United States
       Environmental Protection
       Agency
Office of Water (4305)
EPA-820-R-12-015
  August 2012
&EPA     AQUATOX (RELEASE 3.1)

         MODELING ENVIRONMENTAL FATE
           AND ECOLOGICAL EFFECTS IN
              AQUATIC ECOSYSTEMS
        VOLUME 2: TECHNICAL DOCUMENTATION

-------
{This Page Left Blank, Back of Cover}

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
         AQUATOX (RELEASE 3.1)

           MODELING ENVIRONMENTAL FATE
            AND ECOLOGICAL EFFECTS IN
               AQUATIC ECOSYSTEMS
       VOLUME 2: TECHNICAL DOCUMENTATION

                   Richard A. Park1
                        and
                 Jonathan S. Clough2
                     AUGUST 2012

           U.S. ENVIRONMENTAL PROTECTION AGENCY
                   OFFICE OF WATER
             OFFICE OF SCIENCE AND TECHNOLOGY
                  WASHINGTON DC 20460
                 Modeling, Diamondhead MS 39525
           2Warren Pinnacle Consulting, Inc., Warren VT 05674

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                   DISCLAIMER

This document describes the scientific and technical background of the aquatic ecosystem model
AQUATOX, Release 3.1.  Anticipated users of this document include persons who are interested
in using the model, including but not limited to researchers and regulators.  The model described
in this document is not required, and the document does not change any legal requirements or
impose legally binding requirements  on  EPA, states, tribes or the regulated community.  This
document has been approved for publication by the Office of Science and Technology, Office of
Water, U.S. Environmental Protection Agency. Mention of trade names, commercial products or
organizations does not imply endorsement or recommendation for use.
                              ACKNOWLEDGMENTS

This model has been developed and documented by Richard A. Park of Eco Modeling and by
Jonathan S. Clough of Warren Pinnacle Consulting, Inc. under subcontract to Eco Modeling.
The  work was  funded with Federal funds from the U.S.  Environmental Protection Agency,
Office  of Science and Technology under  contract numbers 68-C-01-0037 to AQUA TERRA
Consultants, Anthony Donigian, Work Assignment Leader; and EP-C-12-006 to Horsley Witten
Group, Inc., Nigel Pickering, Work Assignment Leader. Integration of Interspecies Correlation
Estimation (Web-ICE) was made possible  due to the work of US.  EPA Office of Research and
Development Gulf Breeze, the University of Missouri-Columbia, and the US Geological Survey.

The assistance, advice, and comments of the EPA work assignment manager,  Marjorie Coombs
Wellman of the Standards  and Health Protection Division, Office of Science and Technology
have been of great  value in developing this model and preparing this report.  Further technical
and financial support from David A. Mauriello, Rufus Morison, and Donald Rodier of the Office
of Pollution Prevention and Toxics is gratefully acknowledged.  Marietta Echeverria, Office of
Pesticide Program,  contributed to the integrity of the  model through her careful analysis and
comparison with EXAMS. Amy Polaczyk and Marco Propato,  Warren  Pinnacle Consulting,
made valuable contributions  to the  model formulation and testing. Release 2 of this  model
underwent independent peer review by Donald DeAngelis, Robert Pastorok,  and Frieda Taub;
and Release 3  underwent peer review by  Marty Matlock, Damian Preziosi,  and Frieda Taub.
Their diligence is greatly  appreciated.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	



                             TABLE OF CONTENTS

DISCLAIMER	ii

TABLE OF CONTENTS	iii

PREFACE	viii

1.  INTRODUCTION                                                           1
      1.1  Overview	1
      1.2  Background	4
      1.3 The Multi-Segment Version	6
      1.4 The Estuary Model	6
      1.5 The PFA Model	7
      1.6 AQUATOX Release 3.1 Overview                                       7
      1.7 Comparison with Other Models	9
      1.8 Intended Application of AQUATOX                                     10

2.  SIMULATION MODELING                                                  12
      2.1 Temporal and Spatial Resolution and Numerical Stability	12
      2.2 Results Reporting	15
      2.3 Input Data	16
      2.4 Sensitivity Analysis	17
      2.5 Uncertainty Analysis	19
      2.6 Calibration and Validation                                            24

3.  PHYSICAL CHARACTERISTICS                                           41
      3.1  Morphometry	41
            Volume	41
            Bathymetric Approximations	44
            Dynamic Mean Depth	47
            Habitat Disaggregation                                            47
      3.2 Velocity	48
      3.3 Washout	49
      3.4  Stratification and Mixing	50
            Modeling Reservoirs and Stratification Options                        54
      3.5  Temperature	55
      3.6  Light	56
            Hourly Light	58
      3.7  Wind	59
      3.8  Multi-Segment Model                                                60
            Stratification and the Multi-Segment Model                           62
            State Variable Movement in the Multi-Segment Model	62

4.  BIOTA	64
      4.1  Algae	65

                                    iii

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	


            Light Limitation	69
            Adaptive Light	74
            Nutrient Limitation	75
            Current Limitation	77
            Adjustment for Suboptimal Temperature	78
            Algal Respiration	80
            Photorespiration	81
            Algal Mortality	82
            Sinking	83
            Washout and Sloughing	84
            Detrital Accumulation in Periphyton	89
            Chlorophyll a	89
            Phytoplankton and Zooplankton Residence Time	90
            Periphyton-Phytoplankton Link	91
      4.2 Macrophytes	92
      4.3 Animals	96
            Consumption, Defecation, Predation, and Fishing	98
            Respiration	102
            Excretion	105
            Nonpredatory Mortality	106
            Suspended Sediment Effects	107
            Gamete Loss and Recruitment	123
            Washout and Drift	125
            Vertical Migration	126
            Migration Across Segments	127
            Anadromous Migration Model	128
            Promotion and Emergence	129
      4.4 Aquatic Dependent Vertebrates	130
      4.5 Steinhaus Similarity Index	130
      4.6 Biological Metrics	131
            Invertebrate Biotic Indices	135

5.  REMINERALIZATION                                                     137
      5.1 Detritus	137
            Detrital Formation	141
            Colonization	142
            Decomposition	144
            Sedimentation	147
      5.2 Nitrogen	150
            Assimilation	152
            Nitrification and Denitrification	153
            lonization of Ammonia	155
            Ammonia Toxicity	157
      5.3 Phosphorus	158
      5.4 Nutrient Mass Balance	160
            Variable Stoichiometry	160
            Nutrient Loading Variables                                         161
                                     iv

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	


            Nutrient Output Variables	161
            Mass Balance of Nutrients	162
      5.5 Dissolved Oxygen	169
            Diel Oxygen	173
            Lethal Effects due to Low Oxygen	174
            Non-Lethal Effects due to Low Oxygen                               180
      5.6 Inorganic Carbon	181
      5.7 Modeling Dynamic pH                                                184
      5.8 Modeling Calcium Carbonate Precipitation and Effects	187

6.  INORGANIC SEDIMENTS                                                  189
      6.1 Sand Silt Clay Model                                                  189
            Deposition and Scour of Silt and Clay	191
            Scour, Deposition and Transport of Sand	194
            Suspended Inorganic Sediments in Standing Water                    196
      6.2 Multi-Layer Sediment Model                                          197
            Suspended Inorganic Sediments	199
            Inorganics in the Sediment Bed	199
            Detritus in the Sediment Bed	201
            Pore Waters in the Sediment Bed	201
            Dissolved Organic Matter within Pore Waters	202
            Diffusion within Pore Waters	203
            Sediment Interactions	204

7.  SEDIMENT DIAGENESIS	207
      7.1 Sediment Fluxes	209
      7.2 POC	212
      7.3 PON	214
      7.4 POP	214
      7.5 Ammonia	214
      7.6 Nitrate	216
      7.7 Orthophosphate	217
      7.8 Methane	218
      7.9Sulfide	220
      7.10 Biogenic Silica	221
      7.11 Dissolved Silica	222

8.  TOXIC ORGANIC CHEMICALS                                             224
      8.1 lonization	231
      8.2 Hydrolysis	232
      8.3 Photolysis	234
      8.4Microbial Degradation                                                236
      8.5 Volatilization	237
      8.6 Partition Coefficients                                                  240
            Detritus	240
            Algae 	243
            Macrophytes	244

                                     v

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	


             Invertebrates	245
             Fish   	245
      8.7 Nonequilibrium Kinetics	246
             Sorption and Desorption to Detritus	247
             Bioconcentration in Macrophytes and Algae	248
                   Macrophytes	248
                   Algae 	249
             Bioaccumulation in Animals	252
                   Gill Sorption	252
                   Dietary Uptake	255
                   Elimination	256
             Bioaccumulation Factor	260
             Linkages to Detrital Compartments	261
      8.8 Alternative Uptake Model: Entering BCFs, Kl, and K2                    261
      8.9 Half-Life Calculation, DT50 and DT95                                   262
      8.10 Chemical Sorption to Sediments	263
      8.11 Chemicals in Pore Waters	265
      8.12 Mass Balance Capabilities and Testing                                 267
      8.13 Perfluoroalkylated Surfactants Submodel                               269
             Sorption	269
             Biotransformation and Other Fate Processes	269
             Bioaccumulation	269
                   Gill Uptake	270
                   Dietary Assimilation	271
                   Depuration	272
                   Bioconcentration Factors                                     273

9.  ECOTOXICOLOGY                                                         275
      9.1 Lethal Toxicity of Compounds	275
             Interspecies Correlation Estimates (ICE)	275
             Internal Calculations	277
      9.2 Sublethal Toxicity	280
      9.3 External Toxicity	283

10. ESTUARINE SUBMODEL                                                  286
      10.1 Estuarine Stratification                                               286
      10.2 Tidal Amplitude	287
      10.3 Water Balance	288
      10.4 Estuarine Exchange	289
      10.5 Salinity Effects	290
             Mortality and Gamete Loss                                          290
             Other Biotic Processes	290
             Sinking	291
             Sorption	293
             Volatilization	293
             Reaeration	293
             Migration	295

                                     vi

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	






     10.6 Nutrient Inputs to Lower Layer	295




11. REFERENCES	296




APPENDIX A. GLOSSARY OF TERMS                                   315




APPENDIX B. USER SUPPLIED PARAMETERS AND DATA                  318
                               vn

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                      PREFACE

The Clean Water Act- formally the Federal Water Pollution Control Act Amendments of 1972
(Public Law 92-50), and subsequent amendments in 1977, 1979,  1980,  1981, 1983, and 1987-
calls for the  identification, control, and prevention of pollution of the nation's waters. Data
submitted by the States  to the U.S. Environmental  Protection Agency's WATERS (Watershed
Assessment,  Tracking  & Environmental  Results)  database  (http://www.epa.gov/waters/)
indicate that a very high percentage of the Nations waters continue to be impaired.  As of early
2009, of the waters that have been assessed,  44% of rivers and streams, 59% of lakes, reservoirs
and ponds,  and 35% of estuaries were impaired  for one or more of their designated uses.  The
five most commonly reported causes of impairment in rivers and  streams were:  pathogens,
sediment, nutrients, habitat alteration and organic enrichment/dissolved oxygen depletion.  In
lakes  and  reservoirs  the five  most common  causes  were  mercury,  nutrients,  organic
enrichment/dissolved oxygen  depletion, metals, and  turbidity.   In  estuaries  the five most
common causes were pathogens, mercury, organic enrichment/oxygen depletion, pesticides and
toxic organics.  Many waters are impaired for multiple uses, by multiple causes, from multiple
sources.

New approaches and tools, including appropriate technical guidance  documents, are needed to
facilitate ecosystem analyses of watersheds  as required by the Clean Water Act. In particular,
there  is a pressing  need for  refinement and release of an ecological  risk methodology that
addresses the direct,  indirect,  and  synergistic effects  of nutrients,  metals, toxic organic
chemicals, and  non-chemical  stressors on aquatic ecosystems, including streams, rivers, lakes,
and estuaries.

The  ecosystem  model  AQUATOX is one of  the few general ecological risk  models that
represents the combined environmental fate and effects of toxic chemicals.  The model also
represents conventional pollutants, such as nutrients  and sediments, and considers several trophic
levels, including attached and planktonic algae,  submerged aquatic vegetation,  several types of
invertebrates, and several types of fish. It has been implemented for experimental tanks, ponds
and pond enclosures, streams, small  rivers,  linked river segments,  lakes, reservoirs,  linked
reservoir segments, and estuaries.
                                       Vlll

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1



                                  1. INTRODUCTION

1.1 Overview

The  AQUATOX model  is  a general  ecological risk  assessment model  that represents  the
combined environmental fate and effects of conventional pollutants,  such as nutrients and
sediments,  and toxic chemicals in aquatic  ecosystems. It  considers  several trophic  levels,
including attached and planktonic algae and submerged aquatic vegetation, invertebrates, and
forage, bottom-feeding, and game fish; it also represents associated organic toxicants (Figure 1).
It can be implemented as a simple model (indeed, it has been used to simulate an abiotic flask) or
as a truly complex food-web model.  Often it is desirable to model a food web rather than a food
chain, for example to examine the possibility of less tolerant organisms being replaced by more
tolerant organisms as environmental perturbations occur.  "Food web models provide a means
for validation because they mechanistically describe  the bioaccumulation process and ascribe
causality to observed relationships between biota and sediment or water" (Connolly and Glaser
1998).  The best way to accurately assess bioaccumulation is to use more complex models, but
only if the data needs of the models can be met and there is sufficient time (Pelka 1998).

