
(-C)/(-C)
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ C)/(+ C)
(+ C)/(+ C)
(+ B )/(+ B )
(+ B )/(+ C)
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(+ B )/(+ B )
(+ B )(+ B )
(NC)/(-C)
(-C)/(-C)
(-B )/(-B )
(-B )/(-B )
(NC)/(-C)
(-B )/(-B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ C)/(+ C)
(+ C)/(+ C)
(+ B )/(+ B )
(+ B )/(+ B )
(+ C)/(+ C)
(+ B )/(+ B )
(A )/(A )
(+ B )/(+ B )
(A )/(A )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(+ B )/(+ B )
A /A
(NC)/(-E )
(-D)/(-D)
(-C)/(-C)
(-B )/(-B )
(NC)/(-E )
(-D)/(-D)
(+ B )/(+ B )
(+ B )/(+ B )
(+ C)/(+ C)
(+ D)/(+ D)
(+ D)/(+ D)
(+ C)/(+ C)
(+ D)/(+ D)
(+ C)/(+ C)
(+ C)/(+ C)
(+ C)/(+ C)
(+ C)/(+ C)
(+ C)/(+ B )
(-D)/(-D)
(-C)/(-C)
(-D)/(-D)
(-D)/(-D)
(+ C)/(+ (C)
(+ C)/(+ C)
(NC)/(-C)
(-D)/(-C)
(-B )/(-B )
(-B )/(-B )
(NC)/(-C)
(-C)/(-C)
(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ C)/(+ C)
(+ C)/(+ C)
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(+ B )/(+ B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(+ B )/(+ B )
(+ B )/(+ B )
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
(-D)/(+ D)
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
A /A
(-D)/(+ D)
23.5
95
10000
5000
3.29
3.29
9900000
700000
0.125
0.00002
0.00006
0.00006
55000
30000
4000
400
0.3
0.3
0.6
0.75
0.25
300000
15000
0.01
0.05
500000
0.01
(+ E )/(-D)
(+ E )/(+ E )
(-C)/(+ C)
(-B )/(+ C)
(+ C)/(+ C)
(-C)/(+ C)
(-C)/(+ C)
(-C)/(+ C)
(-D)/(-D)
(-B )/(+ B )
(-C)/(+ B )
(-B )/(+ C)
(+ E )/(+ E )
(+ E )/(+ D)
(+ C)/(-E )
(+ B )/(+ C)
(-C)/(+ C)
(-C)/(+ C)
(+ D)/(+ D)
(-C)/(-C)
(-B )/(+ C)
(+ E )/(-E )
(+ D)/(-D)
(-C)/(+ C)
(-C)/(+ C)
(+ E )/(-E )
(-C)/(+ C)
(+ F)/(-B )
(+ D)/(+ D)
(-B )/(+ B )
(-B )/(+ B )
(+ D)/(+ D)
(+ C)/(+ C)
(-B )/(+ B )
(-B )/(+ B )
(-B )/(-B )
(-B )/(+ B )
(-B )/(-B )
(+ C)/(+ C)
(+ C)/(+ C)
(+ B )/(+ C)
(+ D)/(-F)
(+ D)/(+ D)
(-D)/(-D)
(-D)/(-D)
(+ B )/(+ B )
(-B )/(-B )
(-B )/(+ B )
(+ F)/(-F)
(+ F)/(-F)
(-B )/(+ B )
(-B )/(+ B )
(+ F)/(-F)
(-B )/(+ B )
(-C)/(-C)
(-B )/(-B )
(+ D)/(+ D)
(+ D)/(+ D)
(-B )/(-B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(+ C)/(+ C)
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(+ C)
(-B )/(-B )
(+ B )/(+ B )
(+ B )/(+ B )
(-F)/(-F)
(-B )/(+ B )
(-B )/(+ B )
(-C)/(+ C)
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-C)/(+ C)
(-B )/(+ B )
(-E )/(-D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(+ D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(-D)/(+ D)
(+ E )/(+ E )
(+ E )/(+ E )
(-E )/(-E )
(-D)/(+ D)
(+ D)/(+ D)
(+ D)/(+ E )
(-D)/(+ D)
(-C)/(+ C)
(+ C)/(+ E )
(-D)/(+ B ) (-C)/(+ B )
(-C)/(-D)
(-B )/(-B )
(-B )/(-B ) (-B )/(+ B )
(-B )/(-B ) (-B )/(+ B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B ) (-B )/(+ B )
(-B )/A
(-B )/(+ B )
(-B )/(+ B ) (-B )/(+ B )