It  has been implemented for experimental tanks, ponds and pond enclosures, streams,  small
rivers, linked river segments, lakes, reservoirs, linked reservoir segments, and estuaries.  It is
intended to be used to evaluate the likelihood of past, present, and future adverse effects from
various   stressors including  potentially toxic  organic  chemicals,  nutrients,  organic  wastes,
sediments, and temperature. The stressors may be considered individually or together.

The  fate portion  of  the model, which is applicable  especially to organic  toxicants, includes:
partitioning among organisms, suspended  and sedimented detritus, suspended and sedimented
inorganic sediments,  and water; volatilization; hydrolysis; photolysis;  ionization; and microbial
degradation.  The effects  portion of the model  includes:  sublethal and lethal toxicity to the
various  organisms modeled;  and indirect effects  such as release of  grazing and predation
pressure, increase in  detritus and recycling of nutrients from killed organisms, dissolved oxygen
sag due  to increased decomposition, and loss of food base for animals.

AQUATOX represents the aquatic  ecosystem by simulating the  changing concentrations (in
mg/L or g/m3)  of organisms,  nutrients, chemicals, and sediments  in a unit volume  of water
(Figure  1).  As such, it differs from population models, which represent the changes in numbers
of individuals. As O'Neill et al. (1986) stated,  ecosystem models  and  population models are
complementary; one  cannot take the place of the other.  Population models excel at modeling
individual species at  risk and modeling fishing pressure  and other age/size-specific aspects;  but
recycling of nutrients,  the   combined  fate  and  effects  of toxic  chemicals,  and  other
interdependences in the aquatic ecosystem are important aspects that AQUATOX represents and
that cannot be addressed by a population model.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 1
               Figure 1. Conceptual model of ecosystem represented by AQUATOX
                                                             Suspended and
                                                            bedded sedim ents
                                                                    Settling, scour
                                                                 Plants
                                                               Phytoplankton
                                                               Attached algae
                                                               M acrophytes
        o
        *=
        O.
Any ecosystem model consists of multiple  components requiring  input data.   These are the
abiotic and biotic state variables or compartments being simulated (Figure 2).  In AQUATOX
the biotic state variables may represent trophic levels, guilds, and/or species.  The model can
represent a food web with both detrital-  and algal-based trophic linkages. Closely related are
driving variables, such as temperature, light, and nutrient loadings, which force the system to
behave in certain ways.  In AQUATOX state variables and driving variables are treated similarly
in the code.   This  provides flexibility because external loadings  of state variables,  such as
phytoplankton carried into a reach from upstream, may function as driving variables; and driving
variables, such  as  temperature, could  be  treated  as dynamic state variables in  a future
implementation.   Constant,  dynamic,  and  multiplicative loadings  can  be specified  for
atmospheric, point- and  nonpoint sources.  Loadings of pollutants can be turned off at the click
of a button to obtain a control simulation for comparison with the perturbed simulation.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                               CHAPTER 1
         Figure 2. State variables in AQUATOX as implemented for Cahaba River, Alabama.
                                       Forage Fish
                                         shiner,
                                         bluegill
Bottom Fish
stoneroller
                                                      Piscivore
                                                        bass
       Zoobenthos
         grazers
         mayfly,
       riffle beetle
              Zoobenthos
              susp. feeders
               caddisfly
                                                                   Predatory
                                                                  Invertebrate
                                                                    crayfish,
                                                                    stonefly
                                             Zoobenthos
                                              molluscs
                                            snail, mussel
                                              Corbicula
                          Penphyton diatom,
                               green
                                  Phytoplankton
                                     diatom,
                                      green
                                                                  Macrophyte
                                                                     moss
  Periphyton
  blue-green
                                        Nitrate & Nitrite
                                          Carbon Dioxide
                          Labile
                       Diss. Detritus
                                            Refractory
                                           Susp. Detritus
                                                                       Labile
                                                                    Susp. Detritus
 Refractory
Diss. Detritus
                          Labile
                       Sed. Detritus
                          Buried Refrac
                           Sed. Detritus
 Refractory
Sed. Detritus
                                                                      Sediments
                                                                    sand, silt, clay
                                                                       (or TSS)
The  model  is  written  in  object-oriented  Pascal using the Delphi  programming  system for
Windows.  An object is a  unit of computer code that can be duplicated; its characteristics and
methods also can be inherited by higher-level  objects.  For  example, the organism object,
including variables such as the LC50 (lethal concentration of a toxicant) and process functions
such as respiration, is inherited by the plant object; that is enhanced by plant-specific variables
and functions and is duplicated for four kinds of algae; and the plant object is inherited and
modified slightly for macrophytes  and moss.  This modularity forms the basis for the remarkable
flexibility of the model,  including the ability to add and delete given state variables interactively.

AQUATOX  utilizes differential  equations to  represent changing values  of state variables,
normally with a reporting time step of one day.  These equations require starting values or initial
conditions for the beginning of the simulation. If the first day of a simulation is changed, then
the initial conditions may need to be changed.  A  simulation can begin with any date  and may be
for any length of time from a few days, corresponding to a microcosm experiment,  to decades,
corresponding to an extreme event followed by long-term recovery.

The process  equations  contain another class of input variables: the parameters or coefficients
that  allow the  user to specify  key  process characteristics.   For  example,  the  maximum
consumption rate is a  critical  parameter characterizing various consumers. AQUATOX is a

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 1
mechanistic model with many parameters;  however, default values  are available so that the
analyst only has to be concerned with those parameters necessary for a specific risk analysis,
such as characterization of a new chemical.  In the pages that follow, differential equations for
the state variables will be followed by process equations and parameter definitions.

Finally,  the system  being  modeled is characterized by  site constants,  such as mean and
maximum depths.  At present one can model lakes, reservoirs, streams, small rivers, estuaries,
and ponds- and even enclosures and tanks.  The "Generalized Parameter" screen is used for all
these site types, although some, such as the  hypolimnion and estuary  entries, obviously are not
applicable to all.  The temperature and light constants  are used for  simple forcing  functions,
blurring the distinctions between site constants and driving variables.

                  Table 1.  Model Overview Summary (also see Section 2.1)
Category:
Reporting Time Step
Differentiation
Output Averaging
Conceptual Approach
Horizontal Spatial
Resolution
Vertical Spatial
Resolution
Sediment Bed
Boundary Conditions
Ecological Complexity
Chemical Complexity
Mass Balance Tracking
Summary:
Daily or Hourly
Variable time-step Runge Kutta
(with fixed time-step option)
Variable
Kinetic; biomass model
Point model, or ID and 2D with
linked segments
Vertically stratified water
column when relevant
Multiple sediment bed options
Inflows and outflows of all state
variables (dissolved oxygen,
nutrients, biota, detritus, and
toxic organics)
Variable— user can model
representative groups or
individual species
Zero to 20 organic chemicals
For nutrients and chemicals
Notes:
time-step over which equations are solved
smaller step sizes than the reporting time-
step may be utilized to reduce relative error
editable by user
no longer a fugacity option for chemicals;
individual organisms are not modeled
modeled units can be a lake, river, reservoir,
stream segment, estuary, or enclosure
user-specified or model calculated dates of
stratification
active layer only, multi-layer sediments,
sediment diagenesis submodels
water inflow, point sources, nonpoint
sources, direct precipitation, separate
tributary inputs
can model abiotic conditions or single
macrophyte species in a water tank up to
dozens of plant and animal species in a
complex river or reservoir system
biotransformation to daughter products
may be modeled

1.2 Background

AQUATOX 3.1 is an update to AQUATOX Release 3 (see section 1.6 below for a list of updates
in Release 3.1). AQUATOX Release 3 was the result of an effort to combine all of the various
versions of  AQUATOX into a single consolidated version.   Models that were combined to
produce Release 3 included:
       AQUATOX Multi-Segment version
       AQUATOX Estuarine Version

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1


   .   AQUATOX PFA Model (Perfloroalkylated Surfactants)

Each of these versions is discussed in a separate section below.

AQUATOX is the latest in a long series of models, starting with the aquatic ecosystem model
CLEAN  (Park et al.,  1974)  and  subsequently  improved in  consultation with numerous
researchers at various European hydrobiological laboratories, resulting in the CLEANER series
(Park et al.,  1975, 1979, 1980; Park, 1978; Scavia and Park, 1976) and LAKETRACE (Collins
and Park, 1989).  The MACROPHYTE model, developed for the U.S. Army Corps of Engineers
(Collins  et  al.,  1985),  provided additional capability for representing  submersed  aquatic
vegetation.   Another series started with the toxic fate model PEST,  developed to complement
CLEANER (Park et al., 1980, 1982), and continued with the TOXTRACE model (Park, 1984)
and the spreadsheet equilibrium fugacity PART model.  AQUATOX combined algorithms from
these models with ecotoxicological constructs; and additional code was written as required for a
truly integrative fate and effects model (Park, 1990, 1993). The model was then restructured and
linked to Microsoft Windows interfaces to provide greater flexibility, capacity  for additional
compartments, and user friendliness (Park et al.,  1995). The current version has been improved
with the addition of constructs  for  sublethal effects  and  uncertainty  analysis, making  it a
powerful tool for probabilistic risk assessment.

This technical documentation is  intended to provide verification of individual constructs or
mathematical and programming formulations used within AQUATOX.  The scientific basis of
the constructs reflects empirical and theoretical support; precedence in the open literature and in
widely used models is  noted.   Units  are  given  to  confirm  the dimensional  analysis.  The
mathematical formulations have been programmed and graphed in spreadsheets and the results
have been  evaluated in terms of behavior consistent with our understanding  of ecosystem
response; many of those graphs are given in the following documentation.  The variable names in
the documentation  correspond  to  those used  in the program so  that  the  mathematical
formulations  and code  can be  compared,  and  the  computer  code  has been checked  for
consistency with those formulations.   Much of this has been done as  part of the continuing
process of internal review. Releases 2 and 3 of the AQUATOX model and documentation have
undergone successful peer reviews by  external  panels convened by the U.S. Environmental
Protection Agency.  Release 3 has also been described in the peer-reviewed literature (Park et al.
2008).

Release 3 has significant additional capabilities compared to Release 2.2:

   •   Link to WEB-ICE (Interspecies Correlation Estimates) database and graphics
       Sediment diagenesis based on the Di Toro  model
       Optional hourly time step with diel oxygen, light, and photosynthesis;
   •   Low oxygen effects
       Toxicity due to ammonia
       Suspended and bedded sediment effects on organisms; % embeddedness
       Calcium carbonate precipitation and removal of phosphorus
   •   Adaptive light limitation for plants

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1


   •   Linked periphyton and phytoplankton compartments
   •   Conversions for many units in input screens
   •   User-specified seasonally varying thermocline depth
   •   User-specified reaeration constant in addition to alternative estimation procedures
   •   Improved CBOD to organic matter estimation
   •   Estuarine reaeration incorporating salinity
   •   Sensitivity analysis with tornado diagrams
   •   Correlation of variables in uncertainty analysis
   •   Sediment oxygen demand (SOD) output
   •   Enhanced graphics including log  plot, duration and exceedance graphs, and threshold
       analysis
       Option to export all graphs to Microsoft Word
       Output of statistics for all graphed model results
       Output of trophic  state  indices and  ecosystem bioenergetics such as gross primary
       productivity and community respiration
   •   Integrated user's manual and context-sensitive help files
1.3 The Multi-Segment Version

The AQUATOX Multi-Segment version was developed and applied for the EPA Office of Water
in support of the Modeling Study of PCB Contamination in the Housatonic River.  Capabilities
introduced with this version include the linkage of individual AQUATOX segments into a single
simulation. Segments can be linked together in a manner that allows feedback into the upstream
segment or a one-way "cascade" linkage can be created.  More information about the physical
characteristics of linked segments may be found in Section 3.8 of this document.

Additionally, a  sediment submodel was added to the AQUATOX model to enable tracing the
passage of toxicants  within a multi-layered sediment bed.  Specifications for this multi-layer
sediment model may be found in section 6.2 of this document.
1.4 The Estuarine Submodel

The  Risk Assessment Division (RAD),  EPA Office of Pollution Prevention  and Toxics,  is
responsible for assessing the human health and ecological risks of new and existing chemicals
that are regulated under the Toxic Substances Control Act (TSCA).  RAD has partially funded
AQUATOX from its initial conceptualization. Many of the industrial chemicals regulated under
TSCA are discharged into estuarine environments.

Therefore, AQUATOX's capabilities were enhanced by adding salinity and other components
(including shore birds) that would be needed to simulate an estuarine environment. The estuarine
version  of AQUATOX is intended to be an exploratory  model for evaluating the possible fate

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1


and effects of toxic chemicals and other pollutants in estuarine ecosystems.  The model is not
intended to represent detailed, spatially varying site-specific conditions, but rather to be used in
representing the potential behavior of chemicals under average conditions.  Therefore, it is best
used as a screening-level model applicable to data-poor evaluations in estuarine ecosystems.