(+ B )/(+ B ) (-B )/(+ B )
(-B )/(+ B ) (-B )/(+ B )
(-B )/(-B ) (-B )/(+ B )
(-B )/(-B )
(-B )/(-B )
(-C)/(-C)
(-B )/(-B )
(-C)/(-C)
(-B )/(-B )
(-B )/(+ C) (-B )/(+ C)
(-B )/(-B )
(-B )/(-B )
(+ B )/(+ B ) (+ B )/(+ B )
(+ B )/(+ B ) (-B )/(+ B )
(-B )/(+ B ) (-B )/(+ B )
(+ B )/(+ B ) (-B )/(+ B )
(+ B )/(+ B ) (-B )/(+ B )
(-D)/(+ D) (-C)/(+ C)
(-C)/(+ C) (-B )/(+ B )
(-B )/(+ B ) (-B )/(+ B )
(-B )/(+ B ) (-B )/(+ B )
(-D)/(+ D) (-C)/(+ C)
(-B )/(+ B ) (-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
A /A
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B /(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
A /A
(-B )/(+ B )
(-B )/(+ B )
(-B )/(+ B )
(-B /(+ B )
(-B )/(+ B )
5000
8635000
0.4
2000
300
0.7
(+ B )/(-B )
A /(-B )
(-B )/(-B )
(-B )/(-B )
A /(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B ) (+ B )/(+ B ) (+ D)/(+ D) (-C)/(-C) (+ D)/(+ D) (+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B )
(+ B )/(+ B )
(+ B )/A
(+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B ) (+ B )/(+ B )
(+ C)/(+ C) (+ B )/(+ B ) (+ B )/(+ B ) (-B )/(-B ) (+ B )/(+ B ) (+ B )/(+ B ) (-B )/(-B ) (+ B )/(+ B ) (+ B )/(+ B )
(+ B )/(+ B ) (+ B )/(+ B ) (-B )/(-B )
(-C)/(-C) (+ B )/(+ B ) (-B )/(-B )
A /A
A /A
A /A
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B )
(-B )/(-B ) (+ B )/(+ C) (-B )/(-B )
(-B )/(-B )
244
Base Value
0.7
PRATEUP
0.4
PSTDUW
6000000
STANDIN
5000
STDDENS
CARBON
0.47
BIOCCONT
0.8
BODCFRAC *
0.5
BODPFRAC
60000
DOCINITB
50000
DOCINITW
0.01
LEACHR
0.53
M ICROBEC
0.00004
M TCDOC
0.8
PEATCC *
0.3
POCCOUT *
0.45
POCFALL
3000000
POCINITB
200000
POCINITW
0.001
POCRES
0.2
POCSIZE
15000000
REFCINIT
DISSOLVED OXYGEN
0.001
DOCONCP
15000
DOINITB
15000
DOINITW
8.5
DOXYCSAT
0.15
HTDOYB
0.15
HTDOYW
0.0001
M TDOX
0.00008
M TFW SDOC
0.2
NSDOYB
0.2
NSDOYW
PHOSPHOROUS
300
BIOM PP
250000
BTPHOSI
40000
DTPHOSIB
38000
DTPHOSIW
0.00006
M TCPHOS
0.05
PM INPPC
0.00005
PRMINBPC
0.5
PSEDDEP
SEDIMENT
0.1
DECOMPR
2.5
M ANNC (SED)
0.5
PSEDDEP
0.0011
RESTHICK (SED)
0.3
SEDFALL1
0.7
SEDFALL2
26000000
SEDINITB1
10000000
SEDINITB2
190000
SEDINITW 1
10000
SEDINITW 2
0.1
SEDRES
0.25
SEDSIZE1
0.5
SEDSIZE2
1.1
SEDSPG1
2.65
SEDSPG2
NH4
NO3
DON
PON
DO
BOD5
T SS
DP
TP
(-C)/(-C)
(-C)/(-C)
(+C)/(+C)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(+B)/(+B)
(+B)/(+C)
(-C)/(-C)
(-B)/(-B)
(+B)/(+B)
(-B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(+C)/(+C)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(-B)
(+B)/(+C)
(+B)/(+B)
(+B)/(+B)
(-D)/(+D)
(-B)/(+B)
(+D)/(+E)
(-D)/(+C)
(-D)/(+D)
(-C)/(+C)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+C)
(-D)/(+D)
(-D)/(+D)
(-C)/(+D)
(-D)/(+D)
(-D)/(+D)
(-C)/(+C)
(+C)/(+C)
(-C)/(-C)
(-D)/(-D)
(+C)/(+C)
(-C)/(-C)
(-C)/(-C)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(+B)/(+B)
(+C)/(+C)
(-C)/(-C)
(-C)/(-C)
(+B)/(-B)
(-B)/(-B)
(-B)/(+B)
(+D)/(+D)
(-E)/(-E)
(-E)/(-E)
(+D)/(+D)
(+D)/(+D)
(+D)/(+C)
(+B)/(+B)
(+C)/(+C)
(+B)/(+B)
(+E)/(+E)
(-E)/(-E)
(+B)/(+C)
(+C)/(+C)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-C)/(+C)
(+B)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/)(+C)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-F)/(-F)
(+D)/(+D)
(+C)/(+C)
(+E)/(+E)
(+E)/(+E)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(-C)/(-C)
(+D)/(+D)
(+C)/(+C)
(+E)/(+E)
(+E)/(+E)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(-C)(-C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(-C)/(+C)
(-B)/(+B)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-C)/(+C)
(-D)/(+D)
(-C)/(+C)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(+B)
(-B)/((+B)
(-B)/((+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/((+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+E)
(-F)/(-F)
(+E)/(+E)
(-F)/(-F)
(-B)/(-E)
(+E)/(+E)
(+B)/(+B)
(+E)/(+E)
(-B)/(+B)
(+E)/(+E)
(-B)/(+E)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(-B))/(+C)
(-C)/(-C)
(+C)/(+C)
(-C)/(-C)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(-B)
(-B)/(+C)
(-C)/(-C)
(+C)/(+C)
(-C)/(-C)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(+B)
(-B)/(-B)
(+C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(-B)
(+B)/(+B)
- Results presented as the RS values for the +10% change in base value followed by the 10% change.
- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.
- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible
results)
245
TABLE F.2: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 25% CHANGE IN BASE VALUES
Parameter
BACTERIA
AEMAXGRB
AEMAXGRW
ANMAXGRB
ANMAXGRW
HETROIB
HETEROIW
HNO3HSCB
HNO3HSCW
HORGHSCB
HORGHSCW
HTDOHSCB
HTDOHSCW
HTDRB
HTDRW
NDOHSATB
NDOHSATW
NDRATEB
NDRATEW
NITROSIB
NITROSIW
NMAXGRB
NMAXGRW
NNH4HSCB
NNH4HSCW
NITROGEN
BIOMCN
BIOMPN
DONINITB
DONINITW
HTNO3YB
HTNO3YW
IMMINITB
IMMINITW
MICRONC
MTCDON
MTCNH4
MTCNO3
NH4INITB
NH4INITW
NO3INITB
NO3INITW
NSYIELDB
NSYIELDW
ONPARTF
PONCOUT
PONFALL
PONINITB
PONINITW
PONRES
PONSIZE
REFNINIT
RESTHN (NIT)
VEGETATION
BIODENS
BIOINIT
PBIOUW
PEATACRB
PEATACRW
PEATDENS
Base Value
NH4
NO3
DON
PON
DO
BOD5
T SS
DP
TP
0.01
0.05
0.01
0.05
240000
25000
0.15
0.15
50
50
0.15
0.5
0.00125
0.001
1
1
0.002
0.002
2400
1000
0.005
0.005
1
1
(NC)/(+E)
(+D)/(+D)
(+D)/(+D)
(+B)/(+B)
(NC)/(+E)
(+D)/(+D)
(-C)/(-C)
(-B)/(-B)
(-C)/(-D)
(-D)/(-D)
(-D)/(NC)
(-C)/(-C)
(-D)/(-E)
(-C)/(-C)
(-C)/(-C)
(-B)/(-B)
(-C)/(-C)
(-B)/(-B)
(+D)/(+D)
(+C)/(+C)
(+D)/(+D)
(+C)/(+C)
(-C)/(-C)
(-B)/(-B)
(NC)/(-E)
(-C)/(-C)
(-D)/(-D)
(-C)/(-C)
(NC)/(-E)
(-C)/(-C)
(+D)/(+D)
(+B)/(+B)
(+C)/(+C)
(+C)/(+C)
(-C)/(NC)
(-C)/(-C)
(+D)/(+D)
(+B)/(+B)