Complete documentation for the AQUATOX estuarine submodel may be found in Chapter 10 of
this document.
1.5 The PFA Submodel

The bioaccumulation and effects of a group of chemicals known as perfluorinated surfactants has
been of recent interest. There are two  major types  of  perfluorinated surfactants:  perfluoro-
alkanesulfonates and  perfluorocarboxylates.   The perfluorinated compounds  of interest as
bioaccumulators are the perfluorinated acids (PFAs).  Perfluoroctane sulfonate (PFOS) belongs
to the sulfonate group and perfluorooctanoic  acid (PFOA) belongs to the carboxylate group.
These persistent chemicals have been found in humans,  fish, birds, and marine and terrestrial
mammals throughout the world. PFOS has an especially high bioconcentration factor in fish.  The
principal focus was on PFOS because of its prevalence and the availability of data.  Because both
chemical classes contain high- and low-chain homologs, AQUATOX will be useful in estimating
the fate  and effects of a wide  range of molecular weight  components where actual data are not
available for every homolog.

Complete documentation for the AQUATOX Perfloroalkylated Surfactants model may be found
in Section 8.13 of this document.

1.6 AQUATOX Release 3.1 Overview

Additional capabilities are available in Release 3.1 as compared to Release  3.  Some highlights
follow:

   •  Addition of sediment-diagenesis  "steady-state" mode to  significantly  increase model
      speed;
   •  Modification of denitrification code in  order to simplify calibration  and to achieve
      alignment with other models;
   •  Enabled importation of equilibrium CO2 concentrations to enable  linkage to CO2SYS
      and  similar models;
   •  New CBOD to organic matter conversion relying on percent-refractory detritus  input;
      Input and output BOD is clarified to be "carbonaceous" BOD.
   •  Floating plants refinements
          •  Added floating option for plants other than cyanobacteria  (formerly known as
             "blue-green algae)
          •  Converted the averaging depth for floating plants to the top three meters to more
             closely correspond to monitoring data
          •  Floating plants now explicitly move from the hypolimnion to  the epilimnion when

                                       7

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1

             a system is stratified.
   •   Modifications to PFA (perfluoroalkylated surfactants) model to increase flexibility:
          •  Uptake rates (Kls) and elimination rates (K2s) are visible and editable for animals
             and plants
          •  New interface to estimate animal Kls and K2s as a function of chain length
          •  Improved gill-uptake equation for invertebrates.
   •   Bioaccumulation and toxicity modeling improvements:
             Optional alternative elimination-rate estimation for  animals  based on Barber
             (2003);
          •  Updated ICE (toxicity  regressions), based on new EPA  models released  in
             February 2010 and improved AQUATOX ICE interface;
          •  Addition of output of Kl, K2, and BCF estimates.
   •   Improved sensitivity and uncertainty analyses
          •  "Output to CSV" option for uncertainty runs so that complete results for  every
             iteration may be examined;
          •  Option for non-random sampling for "statistical sensitivity analyses";
          •  A "reverse tornado" diagram (effects diagram) that  shows the  effects  of each
             parameter change on the overall simulation;
          •  Nominal  range  sensitivity analysis  has  been  added  for linked  segment
             applications.
   •   Database Improvements
          •  AQUATOX database search functions dramatically improved.
          •  "Scientific Name" field added to Animal and Plant databases.
   •   Interface and Data Input Improvements
             Software and software installer is 64-bit OS compatible;
          •  Added an option in the "Setup" screen to trigger nitrogen fixation based on  the N
             to P ratio.
          •  Addition of output variables to clarify whether photosynthesis is sub-optimal due
             to high-light or low-light conditions.
          •  Time-varying evaporation  option in the  "Site" screen with  linkage from  the
             "Water Volume" screen
             Grid mode within a study so that all animal, plant, and chemical  parameters in a
             study can be examined, edited, and then exported to Excel
          •  Added capability to input time-series loads of organisms based on fish stocking
          •  Updated HSPF WDM file linkage to  be  more generally  applicable  (does  not
             require use of WinHSPF).
          •  Enabled hourly loadings for the following variables:  all nutrients, CO2, Oxygen,
             Inorganic suspended sediments (sand/silt/clay), TSS, Light, Organic Matter
          •  "Graph Setup" window now enabled for linked-mode  graphics.
          •  Other minor interface improvements.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1
Documentation for each of these enhancements may be found in this technical documentation
volume or in the User's Manual.
1.7 Comparison with Other Models

The following comparison is taken from Park et al. (2008):

The model is perhaps the most comprehensive aquatic ecosystem simulation model available, as
can be seen by comparison  with other representative  dynamic models being used for risk
assessment  (Table 2).   All the models,  with the exception  of QSim and CASM, are public
domain. The closest to AQUATOX in terms of scope is  the family of CATS models developed
by Traas and others  (Traas  et  al.,  1996; Traas  et al., 1998;  Traas  et  al.,  2001);  these
ecotoxicology  models have  simple  representations of growth  and are not as  suitable as
AQUATOX for detailed analyses of eutrophication effects.  CASM (DeAngelis et al.,  1989;
Bartell et al., 1999) is similar to CATS, with simplified growth terms, but it lacks a toxicant fate
component.  QUAL2K (Chapra et al., 2007) and WASP (Di Toro et al., 1983; Wool et al., 2004)
are water quality models that share many functions with AQUATOX, including benthic algae
(Martin et al.,  2006); WASP also models fate of toxicants.  The hydraulic and water quality
models EFDC  (Tetra Tech Inc., 2002) and HEM3D (Park et al., 1995a) are often combined;
EFDC has also been used to provide the flow field for linked segments in AQUATOX, resulting
in a  similar representation. AQUATOX, QUAL2K, WASP,  and EFDC  include the  sediment
diagenesis model for remineralization (Di Toro, 2001). WASP  and the bioaccumulation model
QEAFdChn (Quantitative Environmental  Analysis, 2001) have been combined in the Green Bay
Mass Balance (GBMB) study (U.S. Environmental Protection Agency, 1989), which Koelmans
et al.  (2001) considered to be more accurate for portraying bioaccumulation than AQUATOX.
However, GBMB does not include an ecotoxicology component. BASS (Barber, 2001) is a very
detailed bioaccumulation  and  ecotoxicology   model;  it  provides better  resolution than
AQUATOX in modeling single species, but so far it has  only been applied to fish and does not
include ecosystem dynamics.  The German model QSim  (Schol et al., 1999; Schol et al., 2002;
see also Rode et al., 2007) has detailed ecosystem functions  and has been applied  in studying
impacts of both eutrophication and hydraulics on river ecosystems.   Similar to AQUATOX, it
has been used  to analyze relationships between  plankton and mussels and impacts of oxygen
depletion. Further comparison of models can be found in a book by Pastorok et al. (2002).

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                             CHAPTER 1
Table 2.   Comparison of AQUATOX with  other representative dynamic models used for risk
assessment (Park et al. 2008).
State variables and
processes.
Nutrients
Sediment diagenesis
Detritus
Dissolved sicyger.
DC effects on biota
PH
Ntti toxitity
Sand'silt/day
Sediment effects
Hydraulics
Heat budget
salinity
Phytopiaritton
Periphyton
Macrophytes
ZoaplanJtton
Zoobenthos
Fish
Bacteria
Pathogens
Organic toxicant late
Organic toxicants in
Sediments
Stratified sediments
Phyttiplanktan
Periphytosn
Macropbytes
Zooplafikton
Zoobenthos
Fish
Birds oc other animals
Eoatoxicity
Linked segments
AQUATOX
X
X
X
X
X
X
X.
X
X


X
X
X
X
X
X
X


X,

X
X
X
X
X
X
X
X
X
X
X
CATS CASM Qual2K WASP7 EFDC-HEM3D QEAFdChn
XXIX X
XX X
XXIX X
XXX X

I

X X

X
XX X
X X
X X X X X
XXX X
X 3C
X X
X X
X X
X
I X
X X

X XX
X X
X
X
X
X X
X X
X I
X
X X
XXX X
BASS QSim
X

J.
X
X
X



X
X

X
X
X
X
X
X X
X

X








X

X
X
1.8 Intended Application of AQUATOX

AQUATOX is intended to be used at any one of several levels of application. Like any model, it
is  best used as one of several tools in a weight-of-evidence approach.  The level of required
precision, rigor, data requirements  and user effort depend  upon the goals of the modeling
exercise and the potential consequences of the model results.

Perhaps its most widespread use is as  a screening-level model requiring few changes to default
studies and parameters.  In fact, it was originally developed as an evaluative model to assess the
fate and  effects of pesticides and industrial organic chemicals in representative or "canonical"
environments;   these  include  ponds  and  pond  enclosures,  experimental streams, and a
representative  estuary. It is especially useful in taking the place of expensive, labor-intensive
mesocosm  tests.  It  has been calibrated  and validated  with data  from  pond enclosures,
experimental streams, and a polluted harbor. In one early application, AQUATOX was driven
with predicted pesticide  runoff into  a farm pond adjacent to  a corn field using the field model
PRZM.  Also,  with little effort the  model can  provide insights into the potential impacts of
invasive  species and  the possible effects of control measures, such as pesticide application, on
the aquatic ecosystem.
In recent years
AQUATOX has been applied as part of the process of developing water quality
                         10

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 1


targets for nutrients,  and comparing model-derived values with  regional criteria developed
empirically.  This application has involved setting up the model and calibrating with available
data for rivers and reservoirs receiving nutrients from wastewater treatment plants, agricultural
runoff, and background  "natural" loadings.   It has been our experience that  this entails a
substantial level of effort, especially if the system is spatially heterogeneous, which then requires
application of linked segments.  A certain amount of site-specific biotic, water quality and flow
data is required, as well as pollutant loading data, for calibration.  However, once the model is
set up and calibrated for a site, it is relatively easy to represent a series of loading scenarios and
determine threshold nutrient levels for deleterious impacts such as nuisance algal  blooms and
anoxia.  This process is facilitated by the fact that the model has been calibrated across nutrient,
turbidity, and discharge gradients,  resulting in  robust parameter sets that span these conditions.
This is  important because the  intent of setting water quality  targets  is to model ecological
communities under changing conditions as a result of environmental management decisions; this
would give better assurance that the  sometimes costly nutrient  reduction actions would render
the desired environmental result.

The most intensive, time-consuming application of AQUATOX is in environmental remediation
projects,  such as SUPERFUND.   Because of the likely litigation and  the potential for costly
remediation,  this level of application requires  site-specific calibration  and validation using
quality-assured data collected specifically for  the model.  In dynamic systems,  linkage to an
equally well calibrated and validated hydrodynamic model is essential to represent, for example,
burial and exhumation of contaminated sediments.  Several of the more powerful features of the
model, such as the linked segments and IPX layered-sediment submodel, were developed for this
type of application.  Unfortunately, the one remediation application performed  by the model
developers cannot be published because of continuing litigation.
                                        11

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 2


                             2.  SIMULATION MODELING

2.1 Temporal and Spatial Resolution and Numerical Stability

AQUATOX Release 3 is designed to be  a general, realistic
model of  the fate  and  effects  of pollutants  in  aquatic
ecosystems. In order to be fast, easy to use, and verifiable, it
was   originally  designed with  the  simplest spatial  and
temporal resolutions consistent with this objective.  Releases      ™1X?..      ...  ...    ,   .
   r                                    J                    • Model is run with a daily or hourly
                                                            Simulation Modeling:  Simplifying
                                                            Assumptions
                                                              Each modeled segment is well-
                                                              maximum time-step.
                                                             • Results are trapezoidally integrated
may still be run as a non-dimensional point model.  However,
unlike previous versions of AQUATOX, in Release 3 spatial
segments may be linked together to  form a two- or three-
dimensional model if a more complicated spatial resolution is
desired.

The model  generally represents average daily conditions for a well-mixed aquatic system.  Each
segment in a multi-dimensional run is  also  assumed to be well-mixed in each time-step.
AQUATOX also represents one-dimensional vertical epilimnetic and hypolimnetic conditions
for those systems that exhibit stratification on a seasonal basis.  Multi-segment systems also can
be  set  up  with  vertical  stratification.    Furthermore, the effects  of run,  riffle,  and pool
environments can be represented for streams. Results may be plotted in the AQUATOX output
screen with the capability to import observed data to examine against model predictions.

While the model is generally run with a daily maximum time-step, the temporal resolution of the
model can  also be reduced to an hourly maximum time-step. This capability was added so that
AQUATOX can represent diel oxygen.  See sections 3.6  and 5.5 for more information on how
this choice  of hourly time-step affects AQUATOX equations.  The reporting step can be as long
as several years or as short as one hour; results  are integrated to obtain the desired reporting time
period.

According  to Ford and Thornton (1979), a one-dimensional model is appropriate for reservoirs
that are between 0.5 and 10 km in length; if larger,  then a two-dimensional model disaggregated
along the long axis is indicated.   The one-dimensional assumption is also appropriate for many
lakes (Stefan and Fang, 1994). Similarly, one can consider a single reach or stretch of river at a
time.