(-C)/(-C)
(-C)/(-C)
(-C)/(-C)
(-C)/(-C)
(+D)/(+D)
(+D)/(+D)
(+D)/(+D)
(+D)/(+D)
(-C)/(-C)
(-C)/(-C)
(NC)/(-D)
(-D)/(-C)
(-B)/(-B)
(-B)/(-B)
(NC)/(-C)
(-C)/(-C)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+C)/(+D)
(+C)/(NC)
(+B)/(+B)
(+B)/(+C)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(NC)/(-C)
(-C)/(-C)
(-B)/(-B)
(-B)/(-B)
(NC)/(-C)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+C)/(+C)
(+C)/(NC)
(+B)/(+B)
(+B)/(+C)
(+C)/(+C)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(NC)/(-E)
(-D)/(-D)
(-C)/(-C)
(-B)/(-B)
(NC)/(-E)
(-D)/(-D)
(+B)/(+B)
(+B)/(+B)
(+C)/(+C)
(+D)/(+D)
(+D)/(NC)
(+C)/(+C)
(+D)/(+D)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(+C)/(+B)
(-D)/(-D)
(-C)/(-C)
(-D)/(-D)
(-D)/(-D)
(+C)/(+(C)
(+C)/(+C)
(NC)/(-C)
(-D)/(-C)
(-B)/(-B)
(-B)/(-B)
(NC)/(-C)
(-C)/(-C)
(+B)
(+B)/(+B)
(+B)/(+B)
(+C)/(+C)
(+C)/(NC)
(+B)/(+B)
(+B)/(+C)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-C)/(-B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/(-D)
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
(-D)/(+D)
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
A/A
(-D)/(+D)
23.5
95
10000
5000
3.29
3.29
9900000
700000
0.125
0.00002
0.00006
0.00006
55000
30000
4000
400
0.3
0.3
0.6
0.75
0.25
300000
15000
0.01
0.05
500000
0.01
(+E)/(-D)
(+E)/(+E)
(+B)/(+C)
(-B)/(+C)
(+C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(-D)
(-B)/(+B)
(-C)/(-C)
(+C)/(+C)
(+E)/(+E)
(+D)/(+D)
(+C)/(-E)
(+B)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(-C)
(-C)/(-C)
(+C)/(+C)
(+E)/(-D)
(+E)/(-E)
(-C)/(+C)
(-C)/(+C)
(+E)/(-E)
(-C)/(+C)
(+F)/(-B)
(+D)/(+D)
(-B)/(+B)
(-B)/(+B)
(+D)/(+D)
(+C)/(+C)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+C)/(+D)
(+C)/(+C)
(+B)/(+C)
(+D)/(-F)
(+D)/(+D)
(-D)/(-D)
(-D)/(-D)
(+B)/(+B)
(-B)/(-B)
(-B)/(+B)
(+F)/(-F)
(+F)/(-F)
(-B)/(+B)
(-B)/(+B)
(+F)/(-F)
(-B)/(+B)
(-C)/(-C)
(-B)/(-B)
(+D)/(+D)
(+D)/(+D)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+C)/(+C)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+C)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(-F)/(-F)
(-B)/(+B)
(-B)/(+B)
(-D)/(+C)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-D)/(-E)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(+D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(+E)/(+E)
(+E)/(+E)
(-E)/(-F)
(-D)/(+D)
(+D)/(+D)
(+D)/(+D)
(-D)/(+D)
(-C)/(+C)
(+D)/(+D)
(-C)/(-B)
(-C)/(-D)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-C)/(-C)
(-C)/(-C)
(-B)/(+C)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(-B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(-B)/(+B)
(-C)/(+C)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
A/A
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
A/A
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B/(+B)