Usually the reporting time step is one day, but numerical instability is avoided by allowing the
step size of the integration to vary to achieve a predetermined accuracy in the solution. (This is a
numerical  approach, and the step  size is not directly related  to the  temporal  scale  of the
ecosystem  simulation.)  AQUATOX uses a very efficient fourth- and  fifth-order Runge-Kutta
integration  routine with adaptive step size to solve  the differential equations (Press et al.,  1986,
1992). The routine uses the fifth-order solution to determine the error associated with the fourth-
order solution; it decreases the step size (often to 15 minutes or less) when rapid changes occur
and increases the step size when there are slow changes,  such  as in winter. However, the step
size is constrained to a maximum of one day (or one hour in hourly simulations) so that short-
term pollutant loadings are always detected.  The reporting step, on the other hand, can be  as

                                        12

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 2
long as several years or as short as one hour; results are integrated to obtain the desired reporting
time period.

As an alternative, the user may specify an exact step size that is used thoughout the simulation.
This is similar to the way that many models solve the differential equations. The disadvantage is
that  the accuracy of the solution may not be maintained. However, it is  useful under  some
circumstances and is discussed more fully later in this section.

The  temporal and spatial resolution is in keeping with the generality and realism of the model
(see  Park and Collins, 1982). Careful consideration has been given to the hierarchical nature of
the  system.  Hierarchy theory tells us that models  should have resolutions appropriate to the
objectives; phenomena with temporal and spatial scales that are significantly longer than those of
interest should be treated as constants,  and phenomena with much smaller temporal and spatial
scales should be treated as steady-state properties or parameters (Figure 3; also see O'Neill et al.,
1986).  AQUATOX uses a longer time step than dynamic hydrologic models that are concerned
with representing short-term phenomena  such as storm hydrographs,  and it  uses a shorter time
step  than  fate  models  that may  be  concerned only  with  long-term   patterns such  as
bioaccumulation in large fish.

Figure 3. Position of ecosystem models such as AQUATOX in the spatial-temporal hierarchy of models.
                        Rule-based habitat models
                          succession, urbanization, sea-level rise
                                   Ecosystem models
                                   Population models
                                  High-resolution
                                  process models
                                   flood hydrograph
                                   diurnal pH
Changing the permissible  relative  error (the difference between the fourth- and  fifth-order
solutions) of the simulation can affect the results.  The model allows the user to set the relative
error, usually between 0.005 and 0.01.  Comparison of output shows that up to a point a smaller
error can yield a marked improvement in the simulation,  although execution time is longer.  For
example,  simulations of two pulsed doses of chlorpyrifos in a pond exhibit a spread in the first
pulse of about 0.6 ug/L dissolved toxicant between the simulation with 0.001 relative error and
the simulation with 0.05 relative error (Figure 4); this is probably due in part to differences in the
timing of the reporting step.  However,  if we  examine the dissolved oxygen  levels, which
combine the effects  of photosynthesis,  decomposition,  and reaeration, we find that there are
pronounced differences over the  entire simulation period. The simulations with 0.001 and 0.01
relative error give almost exactly the same results, suggesting that the more efficient 0.01 relative
error should be used; the simulation with 0.05 relative error exhibits  instability in the oxygen
                                        13

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                              CHAPTER 2
simulation; and the simulation with 0.1 error gives quite different values for dissolved oxygen
(Figure 5). The observed mean daily maximum dissolved oxygen for that period was 9.2 mg/L
(U.S. Environmental Protection Agency, 1988), which corresponds most closely with the results
of simulation with 0.001 and 0.01  relative error.
Figure 4.  Pond with chlorpyrifos in dissolved
phase.
Figure 5.  Same as Figure  4  with  Dissolved
Oxygen.
  0)
   06/19/88       06/30/88       07/12/88
         06/24/88        07/06/88        07/18/88
            0.001 -—0.01
                          0.05
                                                  12 r
                                                  11
                                                — 10
                                                o
                                                O
   06/19/88       06/30/88       07/12/88
          06/24/88       07/06/88       07/18/88
                                                          0.001 —0.01
                                                                        0.05 —0.1
A common use of AQUATOX is to determine the impact of a perturbation in a perturbed
simulation when compared with a control simulation. For example, the model is often run with
and  without a potentially  toxic organic chemical, and  a percent  difference graph is  plotted
showing how the two simulations differ. Of particular interest is whether there are likely to be
significant  differences   in  state  variables  or  other environmental  indices   at  very  low
concentrations of the chemical.  Because a simulation with the toxicant may require a decrease in
step size to capture the dynamics of the fate of the toxicant as opposed to a simulation without
the toxicant,  there may be  a  mismatch in the step sizes of the two  simulations,  and the
simulations may differ solely on the  basis of the difference in numerical resolution. Although
decreasing the relative error may decrease the mismatch, there may still be a difference that
prevents determination of the "no effects" level of the chemical.

In the example that follows, toxicity of PFOS has been turned off by setting all LC50 and EC50
parameter values to 0. In Figure 6 the default variable step size option has been used with a very
small relative error. In Figure 7 the simulation is the same except the constant step size  option
has been used, and it is  readily apparent that there is no difference between the  perturbed and
control runs.
                                        14

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 2
Figure 6. Percent differences in fish biomass between perturbed simulation with PFOS (toxicity turned
off) and control simulation without PFOS, using variable step size  (relative  error = 0.0001). The
differences are purely artifacts of the numerical method.
50.0
40.0
30.0
20.0
LU
£ 10.0
LU
aL
t °-°
J -10.0
-20.0
-30.0
-40.0
-50.0
Conasauga River, GA (Difference)





"-N /
^~J




	 Stoneroller
	 Spotted Bass
	 Catfish
	 Shiner
Bluegill

3/3/2008 6/1/2008 8/30/2008 11/28/2008 2/26/2009
Figure 7. There is no difference in fish biomass between perturbed simulation with PFOS (toxicity turned
off) and control simulation without PFOS using a constant step size (0.1 day).
5.00E+01
4.00E+01
3.00E+01
2.00E+01
LU
g 1.00E+01
oL
LL O.OOE+00
LL
° -1.00E+01
5*
-2.00E+01
-3.00E+01
-4.00E+01
-5.00E+01
Conasauga River, GA (Difference)











	 Stoneroller
	 Spotted Bass
	 Catfish
	 Shiner
• Bluegill

3/3/2008 6/1/2008 8/30/2008 11/28/2008 2/26/2009
2.2 Results Reporting

The AQUATOX results reporting time step may be set to any desired frequency, from a fraction
of an hour to multiple years. The Runge-Kutta differential equations solver produces a series of
results of variable frequency; this frequency may be either greater than or less than the reporting
time-step.   To standardize AQUATOX  output,  the user  has two  options, the trapezoidal
                                        15

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                            CHAPTER 2
integration of results (default) or the output of "instantaneous" concentrations.  Using either of
these options, AQUATOX will produce output with time-stamps that match the reporting time-
step precisely.

When  instantaneous concentrations  are requested (in the model's setup screen) AQUATOX
returns output precisely at the requested reporting time-step through linear interpolation of the
nearest Runge-Kutta results that occur before and after the relevant reporting time-step.

When results are trapezoidally integrated,  AQUATOX calculates results by summing all of the
trapezoids that can be produced by linear interpolation between Runge-Kutta results and dividing
by the results-reporting step-size to get an average result over the reporting step. In Figure 8, for
example, the areas of the four shaded trapezoids are summed together and this sum is divided by
the results reporting step to achieve an average result over that reporting step.  When trapezoidal
integration is selected, AQUATOX output is time-stamped at the end of the interval over which
the integration is taking place. For example, if a user selects a 366.25 day time-step, the results
at the end of the first year will be reflective of all time-steps calculated within that year.
                        Figure 8. An example of trapezoidal integration.
                                   Runge Kutta
                                    Step Size
                                    (Variable)
  Results-
Reporting Step
Results may be plotted  in the AQUATOX output  screen including the capability to import
observed data to examine against model predictions.

2.3 Input Data

AQUATOX accepts several forms of input data, a partial list of which follows:

   •   Point-estimate   parameters   describing   animals,   plants,  chemicals,   sites,   and
       remineralization.   Default values for these  parameters are generally available  from
       included databases (called "libraries"). The full list of these parameters, their units, and
                                        16

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                             CHAPTER 2
       their manner of reference in the interface, this document, and the source code  may be
       found in Appendix B of this document.
   •   Time series (or  constant values) for nutrient-inflow, organic matter-inflow, and gas-
       inflow loadings.
   •   Time series for inorganic sediments in water, water volume variables, and the pH, light,
       and temperature climates.
   •   Time series of chemical inflow loadings and initial conditions.
   •   A feeding preference matrix must be specified to describe the food web in the simulation.
   •   Additional  parameters may be required depending on which  submodels are included (e.g.
       additional sediment diagenesis parameters.)
   •   Nearly all point-estimate parameters may be represented by distributions when the model
       is run in uncertainty mode (see  section 2.5).

For more discussion  of AQUATOX  data requirements please  see the  "Data Requirements"
section in the AQUATOX Users Manual (or in the context sensitive help files included with the
model software). Furthermore, a Technical Note on Data Requirements is available.

For time-series loadings, when a value is input for every day of a simulation, AQUATOX will
read the relevant value on each day.  If missing values  are encountered by the model,  a  linear
interpolation will be performed between the surrounding dates.  If the AQUATOX simulation
time includes dates before or after the input time-series the model assumes an annual cycle and
tries to calculate the appropriate input  value accordingly.  Please see the "Important Note about
Dynamic Loadings" in  the AQUATOX Users  Manual (integrated help-file) for a complete
description of this process.
2.4 Sensitivity Analysis

"Sensitivity"  refers  to  the  variation  in  output  of  a
mathematical model with respect to changes in the values of
the model inputs (Saltelli 2001).  It provides a ranking of the
model input  assumptions  with  respect  to  their  relative
contribution to model output variability or uncertainty (U.S.
Environmental Protection Agency 1997).
                                                            Simplifying Assumptions:

                                                             • Parameters are treated as
                                                               independent
                                                             • Feeding preference matrices are not
                                                               included
                                                             • Sensitivity is compared for the last
                                                               step of the simulation
                                                            Caution
                                                             • 10% change is appropriate, a large
                                                               change can exceed reasonable
                                                               values and give misleading results
AQUATOX  includes  a built-in nominal range  sensitivity
analysis (Frey and Patil 2001), which may be used to examine
the sensitivity of multiple  model outputs to  multiple model
parameters.  The user first selects which model parameters to
vary and which output variables to track.  The model  iteratively steps through  each of the
parameters and varies them by a given percent in the positive and negative direction and saves
model results in an Excel file.

A sensitivity statistic may then be calculated  such that when  a 10% change in the parameter
results in a 10% change in the model result, the sensitivity is calculated as 100%.
                                        17

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 2
         Sensitivity =
                     Result Pos-Result Basetine +
                 ResultNeg-ResultBaselme
100
                                     2- Result,
                                             'aseline
                                        PctChanged
where:
       Sensitivity
       J\eSUltScenario
       PctChanged
normalized sensitivity statistic (%);
averaged AQUATOX result for a given endpoint given a positive
change  in the input parameter,  a negative change in the input
parameter or no change in the input parameter (baseline)
percent  that the input parameter is modified in the positive  and
negative directions.
Sensitivity is computed for the last time step of the simulation, so one usually sets the reporting
time step  to encompass a year or the entire period of the simulation. For each output variable
tracked, model parameters may be sorted on the average sensitivity (for the positive and negative
tests) and  plotted on a bar chart. The end result is referred to as a "Tornado Diagram."  Tornado
diagrams  may automatically be produced within the AQUATOX  output  window (Figure  9).
When interpreting a tornado diagram, the vertical line at the middle of the diagram represents the
deterministic model  result.  Red lines represent model results when the given parameter is
reduced by  the  user-input percentage while  blue lines represent a positive  change in the
parameter. An "effects  diagram"  that illustrates the effects  of a single parameter change on all
tracked outputs can also be created. See the User's Manual  (or context-sensitive help) for more
information on how to create and interpret these types of output.

When  sensitivity analysis is  run on  a  multi-segment model, the user  must choose either
parameters that are relevant to all segments (e.g.  animal  or plant parameters) or individual
segments  (e.g. state-variable initial conditions,  or the segment's physical characteristics). The
segment for which the parameter is  relevant may be selected in the sensitivity analysis setup
window (see the User's Manual  for more information)  Any number of global  or segment-
specific parameters may be selected for a single sensitivity-analysis; output files will be written
for each segment in the simulation.
                                        18

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                 CHAPTER 2
            Figure 9. An example tornado diagram showing calculated sensitivity statistic.
Sensitivity of Chironomid (g/m2 dry) to a 20% change in the 15 most sensitive (tested) parameters

220% - Site: Ave. Epilimnetic Temperature (deg C)
163% - Chironomid: Maximum Temperature (deg. C)

137% - Peri, Green: Max Photosynthetic Rate (1/d) -
1 1 5% - Temp: Multiply Loading by













I 	 '
=^i

^^^™


^^^^^™

•
•

i —



I 	 1

•
	 	 1
^^^~

^^^™




1.5 2 2.5 3 3.5 4 4.5 5
Chironomid (g/m2 dry)

2.5 Uncertainty Analysis

There  are numerous sources of uncertainty and variation in
natural systems.   These include: site characteristics such as
water depth, which may vary seasonally and from site to site;
environmental loadings such as water flow, temperature,  and
light, which may have a  stochastic component; and critical
biotic  parameters  such  as  maximum  photosynthetic  and
consumption  rates,  which  vary  among  experiments  and
representative organisms.

In addition,  there are sources of uncertainty and variation with
regard  to  pollutants, including:   pollutant  loadings  from
runoff, point sources, and atmospheric deposition, which may
vary stochastically from day to day and year to year; physico-chemical characteristics such as
octanol-water partition coefficients and Henry Law constants that cannot be measured easily;
chemodynamic parameters  such as microbial degradation, photolysis,  and hydrolysis  rates,
which may be subject to both measurement errors and indeterminate environmental controls.

Increasingly, environmental  analysts and decision makers are requiring probabilistic modeling
approaches  so that they can consider the implications of uncertainty in the analyses.  AQUATOX
Uncertainty Analysis: Strengths

 • Use of Latin hypercube sampling
   is more efficient than brute-force
   Monte Carlo analysis
 • Nearly    all    variables   and
   parameters may be represented as
   distributions
 • Variables can be correlated

Simplifying Assumptions:

 • Feeding preference matrices are
   not included
 • Modeled correlations  cannot be
   perfect (e.g. 1.0) due to limitations
   of the Iman & Conover method
                                         19

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                              CHAPTER 2
provides this  capability by  allowing the user to specify  the  types  of distributions  and key
statistics for almost all input variables.  Depending on the  specific variable and the amount of
available information, any one of several distributions may be  most appropriate.  A lognormal
distribution is the default for environmental and pollutant loadings.  In the uncertainty  analysis,
the distributions for constant loadings are sampled daily, providing day-to-day variation within
the limits of the distribution, reflecting the stochastic nature of such loadings. A useful tool in
testing  scenarios is  the  multiplicative loading  factor,  which can be applied  to all loads.
Distributions for dynamic loadings may employ multiplicative factors that are sampled once each
iteration (Figure 10). Normally the multiplicative factor for a loading is set to 1, but, as seen in
the example, under extreme  conditions the loading may be ten times  as great.  In this way the
user could represent unexpected conditions such  as pesticides  being applied inadvertently just
before each large storm of the season.  Loadings usually exhibit a lognormal distribution, and
that is suggested in these applications, unless there is information to the contrary.  Figure  11
exhibits the result of such a loading distribution.
             Figure 10. Distribution screen for point-source loading of toxicant in water.
                                                       Distribution Type:
                                                         i  Triangular
                                                           Uniform
                                                         C Normal
                                                         '••' Lognormal
                    0.673       5.83
                                                       Distribution Parameters:

                                                              Mean  [i
                                                        Stcl. Deviation  fo.6
'• Probability   r Cumulative Distribution
                 In an Uncertainty Run:
                    <•'  Use Above Distribution
                       Use Point Estimate
Choice of  distribution:  A  sequence of  increasingly  informative  distributions  should be
considered for most parameters.   If only two  values  are  known and  nothing more can be
assumed,  the two values may be used  as  minimum  and  maximum  values  for  a uniform
distribution (Figure 12); this is often used for parameters where only two values are known.  If
minimal information is available but there is reason to accept a particular value as most likely,
perhaps based on calibration, then a triangular distribution may be most suitable (Figure 13).
Note that  the minimum and maximum values for the distribution are constraints that have zero
probability of occurrence.  If additional data are available indicating both  a central tendency and
spread of response,  such as  parameters for well-studied processes, then a normal distribution

                                         20

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                           CHAPTER 2
may be most appropriate (Figure 14). The result of applying such a distribution in a simulation of
Onondaga Lake,  New York, is shown in Figure  15, where simulated benthic  feeding affects
decomposition  and subsequently the predicted hypolimnetic anoxia.  Most distributions  are
truncated at zero because negative values would have no meaning (Log Kow is one exception).

Figure 11. Sensitivity of bass (g/m2) to variations in loadings of dieldrin in Coralville Lake, Iowa.

                          Largemouth Ba2 (g/m2
                           4/17/2009 2:16:20 PM
   0.7
                                                                         • Mean
                                                                         •Minimum
                                                                          Maximum
                                                                          Mean - StDev
                                                                          Mean + StDev
                                                                          Deterministic
       10/24/1969   12/23/1969   2/21/1970    4/22/1970    6/21/1970    8/20/1970
Figure 12. Uniform distribution for Henry's Law
constant for esfenvalerate.
     0.00
       6.1 E-8
1.53E-6
3E-6
                              Figure 13. Triangular distribution for maximum
                              consumption rate for bass.	
                                                     0.04
                                                     0.00
0.015
0.0425
0.07
                                        21

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                        CHAPTER 2
                 Figure 14. Normal distribution for maximum consumption rate for
                              the detritivorous invertebrate Tubifex.
           (Distribution Information
                   0.04
                   0.03 •
                   0.02
                   0.01
                               0.25
                                        048.3
                  •V Probability  r Cumulative Distribution
Distribution Typo:

  r Triangular
  r uniform
  (• Normal
  C Lotjnormal
                  In an Uncertainty Run:
                      ff Use Above Distribution
                      r Use Point Estimate
Distribution Parameters:
       Mean  ]0.25
 Stii. Deviation  10.1
                                                         V  OK
              X Cancel
                Figure 15.  Sensitivity of hypolimnetic oxygen in Lake Onondaga to
               	variations in maximum consumption rates of detritivores.	
                     01/01/89          09/24/89         06/17/90
                              05/14/89          02/04/90         10/28/90

                               — Minimum         Mean
                               — Maximum     - - Deterministic
Efficient sampling from the distributions is obtained with the Latin hypercube method (McKay
et al., 1979; Palisade Corporation,  1991).  Depending on how many iterations are chosen for the
analysis, each  cumulative distribution is  subdivided into  that many  equal segments.  Then a
                                          22

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 2
uniform random value is chosen within each  segment and used  in  one of the subsequent
simulation runs. For example, the distribution shown in Figure 14 can be sampled as shown in
Figure 16.   This method is particularly advantageous because all regions of the distribution,
including the tails, are sampled.  A non-random seed can be used for the random number
generator, causing the same sequence of numbers to be picked in successive applications; this is
useful if you want to be able to duplicate the results exactly.  The default is twenty iterations,
meaning that twenty  simulations will be performed with sampled input values; this should be
considered  the  minimum number  to  provide  any  reliability. The optimal number  can be
determined experimentally by noting  the number required  to  obtain  convergence of mean
response values for key state variables; in other words, at what point do additional iterations not
result in significant  changes  in the  results?  As many  variables may  be represented by
distributions as  desired.  Correlations may be imposed using the method of Iman and Conover
(1982). By varying one parameter at a time the sensitivity of the model to individual parameters
can be determined in a more rigorous way than nominal range sensitivity offers. This is done for
key parameters in the following documentation.
                   Figure 16.   Latin hypercube sampling  of a  cumulative
                   distribution with a mean of 25 and standard deviation of 8
                   divided into 5 intervals.
                        4
                      0.8

                      0.6

                      0.4
                      0,2  t

                        0  -
An alternate way of presenting uncertainty is by means of a biomass risk graph, which plots the
probability that biomass will be reduced by a given  percentage by the end of the  simulation
(Mauriello and Park 2002).  In practice, AQUATOX compares the end value with the initial
condition for each state variable, expressing the result as a percent decline:
                              Decline = \ 1 -
                 EndVal
                 StartVal
\-100
(1)
where:
       Decline
       EndVal
percent decline in biomass for a given state variable (%);
value at the end  of the simulation for a given state variable (units
depend on state variable);
                                       23

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                  CHAPTER 2
       StartVal
initial condition for given state variable.
The  results from each iteration are sorted  and plotted in a cumulative distribution so that the
probability that a particular percent decline  will be exceeded can be evaluated (Figure 17).  Note
that there are ten points in this example, one for each iteration as the consecutive segments of the
distribution are sampled.
                Figure 17. Risk to bass from dieldrin in Coralville Reservoir, Iowa.

                          Biomass Risk Graph
                          4/17/2009 2:17:42 PM                          r
                                                                       • Largemouth Ba2
                  -40
                         -30      -20      -10       0
                         Percent Decline at Simulation End
                                                         10
Uncertainty analysis can also be used to perform statistical sensitivity analysis, which is much
more powerful than the screening-level nominal range sensitivity analysis. Parameters are tested
one at a time using the most appropriate distribution of observed parameter values.  The time-
varying and mean coefficient of variation can be calculated in an exported Excel file using the
mean and standard deviation results for a particular endpoint. Examples will be published in a
separate report.

2.6 Calibration and Validation

Rykiel (1996) defines calibration as "the estimation and  adjustment  of model parameters  and
constants to improve the  agreement between model output and a data set" while "validation is a
demonstration that a model  within its domain of applicability possesses a satisfactory range of
accuracy  consistent with the  intended application  of the model."   A  related  process is
verification, which is "a  subjective assessment of the behavior of the model" (I0rgensen 1986).
The terms are used in those ways in our applications of AQUATOX.

Endpoints for comparison of model results and data should utilize available data for various
ecosystem  components, preferably covering nutrients,  dissolved oxygen, and different trophic
                                        24

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 2


levels, and toxic organics if they are being modeled.  Although AQUATOX models a complete
food web, often the only biotic data  available are chlorophyll a values. The model converts
biomass predictions to chlorophyll a values to facilitate comparison. Likewise, Secchi depth is
computed from  the  overall  extinction  coefficient for  comparison  with  observed  data.
Verification  should consider  process  rates to  confirm that the results were obtained  for the
correct reasons  (Wlosinski and  Collins  1985). Rate information that  can be assessed for
reasonableness and compared with observations includes sediment oxygen demand (SOD), the
fluxes of phosphorus,  nitrogen, and dissolved oxygen, and all biotic process rates. These can be
presented in  tabular and graphical form in AQUATOX.

There are several measures of model performance that  can be used for both calibrations and
validations (Bartell et al.  1992, Schnoor 1996).  The  primary difficulty is  in comparing general
model behavior over long periods to observed data from a few points in time with poorly defined
sample variability.  Recognizing that  evaluation is  limited by the quantity and quality of data,
stringent measures of goodness of fit are often inappropriate; therefore,  we follow a weight-of-
evidence approach with a sequence of increasingly rigorous tests to evaluate performance and
build confidence in the model results:

   •   Reasonable  behavior as demonstrated  by time  plots  of key variables—is the model
       behavior reasonable based on general experience? Are the end conditions similar to the
       initial conditions? This is highly subjective, but when observed data are lacking or are
       sparse and restricted to short time periods it provides a limited reality check (Figure 18,
       Figure 19).

   •   Visual inspections of data points compared to model  plots—do  the observations and
       predictions exhibit a reasonable concordance of values  (Figure 20, Figure 21)?  Visual
       inspection can also take into consideration if there is concordance given a slight shift in
       time.

   •   Do model curves fall within the error bands of observed data (Figure 22)?  Alternatively,
       if there  are limited replicates, how do the model curves compare with  the spread of
       observed data?

   •   Do point observations fall within predicted  model bounds obtained through uncertainty
       analysis? This has the limitation of being dependent on the precision of the model; the
       greater the model uncertainty,  the greater the  possibility of the data being encompassed
       by the error bounds (Figure 23).

   •   Regression of paired data and model results—does the model produce results that are free
       of systematic bias?  What is the correlation  (R2)? See Figure 24, which corresponds to
       the results shown in Figure 20.

       Overlap between data and model distributions  based on relative bias (rB) in combination
       with  the ratio of variances  (F)—how much overlap is there (Figure 25)? Relative bias is
       a robust measure  of how well  central tendencies of predicted and observed results
       correspond; a value of 0 indicates that the means are the same (Bartell et al. 1992). The F
       test is the ratio of the variance of the model and  the variance of the data.  A value of 1
       indicates that the variances are the same.