(-B)/(+B)
5000
8635000
0.4
2000
300
0.7
(+B)/(-B)
A/(-B)
(-B)/(-B) (-B)/(-B) (-B)/(-B)
(-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)
(-B)/(-B)
(-B)/(-B) (+B)/(+B) (+D)/(+D) (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B)
(+B)/(+B) (+B)/A
(+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B)
(+C)/(+C) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B)
(+B)/(+B) (+B)/(+B) (-B)/(-B) (-C)/(-C) (+B)/(+B) (-B)/(-B)
A/A
A/A
A/A
(-B)/(-B)
(-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)
(-B)/(-B) (+B)/(+C) (-B)/(-B) (-B)/(-B)
246
Base Value
0.7
PRATEUP
0.4
PSTDUW
6000000
STANDIN
5000
STDDENS
CARBON
0.47
BIOCCONT
0.8
BODCFRAC *
0.5
BODPFRAC
60000
DOCINITB
50000
DOCINITW
0.01
LEACHR
0.53
MICROBEC
0.00004
MTCDOC
0.8
PEATCC *
0.3
POCCOUT *
0.45
POCFALL
3000000
POCINITB
200000
POCINITW
0.001
POCRES
0.2
POCSIZE
15000000
REFCINIT
DISSOLVED OXYGEN
0.001
DOCONCP
15000
DOINITB
15000
DOINITW
8.5
DOXYCSAT
0.15
HTDOYB
0.15
HTDOYW
0.0001
MTDOX
0.00008
MTFW SDOC
0.2
NSDOYB
0.2
NSDOYW
PHOSPHOROUS
300
BIOMPP
250000
BTPHOSI
40000
DTPHOSIB
38000
DTPHOSIW
0.00006
MTCPHOS
0.05
PMINPPC
0.00005
PRMINBPC
0.5
PSEDDEP
SEDIMENT
0.1
DECOMPR
2.5
MANNC (SED)
0.5
PSEDDEP
0.0011
RESTHICK (SED)
0.3
SEDFALL1
0.7
SEDFALL2
26000000
SEDINITB1
10000000
SEDINITB2
190000
SEDINITW 1
10000
SEDINITW 2
0.1
SEDRES
0.25
SEDSIZE1
0.5
SEDSIZE2
1.1
SEDSPG1
2.65
SEDSPG2
NH4
NO3
DON
PON
DO
BOD5
T SS
DP
TP
(-C)/(-C)
(-C)/(-C)
(+C)/(-C)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(+B)/(+B)
(+B)/(+C)
(-C)/(-C)
(-B)/(-B)
(+B)/(+B)
(-B)/(+B)
(+B)/(+B)
(+B)/(+B)
(+B)/(+B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+B)
(+C)/(+C)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(-B)
(+B)/(+C)
(+B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+D)/(+E)
(-D)/(+C)
(-D)/(+D)
(-C)/(+C)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+C)
(-D)/(+D)
(-D)/(+D)
(-C)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(-D)
(+C)/(+C)
(-C)/(-C)
(-D)/(-D)
(+D)/(-D)
(-C)/(-C)
(-C)/(-C)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(+B)/(+B)
(+B)/(+C)
(-C)/(-C)
(-C)/(-C)
(+B)/(-B)
(-B)/(-B)
(-B)/(+B)
(+D)/(+D)
(-E)/(-E)
(-E)/(-E)
(+D)/(+D)
(+D)/(+D)
(+D)/(+C)
(+B)/(+B)
(+C)/(+C)
(+B)/(+B)
(+E)/(+E)
(-E)/(-E)
(+B)/(+C)
(+C)/(+C)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-D)/(-D)
(-C)/(-C)
(-E)/(NC)
(-E)/(NC)
(-C)/(-D)
(-E)/(NC)
(-D)/(-D)
(+D)/(+D)
(-C)/(-C)
(-B)/(+B)
(+C)/(+C)
(+C)/(+C)
(+D)/(NC)
(+E)/(NC)
(+C)/(+C)
(+D)/(NC)
(+D)/(+D)
(+C)/(+C)
(+B)/(+C)
(-B/(+B)
(+C)/(+C)
(+B)/(+C)
(+C)/(NC)
(+C)/(NC)
(+B)/(+C)
(+C)/(NC)
(+C)/(+C)
(+B)/(+C)
(+B)/(+B)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(NC)
(-D)/(NC)
(-D)/(+D)
(-D)/(NC)
(-D)/(+D)
(-D)/(+D)
(-C)/(+C)
(+B)/(+B)
(+D)/(+D)
(+D)/(+D)
(+E)/(NC)
(+E)/(NC)
(+D)/(+D)