                                        25

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
               CHAPTER 2
       Do the  observed  and predicted values  differ significantly  based on their  cumulative
       distributions (Figure 26)? The Kolmogorov-Smirnov statistic, a non-parametric test, can
       be used; however, the two datasets should represent the same time periods (for example,
       one should not compare predicted values over a year with observed values taken only
       during spring and summer).
      Figure 18. Predicted biomass patterns for animals in a hypothetical farm pond in Missouri.
                    FARM POND, ESFENVAL (CONTROL)
                        Run on 01-25-09 10:48 AM
           0.88
           0.80
           0.72
           0.64
           0.56
           0.48
           0.40
           0.32
           0.24
           0.16
           0.08 -
- Daphnia(mg/Ldry)

• Mayfly (Baetis (g/m2dry)
- Gastropod (g/m2 dry)
•Shiner (g/m2dry)
 Largemouth Bas (g/m2dry)
• Largemouth Ba2 (g/m2 dry)
             5/16/1994
                      8/14/1994
                                11/12/1994
                                          2/10/1995
 Figure 19. Sediment oxygen demand predicted for Lake Onondaga, using Di Toro sediment diagenesis
     	option; this is an example of using rates for a reality check.	
                ONONDAGA LAKE, NY (CONTROL)  Run on 06-20-08 2:47 PM
                              (Hypolimnion Segment)
     |  — SOD(gO2/m2d)|
           0.3
           0.0
           1/12/1989   5/12/1989
                              9/9/1989
                                       1/7/1990
                                                5/7/1990
                                                          9/4/1990
                                                                   1/2/1991
                                          26

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                    CHAPTER 2
      Figure 20. Comparison of predicted and observed (Oliver and Niemi 1988) PCB congener
                     bioaccumulation factors in Lake Ontario lake trout.
        11
        10-

         9 -
     2   8
     ?  7H
         6 -

         5 -

         4
                                   Lake Trout
Prod Obs = 0.97 +/-1.03
                                 *  Observed
                                — Predicted
                                6          7
                                 Log KOW
                  8
          Figure 21. Predicted biomass and observed numbers of chironomid larvae in a
                   Duluth, Minnesota, pond dosed with 6 ug/L chlorpyrifos.
0.40
0.36
0.32
0.28
^0.24
-D
IN 0.20
^0.16
0.12
0.08
0.04
0.00
CHLORPYRIFOS 6 ug/L (PERTURBED)
Run on 01-25-09 3:09 PM

[






I

6/27/1986 7/27/1986 8/26/1986

rti_- -j i i n j i
1000
• Obs. Chironomids (no./sample)
-900

600






                                     27

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 2
         Figure 22. Predicted and observed benthic chlorophyll a in Cahaba River, Alabama;
                       bars indicate one standard deviation in observed data.
                 1/1/01
7/20/01
2/5/02
8/24/02
 Figure 23. Visual comparison of the envelope of model uncertainty, using two standard deviations for
  each of the nutrient loading distributions, with the observed data for chlorophyll a in Lake Onondaga,
                                             NY.
     120
     100
                                                                            -Min Chloroph (ug/L)
                                                                             Mean Chloroph (ug/L)
                                                                            -Max Chloroph (ug/L)
                                                                            -Det Chloroph (ug/L)
                                                                            -ObsChl
                                          28

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                         CHAPTER 2
  Figure 24. Regression shows that the correlation between predicted and observed (Oliver and Niemi
   1988) PCB congener bioaccumulation factors in Lake Ontario trout may be very good, but the slope
 indicates that there is systematic bias in the relationship. See Figure 20 for another presentation of these
                                       same results.
                         10
                      00
                      o
                          9 -
                          8 H
                      •o   7
                      4)   7
                          6 -\
                                  LAKE ONTARIO TROUT
                                                       ^=0.915
6789
      Obs Log BAF
                                                                10
                                       29

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 2
   Figure 25. Relative bias and F test to compare means and variances of observed data and predicted
 results with AQUATOX. The isopleths correspond to the probability that the distributions of predicted
and observed, as defined by the combination of the rB and F statistics,  are similar. The isopleths assume
                                 normal distributions.
            Statistical Comparison of Means and Variances
         rB
                                                       rB = 0.242, f = 0.400
                                                       predicted and observed
                                                       distributionsare similar
          j-j	*J pred
                   obs
Figure 26   Comparison of predicted and observed chlorophyll a in Lake Onondaga, New
York (U.S. Environmental Protection Agency 2000). The Kolmogorov-Smirnov p statistic =
0.319, indicates that the distributions are not significantly different.	
                 100 i
                  90 i
              ^  80 1
               u  70 i
              -£  60:
               I
               3
              o
40 J
30 i
20 i
10 ^
 0 :
          20     40     60     80     100
               Chlorophyll a (ug/L)

        Observed         	 Predicted
                                                               120
                                     30

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 2
Data are often too sparse for adequate calibration at a given site.  However, AQUATOX can be
calibrated simultaneously across  sites using an expanded state variable list representative of a
range of conditions and using the  same parameter set. In this way the observed biotic data can be
pooled and the resulting state variable and parameter sets, being applicable to diverse sites, are
assured to be robust. This is an approach that we have used on the Cahaba River, Alabama (Park
et al. 2002); on three dissimilar rivers in Minnesota (Park et al. 2005); and on 13 diverse reaches
on the Lower Boise River Idaho (CH2M HILL et al. 2008).  The Minnesota rivers application is
discussed below.

Time series of driving variables  for the Minnesota rivers were obtained from several sources
with varying degrees of resolution and reliability. Results of watershed simulations with HSPF
(Hydrologic  Simulation  Program-  Fortran,  a watershed  loading model)  were  linked  to
AQUATOX, providing boundary conditions (site constants and drivers) for the Blue Earth and
Crow Wing Rivers (Donigian et  al. 2005). HSPF was not run for the Rum River; however, a
U.S. Geological Survey (USGS) gage is located at the sample site and both daily discharge and
sporadic water quality data  were available from the USGS  Web  pages (search on "National
Water Information System"). AQUATOX interpolates between points, and this feature was used
to compute daily time series of nutrient  concentrations from USGS National Water Information
System (NWIS) observed data. Total suspended solids (TSS) are critical  because the daily light
climate for algae is affected.  Therefore, we derived a significant relationship by regressing TSS
against In-scaled discharge and used that to generate  a daily time series for the Rum River
(Figure 27).
                                       31

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                             CHAPTER 2
Figure 27.  TSS at Rum River: a) linear regression against daily flow at gage; b) resulting simulated daily
                        time series (line), and observed values (symbols).
          a)
 25
                   20
                   10 •
          b)
140

120 -

100 -
                   60 H
                   40 -
                   20 -
                               400000      800000      1200000

                                          Daily MeanFlow (m3/d)
                                              1600000
                                                       2000000
                    0
01/99     05/99     08/99      12/99
                                       04/00     08/00
                                                                           12/00
After calibration we evaluated the  efficacy of generating daily time  series for TN using  a
regression of TN on discharge. The relationship is statistically significant and yielded a more
realistic  time series than the interpolation  with sparse data that we had used  (Figure 28).
                                         32

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                           CHAPTER 2
However, calculation of the different limitations  on photosynthesis  indicates that N  is not
limiting in the Rum River (Figure 29), so we kept the simpler approach and did not repeat the
calibration (see section 4.1 for an explanation of the reduction factor as an expression of nutrient
limitation) .  TP did not exhibit a statistically significant trend with discharge (R2 = 0.124) so the
simple interpolation was also kept.
 Figure 28. TN at Rum River site: a) In-linear regression against daily flow at gage; b) interpolated TN
            observations (red) and time series (black) estimated from discharge regression.
          ,1!
          b)
   1.8

   1.6

   1-1

~  1.2
|
&   *•

\  0.8

£  0.6

   0.4

   0.3
                         2.50
                         0.00
                                                        R2 = 0.6861
                                   500000       1000000      1500000

                                            Discharge (tni/d)
                                                                       2000000
                                                                       -EstTN

                                                                       -ObsTM
                                         33

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
               CHAPTER 2
       Figure 29. Predicted nutrient limitations for the dominant algal group in the Rum River.
       	Note that N is not limiting.	
           0.9

           0.8

           0.7

           0.6

           0.5
          u
          £o.5

           0.4

           0.3

           0.2

           0.1
                       Rum R. 18 MN (CONTROL)
                        Run on 01-27-09 7:35 AM
• Peri High-Nut NJ.IM (frac)
•Peri High-Nut PO4J.IM (frac)
- Peri High-Nut CO2_LIM (frac)
                3/21/1999 7/19/1999 11/16/1999 3/15/2000 7/13/2000 11/10/2000
In almost all cases parameter values were  chosen from ranges reported in the literature (for
example, Le Cren and Lowe-McConnell 1980, Collins and Wlosinski 1983, Home and Goldman
1994, Jorgensen et al. 2000, Wetzel 2001).  However, because these often are broad ranges and
the model is very sensitive to some parameters, iterative calibration was necessary for a subset of
parameters in AQUATOX.  Conversely, some parameters have well established values and
default values were used with confidence. A few parameters such as extinction coefficients and
critical force for sloughing of periphyton are poorly defined or are unique  to the AQUATOX
formulations and were treated as "free" parameters subject to broad calibration.  For example,
some periphyton species are able to migrate vertically through the  periphyton mat, and others
have open growth forms; therefore, they could be assigned extinction coefficient values without
regard to  the physics of light transmission through biomass fixed in space. As  noted earlier,
sensitivity analysis can help  determine  how much attention needs to  be  paid  to individual
parameters.  Sensitivity analysis of five diverse studies has shown that the model is sensitive to
optimal temperature (TOpf) for algae and fish, maximum photosynthesis (PMax)  for algae, %
lost in periphytic sloughing, and log octanol-water partition coefficient (KOW). It is advisable to
perform  sensitivity  analysis when  the  initial  calibration is  complete  in order to identify
parameters and driving variables requiring additional attention. Although not  used  in this
application, if modeling a toxic chemical,  there are several  published  sources (for example,
Lyman  et al.  1982, Verscheuren  1983,  Schwarzenbach et al. 1993), and  there are a couple
excellent online references, including the US EPA ECOTOX site and the USDA ARS Pesticide
Properties Database, which can be found with an Internet search engine.

Calibration of AQUATOX for the Minnesota rivers used observed chlorophyll a as the primary
target for obtaining best fits.  Because  there were only five to eight sestonic chlorophyll a
observations in each of the two  target  years and only one benthic chlorophyll a observation at
                                        34

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 2


each location, calibration  adequacy was  evaluated subjectively, based  on generally expected
behavior (e.g.  blooms occurring during summer) and approximate concordance with observed
values (in terms of both magnitude and timing), as determined through graphical comparisons of
model output and data (Figure 30).

The central tendencies are similar for predicted and observed distributions for all three sites, as
shown by the relative bias (Figure 31).  Despite the fluctuations in predicted chlorophyll a,  the
predicted and observed  variances are similar for  the Crow Wing River  and Rum River
simulations.  Predicted  periphyton sloughing events played a major role in  determining  the
timing of chlorophyll a peaks in both simulations. The variance in predicted values is too high in
the Blue Earth River simulation, where  summer peak concentrations  in 1999 appear to be
overestimated by a factor of about two.  The reason for this is not known, but may be related to
inherent uncertainties in the simulated flow and TSS values, the sparseness of water chemistry
sampling data, and/or limitations of model algorithms.  Given the wide range in degree of
enrichment among these three rivers, and the fact that the model was  calibrated against all three
data sets using a single set of parameters, a two-fold error during one period of the Blue Earth
River simulation seems to be acceptable.  The combined probability that the Blue Earth River
predictions  and observations have the same  distribution, based on both central tendency and
dispersion,  is greater than 0.8.  For the purpose of this analysis, we judged the  calibration to be
adequate for the three rivers.
                                       35

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                         CHAPTER 2
 Figure 30.  Observed (symbols) and calibrated AQUATOX simulations (lines) of chlorophyll a in three
 Minnesota rivers: a) Blue Earth at mile 54, b) Rum at mile 18, c) Crow Wing at mile 72. Note the order-
                        of-magnitude range in scale among the figures.
            a)
 400
 350
 300
 250

950
                   01/99   05/99   08/99   12/99  04/00  08/00   12/00
             b)
                   01/99   05/99   08/99   12/99  04/00  08/00   12/00
             c)
                   30
                   25 -
                   20 -
                   110 -
                  CJ
                    5 -
                    0
                    01/99   05/99   08/99   12/99   04/00  08/00   12/00
                                      36

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 2
Figure 31. Overlap between model and data distributions based on relative bias and ratio of variances, F;
  1 = Blue Earth River, 2 = Crow Wing River, 3 = Rum River.  Isopleths indicate the probability that the
               predicted and observed distributions are the same, assuming normality.
                                        -202
                                          Relative Bias
The calibrated algal model was also applied to three dissimilar sites on the Lower Boise River,
Idaho, without  modification from the  Minnesota calibration.    This provided additional
verification of the generality of the parameter set. The three sites cover a broad range of nutrient
and turbidity conditions over 90 km.  Eckert is a low-nutrient, clear-water site upstream of Boise;
Middleton receives wastewater treatment effluent and is a nutrient-enriched, clear-water site; and
Parma is  a nutrient-enriched, turbid site  impacted by irrigation return flow from agricultural
areas.  Although the model  overestimated periphyton at the Eckert site, the fit  of the initial
application (Figure 32) provided an excellent basis for further river-specific calibration.
                                        37

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 2
  Figure 32. Predicted (line) and observed (symbols) benthic chlorophyll a (a) at Eckert Road, (b) near
        Middleton, (c) near Parma, Lower Boise River, Idaho, using Minnesota parameter set.
              b)
              c)
                      2'JO
                        1/1/98
250
                        11, yG
                                       36296
                                36796
                                 9/27/00
                 5/16/99
9/27/00
                                        38

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 2
As a limited validation, the calibrated model was applied to a site on the Cahaba River south of
Birmingham, Alabama, with modifications to only two parameters,  critical force for periphyton
scouring and optimal temperature for algae.  The Crow Wing and Rum Rivers have cobbles and
boulders and are more sensitive to higher current velocities than the bedrock outcrops in  the
Cahaba River.  Not only is the bedrock stable, it also provides abundant crevices  and lee sides
that are protected refuges for periphyton. For these reasons greater water velocity is expected to
be required to initiate  periphyton scour in the Cahaba River than in the Crow Wing and Rum
Rivers,  thus  the  critical force  (Fcrit) for scour of periphyton was more than doubled in  the
Cahaba River simulation.  Also, between Minnesota and Alabama one would  expect different
local ecotypes in resident algal species, with differing adaptations to temperature.  Based on
professional judgment, the optimum temperature values (Topt) for green algae and cyanobacteria
were therefore increased by 5°C to 31°C and 32°C respectively.  The resulting fit to observed
data (Figure 22) was good. Furthermore, the fish and zoobenthos fits were acceptable (Figure 33,
see also Figure 70). Note that the bluegill are predicted to exhibit ammonia toxicity in 2001, an
observation made possible by viewing  biotic process  rates.  (Within rates  graphs,  animal
mortality rates may be broken down into their various constituents, see (112)).