(-C)/(NC)
(-D)/(-D)
(+D)/(+D)
(+C)/(+D)
(-B)/(+B)
(+C)/(+C)
(+C)/(+C)
(+C)/(NC)
(+C)/(NC)
(+B)/(NC)
(+B)/(NC)
(+C)/(+C)
(+B)/(+C)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-C)/(+C)
(+C)/(+C)
(-B)/(+C)
(-B)/(+C)
(-C)/(+C)
(-C)/)(+C)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-D)/(+D)
(-C)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-B)/(+B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-F)/(-F)
(+D)/(+D)
(+C)/(+C)
(+E)/(+E)
(+E)/(+E)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(-C)/(-C)
(+D)/(+D)
(+C)/(+C)
(+E)/(+E)
(+E)/(+E)
(+C)/(+C)
(+C)/(+C)
(+C)/(+C)
(-C)(-C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-C)/(+C)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-D)/(+D)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(+B)
(-B)/((+B)
(-B)/((+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/((+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(-B)/(+B)
(-B)/(+E)
(-F)/(-F)
(+E)/(+E)
(-F)/(-F)
(-B)/(-F)
(+E)/(+E)
(+B)/(+B)
(+E)/(+E)
(-B)/(+B)
(+E)/(+E)
(-B)/(+E)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(-B))/(+C)
(-C)/(-C)
(+C)/(+C)
(-C)/(-C)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(+B)
(+B)/(+B)
(-B)/(-B)
(-B)/(+C)
(-C)/(-C)
(+C)/(+C)
(-C)/(-C)
(-B)/(-B)
(-B)/(+B)
(-B)/(-B)
(+B)/(+B)
(-B)/(-B)
(+C)/(+C)
(-B)/(+B)
(+B)/(+B)
(-B)/(-B)
(+B)/(+B)
- Results presented as the RS values for the +10% change in base value followed by the 10% change.
- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.
- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible
results)
247
VITA
Erik Ryan Lee
Erik Ryan Lee was born and mostly raised his entire life in San Francisco, California.
He attended Lowell High School in San Francisco, where he was involved in government, choir,
and track. He then attended the University of California at Davis and received his bachelor of
science in Biological Systems Engineering with an emphasis on ecology in 1997.
He made a drastic change by moving to what he considers the country in Virginia, and for
the past two years has attended Virginia Polytechnic Institute and State University. Here he has
learned to appreciate research, having seasons, and carrying an umbrella. In September of 1999,
he finished his thesis after taking a couple of wrong turns, and realizing that there is a
tremendous amount of research that needs to be done.
At the dawn of a new millenium, his future plans include seeing the world, living a full
and content life, and cleaning his native San Francisco Bay. The past is in the rear view mirror,
only the future lies ahead.
The ultimate measure of a man is not where he stands in moments of comfort and convenience
but where he stands in times of challenge and controversy.
- Martin Luther King Jr.
248