 Figure 33. Predicted and observed fish in Cahaba River, Alabama; predicted shiner mean biomass = 0.6
         	g/m2 compared to observed 0.5 g/m2.	
                       Cahaba River AL (CONTROL)
                         Run on 01-26-09 12:07 PM
                                                              -Shiner (g/m2 dry)
                                                              - Bluegill (g/m2 dry)
                                                              - Stoneroller (g/m2 dry)
                                                              - Smallmouth Bas (g/m2 dry]
                                                              Smallmouth Ba2 (g/m2 dry)
                                                              Obs stonerollers (g/m2dry)
                                                              Obs shiners (g/m2 dry)
                                                              Obs bluegill (g/m2 dry)
                                                              Obs bass (g/m2 dry)
                 8/26/2000  2/24/2001   8/25/2001  2/23/2002  8/24/2002
In another validation, published PCB data from New Bedford Harbor, Massachusetts, were used
to verify the  generality of the estuarine  ecosystem bioaccumulation  model.  The observed
concentrations of total PCBs in the water and bottom sediments in the Massachusetts site were
set as  constant  values in a  simulation  of  Galveston Bay,  Texas.   The predicted  PCB
concentrations  in  the various  biotic compartments at the end  of the simulation  were then
compared to the observed means and standard deviations in New Bedford Harbor (Figure 34).
Considering that  the  sites and  some of the species were different,  the concordance in values
provides a validation of the model for assessing bioaccumulation of chemicals in a "canonical"
or representative estuarine environment.
                                        39

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                 CHAPTER 2
   Figure 34.  Predicted and observed concentrations of PCBs in selected animals based on ecosystem
   calibration for Galveston Bay, Texas and exposure data (Connolly 1991) for New Bedford Harbor,
                                     Massachusetts.
                10
               0.1
CO
u
Q_
>-
T3
O
-Q
          |
              0.01
                                        eS^     £•      4^
                                      x?     x>*      .><>      ^
                                     *v      »o     vCr     vO
                                    x^      v*      <>y     <>y
                                   ^    >^
A third example of a validation is shown in Figure 21, which provides a visual comparison of
predicted biomass  and  observed  numbers per sample of chironomid larvae with dosing by an
insecticide. No calibration was performed for either the fate or toxicity of the chemical.
                                       40

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                          CHAPTER 3
                         3. PHYSICAL CHARACTERISTICS
3.1 Morphometry

Volume

Volume is a state variable and can be computed in several
ways depending on  availability of  data and the  site
dynamics.  It  is important for computing the dilution or
concentration  of pollutants, nutrients,  and organisms; it
may be constant, but usually  it is time varying.  In the
model, ponds, lakes, and reservoirs are treated differently
than  streams,   especially  with  respect  to  computing
volumes.   The  change in volume  of ponds, lakes,  and
reservoirs is computed as:
                         Morphometry: Simplifying
                         Assumptions

                          • Base flow equation assumes a
                            rectangular channel
                          • Site shapes are represented by
                            idealized geometrical
                            approximations
                          • Mean Depth may be held constant
                            or user varying depth may be
                            imported
                           dVolume
                              dt
                                    = Inflow - Discharge - Evap
                                                   (2)
where:
       dVolume/dt
       Inflow
       Discharge
       Evap
derivative for volume of water (m3/d),
inflow of water into waterbody (m3/d),
discharge of water from waterbody (m3/d), and
evaporation (m3/d), see (3).
AQUATOX cannot successfully run if the volume of water in a site falls to zero.  To avoid this
condition, if the site's water volume falls below a minimum value (which is defined as a fraction
of the initial condition using the parameter "Minimum Volume Frac." from the "Site" screen), all
differentiation of state variables is suspended (except for the water volume  derivative) until the
water volume again moves above the minimum value.  Differentiation of all state variables then
resumes.

A time series of evaporation may be entered in the "Site" screen in units of cubic meters per day.
Otherwise, evaporation is converted from an annual value for the site to a daily value using the
simple relationship:
                             Evap =
  MeanEvap
     365
•0.02 54-Area
(3)
where:
       Evap
       MeanEvap
       365
       0.0254
       Area
       mean daily evaporation (m /d)
       mean annual evaporation (in/yr),
       days per year (d/yr),
       conversion from inches to meters (m/in), and
       area of the waterbody (m2).
                                       41

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                   CHAPTER 3
The user is given several options for computing volume including keeping the volume constant;
making the volume a dynamic function of inflow, discharge, and evaporation; using a time series
of known  values;  and, for flowing waters, computing volume as a function of the Manning's
equation.  Depending on the method, inflow and discharge are varied, as indicated in Table 3.
As shown in equation (2), an evaporation term is present in each of these volume calculation
options. In order to keep the volume constant, given a known inflow loading, evaporation must
be subtracted from discharge. This will reduce the quantity of state variables that wash out of the
system. In the dynamic formulation, evaporation is part of the differential equation, but neither
inflow nor discharge is a  function of evaporation as they are both entered  by the user.  When
setting the volume of a water body to a known value, evaporation must again be subtracted from
discharge  for the volume solution to be correct.  Finally, when using the Manning's volume
equation, given a known  discharge  loading, the effects of evaporation must be added  to  the
inflow loading  so  that the  proper  Manning's  volume  is achieved.   (This could increase  the
amount of inflow loadings of toxicants and sediments to the system, although not significantly.)

                   Table 3. Computation of Volume, Inflow, and Discharge
Method
Constant
Dynamic
Known values
Manning
Inflow
InflowLoad
InflowLoad
InflowLoad
ManningVol - State/dt + Discharge + Evap
Discharge
InflowLoad - Evap
DischargeLoad
InflowLoad - Evap + (State -
Known Vals)/dt
DischargeLoad
The variables are defined as:

       InflowLoad
       DischargeLoad
       State
       KnownVals
       dt
       ManningVol
user-supplied inflow loading (m /d);
user-supplied discharge loading (m3/d);
computed state variable value for volume (m3);
time series of known values of volume (m3);
incremental time in simulation (d); and
volume of stream reach (m3), see (4).
Figure 35 illustrates time-varying volumes  and inflow loadings  specified  by the user and
discharge computed by the model for a run-of-the-river reservoir.  Note that significant drops in
volume occur with operational releases, usually in the spring, for flood control purposes.
                                       42

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 3
                  Figure 35. Volume, inflow, and discharge for a 4-year period
                               in Coralville Reservoir, Iowa.
                     6.0E+07
                     O.OE+00
                               2.5E+08
                                                           2.0E+08
                                                           1.5E+08
                                                           1 .OE+08
                                                           5.0E+07
                                                                  E
                                                                  13
                                                                  o
                          Oct-74    Oct-75   Nov-76   Dec-77
                              Apr-75   May-76   Jun-77   Jul-78
                               O.OE+00
                       — Inflow
            Discharge-  Volume
The time-varying volume of water in a stream channel is computed as:
where:
       7
       CLength
       Width
ManningVol = Y • CLength • Width

     dynamic mean depth (m), see (5);
     length of reach (m); and
     width of channel (m).
                                                                                    (4)
In streams the depth of water and flow rate are key variables in computing the transport, scour,
and deposition of sediments. Time-varying water depth is a function of the flow rate, channel
roughness,  slope, and  channel  width using  Manning's  equation  (Gregory,  1973),  which is
rearranged to yield:
                                 Y =
                                       Q • Manning
                                                   -3/5
                                                        (5)
where:
       Q
       Manning
       Slope
       Width
     flow rate (m /s);
     Manning's roughness coefficient (s/m1/3);
     slope of channel (m/m); and
     channel width (m).
The Manning's roughness coefficient is an important parameter representing frictional loss, but it
is not subject to direct measurement.  The user can choose among the following stream types:
                                       43

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3


   .   concrete channel (with a default Manning's coefficient of 0.020);
       dredged channel, such as ditches and channelized streams (default coefficient of 0.030);
       and
   •   natural channel (default coefficient of 0.040).

These generalities are based on Chow's (1959) tabulated values as given by Hoggan (1989). The
user may also enter a value for the coefficient.

In the  absence of inflow data, the flow rate is computed from the initial mean water depth,
assuming a rectangular channel and using a rearrangement of Manning's equation:

                                    IDepth5/3 • JSlope • Width
                            QBase =	                         (6)
                                           Manning
where:

       Idepth              =     mean depth as given in site record (m).
QBase              =     base flow (m3/s); and
The dynamic flow rate is calculated from the inflow loading by converting from m3/d to m3/s:
                                          86400
where:
       Q                  =     flow rate (m3/s); and
       Inflow              =     water discharged into channel from upstream (m3/d).
Bathymetric Approximations

The depth distribution of a water body is important because it determines the areas and volumes
subject to mixing and light penetration. The shapes of ponds, lakes, reservoirs, and streams are
represented in the  model by idealized geometrical  approximations, following the topological
treatment  of Junge  (1966;  see also Straskraba and Gnauck, 1985).   The shape parameter P
(Junge,  1966)  characterizes the site, with  a  shape  that is indicated  by  the ratio of mean to
maximum depth. :
                                          ZMax
Where:
      ZMean       =     mean depth (m);
      ZMax        =     maximum depth (m); and
      P            =     characterizing parameter for shape  (unitless);  P is  constrained
                          between -1.0 and 1.0

Shallow constructed ponds and ditches may be  approximated by an ellipsoid where Z/ZMax =
0.6 and P = 0.6.  Reservoirs and rivers generally are extreme elliptic sinusoids with values ofP

                                       44

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3


constrained to -1.0.  Lakes may be either elliptic  sinusoids, with P between 0.0 and -1.0, or
elliptic hyperboloids with P between 0.0 and 1.0. Not all water bodies fit the elliptic shapes, but
the model generally is not sensitive to the deviations.

Based on these relationships, fractions of volumes and areas can be determined for any given
depth (Junge, 1966).  The AreaFrac function returns the fraction of surface area that is at depth Z
given Zmax and P, which defines the morphometry  of the water body. For example, if the water
body were an inverted cone, when horizontal slices were made through the cone looking down
from the top one could see both the surface area and the water/sediment boundary where the slice
was made.   This would look like a circle within a circle, or a donut (Figure 36).  AreaFrac
calculates the fraction that is the donut (not the donut hole). To get the donut hole, 1 - AreaFrac
is used.


                         AreaFrac = (1-P)- —^— + P • (-^— /                      (9)
                                          ZMax       ZMax


                       6.0 • —	3.0-(1.0-P)-(—^—f-2.0-P-(—^—f
              VolFrac =     ZMax	ZMax	ZMax             / j Q\
                                            3.0 + P

where:
      AreaFrac            =     fraction of area of site above given depth (unitless);
       VolFrac             =      fraction of volume of site above given depth (unitless); and
      Z                   =      depth of interest (m).

For example, the fraction of the volume that is epilimnion can be computed by setting depth Zto
the mixing depth. Furthermore, by setting Z to  the depth of the euphotic zone, where primary
production exceeds respiration, the fraction of the area available for colonization by macrophytes
and periphyton can be computed:

                      „    T    ,     ZEuphotic  „  (ZEuphotic\                   _„.
                      FracLit = (l-P)	+ P-\	                    (11)
                                       ZMax        \   ZMax  )

A relatively deep, flat-bottomed basin would have a small littoral area and a large sublittoral area
(Figure 36).

                                       Figure 36.
                                       45

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 3
If the site is an artificial enclosure then the available area is increased accordingly:

                                             Area + EnclWallArea
                       FracLittoral = FracLit • •
                                                    Area
                                       otherwise
                                                             (12)
where:
       FracLittoral
       ZEuphotic

       Area
       EnclWallArea
          FracLittoral = FracLit

=   fraction of site area that is within the euphotic zone (unitless);
=   depth of the euphotic zone, is assumed to be 1% of surface light
    and calculated as 4.605/Extinct (m) see (40);
=   site area (m2); and
=   area of experimental enclosure's walls (m2).
Figure 37. Area as a function of depth
              RESERVOIR (P = -0.6)
       1   3  5  7  9  11 13 15  17 19  21  23 25
        2  4  6 8  10 12 14 16 18 20 22 24
                   DEPTH(m)
                    Figure 38. Volume as a function of depth
                                                          RESERVOIR (P = -0.6)
                           1   3  5 7  9  11 13  15 17 19  21  23 25
                             2  4  6  8  10 12 14 16 18 20 22 24
                                                              DEPTH (m)
If a user wishes to model a simpler system, the bathymetric approximations may be bypassed in
favor of a more rudimentary set of assumptions via an option in the "site data" screen.

When the user chooses not to "use bathymetry"

    •   the system is assumed to have vertical walls;
    •   the system is assumed to have a constant area as a function of depth;
    •   the system's depth may be  calculated  at any time as water volume divided by surface
       area.

This option may be useful when linking data from other models to AQUATOX as the horizontal
spatial domain of AQUATOX remains unchanged  over time.   However, a system will  not
undergo dynamic stratification based on water  temperature unless the more complex bathymetric
approximations are utilized ((8) to (11)).
                                       46

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3
Dynamic Mean Depth

AQUATOX normally uses an assumption of unchanging mean depth (i.e., mean over the site
area). However, under some circumstances, and especially in the case of streams and reservoirs,
the depth of the system can change considerably over time, which could result in a significantly
different light climate for algae.  For this reason, an option to import mean depth in meters has
been added. A daily time-series of mean depth values may be imported into the software (using
an interface found within the "site" screen by pressing the "Show Mean Depth Panel" button.)  A
time-series of mean depth values can  be  estimated given  known water volumes or  can be
imported from a linked water hydrology model.

The user-input dynamic mean depth affects the following portions of AQUATOX:

    •   Light climate, see (43);
    •   Calculation of biotic volumes for sloughing calculations, see (74);
    •   Calculation of vertical dispersion for stratification calculations, Thick in equation (18);
       Calculation of sedimentation for plants & detritus, Thick in (165);
       Oxygen reaeration, see (190);
    .   Toxicant photolysis and volatilization, Thick in (320) and (331).
Habitat Disaggregation

Riverine environments are seldom homogeneous.  Organisms often exhibit definite preferences
for habitats.  Therefore, when modeling streams or rivers, animal and plant habitats are broken
down into three categories: "riffle,"  "run," and "pool."   The combination of these three habitat
categories make up 100% of the  available habitat within a riverine simulation.  The preferred
percentage of each organism that resides within these three habitat types can be set within the
animal or plant data. Within the site  data, the percentage of the river that is composed of each of
these three habitat categories also can be set.  It should be noted that the habitat percentages are
considered constant  over time, and thus would not capture significant changes in channel
morphology and habitat distribution due to major flooding events.

These  habitats affect  the  simulations  in  two ways: as limitations on photosynthesis  and
consumption and as weighting  factors for water velocity (see 3.2 Velocity). Each animal and
plant is exposed to a weighted average water velocity depending on its location within the three
habitats.  This weighted velocity affects all velocity-mediated processes including entrainment of
invertebrates and  fish, breakage of macrophytes and scour of periphyton.  The reaeration of the
system also is affected by the habitat-weighted velocities.

Limitations on  photosynthesis and consumption  are  calculated  depending  on a species'
preferences for habitats  and the available  habitats within the water body.   If the  species
preference for a particular habitat is equal to zero then the portion of the water body that contains
that particular habitat limits the amount of consumption or photosynthesis accordingly.
                                        47

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                             CHAPTER 3
where:
     HabitatLimit species  =

     Preference habitat   =

     Per cent habitat      =
                                                   s            >
                                                    Per cent habitat
                                                  \     100
                                                                                     (13)
                           fraction  of site available  to  organism  (unitless),  used to limit
                           ingestion, see (91), and photosynthesis, see (35), (85);
                           preference  of animal  or  plant for  the  habitat in  question
                           (percentage); and
                           percentage  of  site  composed  of  the   habitat  in  question
                           (percentage).
It is important to note that the initial condition for an animal that is entered in  g/m  is an
indication of the total mass of the animal over the total surface area of the river. Because of this,
density data for various benthic organisms, which is generally collected in a specific habitat type,
cannot be used as input to AQUATOX until these values have been converted to represent the
entire surface area. This is especially true in modeling habitats; for example, an animal  could
have a high density within riffles, but riffles might only constitute a small portion of the  entire
system.
3.2 Velocity

If the user has site-specific velocity data, this may be entered on the "site data" screen in units of
cm/s. Otherwise, velocity is calculated as a simple function of flow and cross-sectional area:
                             T, ,       AvgFlow
                             Velocity =   s
                                      XSecArea 86400
                                                       -•100
(14)
where
       Velocity
       AvgFlow
       XSecArea
       86400
       100
                           velocity (cm/s),
                           average flow over the reach (m3/d),
                           cross sectional area (m2),
                           s/d, and
                           cm/m.
where:
       Inflow
       Discharge
                              AvgFlow =
                                         Inflow + Discharge
                                                 2
(15)
                           flow into the reach (m /d);
                           flow out of the reach (m3/d).
It is assumed that this is the velocity for the run of the stream (user entered velocities are also
assumed to pertain to the run of the screen).  No distinction is  made in  terms of vertical
differences in velocity in the stream. Following the approach and values used in the DSAMMt
model (Caupp et al. 1995), the riffle velocity is obtained by using a conversion factor that is
                                        48

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 3
dependent  on the discharge.   Unlike the DSAMMt model, pools  also  are  modeled, so a
conversion factor is used to obtain the pool velocity as well (Table 4).

                Table 4. Factors relating velocities to those of the average reach.
Flows (Q = discharge)
Q < 2.59e5 m3/d
2.59e5 m3/d < Q < 5.18e5 m3/d
5. 18e5 m3/d < Q < 7.77e5 m3/d
Q > 7.77e5 m3/d
Run
Velocity
1.0
1.0
1.0
1.0
Riffle
Velocity
1.6
1.3
1.1
1.0
Pool Velocity
0.36
0.46
0.56
0.66
               Figure 39.  Predicted velocities in an Ohio stream according to habitat.
                 400
                 350
                                          CD
                                                   00
                                               . Riffle
                             • Pool
3.3 Washout

Transport out of the system, or washout,  is an  important loss term for  nutrients, floating
organisms, and  dissolved  toxicants in reservoirs  and streams.   Although it  is  considered
separately for several state variables, the process is a general function of discharge:
where:
       Washout
       State
                               Washout=
                                        _ Discharge
                                           Volume
                        • State
loss due to being carried downstream (g/m -d), and
concentration of dissolved or floating state variable (g/m3).
(16)
                                        49

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                 CHAPTER 3
3.4 Stratification and Mixing

Thermal stratification is handled  in  the simplest form
consistent with the goals of forecasting the effects  of
nutrients  and  toxicants.    Lakes and  reservoirs  are
considered in  the model to  have two vertical  zones:
epilimnion and hypolimnion (Figure 40);  the metalimnion
zone  that separates  these  is  ignored.    Instead,  the
thermocline, or plane of maximum temperature change, is
taken as the  separator; this is also known as the mixing
depth (Hanna, 1990).  Dividing the lake into two vertical
zones follows the treatment of Imboden (1973), Park et al.
(1974),  and Straskraba and Gnauck (1983).  The onset of
stratification is considered to  occur when the mean water
temperature  exceeds  4   deg.   and   the difference  in
temperature between  the epilimnion  and  hypolimnion
exceeds 3 deg..  Overturn occurs when the temperature of the epilimnion is less than 3  deg.,
usually  in the fall.  Winter stratification is not modeled, unless manually input. For simplicity,
the thermocline is generally assumed to occur at a constant depth. Alternatively, a user-specified
time-varying thermocline depth may be specified, see the section on modeling reservoirs below.

           	Figure 40. Thermal stratification in a lake; terms defined in text	
Stratification: Simplifying
Assumptions

 • Two vertical zones modeled;
   metalimnion is ignored
 • Flowing waters are assumed not to
   stratify
 • Stratification occurs when vertical
   temperature difference exceeds
   three degrees
 • Winter stratification is not modeled
 • Thermocline occurs at constant
   depth except when user enters time
   series
 • Wind action is implicit in vertical
   dispersion calculations
                                         Epilimnion
                                       	Thermocline
                                    VertDispersion
There  are  numerous  empirical  models  relating thermocline depth  to lake  characteristics.
AQUATOX uses an equation by Hanna (1990), based on the maximum effective length (or
fetch). The dataset includes 167 mostly temperate lakes with maximum effective lengths of 172
to 108,000 m and ranging in altitude from 10  to 1897 m. The equation has a coefficient of
determination r2 = 0.850, meaning that 85 percent of the  sum of squares is explained by the
regression.  Its curvilinear nature is shown in Figure 41, and it is computed as (Hanna, 1990):
                         \og(MaxZMix) = 0.336- \og(Length) - 0.245
                         (17)
where:
                                        50

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 3
      MaxZMix      =    maximum mixing depth under stratified conditions (thermocline
                          depth) for lake (m); and
      Length        =    maximum effective length for wave setup (m, converted from user-
                          supplied km).
                        Figure 41. Mixing depth as a function of fetch
                                  MAXIMUM MIXING DEPTH
                       100        11500       22900       34300
                            5800       17200       28600      40000
                                       LENGTH (m)
Wind action is implicit in this formulation.  Wind has been modeled explicitly by Baca and
Arnett (1976, quoted by Bowie et al., 1985), but their approach requires calibration to individual
sites, and it is not used here.

Vertical dispersion for bulk mixing is modeled as a function of the time-varying hypolimnetic
and epilimnetic temperatures, following the treatment of Thomann and Mueller (1987, p. 203;
see also Chapra and Reckhow,  1983, p. 152; Figure 42):
                VertDispersion = Thick
                                     I      HypVolume
                                 rr,t-l  rr,t+l
                                 -L hypo ~ J- h
                                                                  lypo
                                     \ ThermoclArea • Deltat   T'epi -
                                                       (18)
                                                                  rypo
where:
       VertDispersion =
       Thick          =
vertical dispersion coefficient (m2/d);
distance between the centroid of the epilimnion and the centroid of
the hypolimnion, effectively the mean depth (m);
                                      51

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                      CHAPTER 3
HypVolume
ThermoclArea
Deltat
ji   t-i  ji   t+i
1 hypo ' * hypo

T~T   t T~T  t
1 epi > * hypo
                           volume of the hypolimnion (m );
                           area of the thermocline (m2);
                           time step (d);
                           temperature of hypolimnion one time step before and one time step
                           after present time (deg. C); and
                           temperature  of epilimnion  and  hypolimnion  at present time
                           (deg.C).
Stratification can break down temporarily as a result of high throughflow.  This is represented in
the model by making the  vertical  dispersion coefficient between the  layers  a function  of
discharge for sites with retention times of less than or equal to 180 days (Figure 43), rather than
temperature differences as in equation 11, based on observations by Straskraba (1973) for a
Czech reservoir:
                         VertDispersion = 1.3 7 • J Q4 • Retention
                                                            -2.269
                                                                              (19)
and:
                                Retention = •
                                              Volume
                                            TotDischarge
                                                                              (20)
where:
              Retention     =
              Volume       =
              TotDischarge =
                           retention time (d);
                           volume of site (m3); and
                           total discharge (m3/d).
               Figure 42.  Vertical dispersion as a function of temperature differences
                 25
                                                                       iu
                                                                       O
                                                                       o
                                                                       O
                                                                       (A
                                                                  0.01
                  0 -I	1	1	1	1	1	1	1	1	1	1	1	h1- 0.001
                  12/30    02/28    04/29    06/28    08/27    10/26    12/25
                                         DAY
                        	Epilimnion Temp.      -*-  Hypolimnion Temp.

                        	Vert. Dispersion (sq m/d)     4 degrees
                                        52

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 3
                  Figure 43. Vertical dispersion as a function of retention time
                    100
                  Q
                  w
                  tr
                  LJJ
                  D_
                     10 .
                  OH
                  LU
                    0.1
                               VERTICAL DISPERSION
                      180   162   144   126   108   90   72    54    36
                         171   153  135  117   99    81    63   45   27

                                       RETENTION TIME (d)
The bulk vertical mixing coefficient is computed using site characteristics and the time-varying
vertical dispersion (Thomann and Mueller, 1987):
                      BulkMixCoeff =
          VertDispersion • ThermoclArea
                      Thick
(21)
where:
       BulkMixCoeff =      bulk vertical mixing coefficient (m3/d),
       ThermoclArea =      area of thermocline (m2).
Turbulent diffusion of biota and other material between epilimnion and hypolimnion is computed
separately for each segment for each time step while there is stratification:
              T  hPi'ff  _ BulkMixCoeff
              L UtOLJlJJ e^       ^        ' ( (^OTICcompartment, hypo ~ ^OWC compartment, epi/
                             Volumeepi
                          _ BulkMixCoeff
               _/ Hi ULJIJJ -faypO                \ \^ OTIC compartment, epi ~ ^ OTIC compartment, hypo/
                                                        (22)


                                                        (23)
where:
       TurbDiff
       Volume
       Cone
turbulent diffusion for a given zone (g/m3-d);
volume of given segment (m3); and
concentration of given compartment in given zone (g/m3).
                                       53

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
               CHAPTER 3
The effects of stratification, mixing due to high throughflow, and overturn are well illustrated by
the pattern of dissolved oxygen levels in the  hypolimnion of Lake Nockamixon, a eutrophic
reservoir in Pennsylvania (Figure 44).

                   Figure 44.  Stratification and mixing in Lake Nockamixon,
             	Pennsylvania as shown by hypolimnetic dissolved oxygen
                    14
                    10
                  S, 8
                  o>

                  "o  4
                  00
                  CO
                                        onset of
                                        stratification
   ,
overturn
      jf
      ''
                                         high
                                         throughflow
                    01/01182 03/07/82 05/11182 07/15/82 09/18/82 11 /22/S2
Modeling Reservoirs and Stratification Options

Stratification assumptions and equations based on lake characteristics may not be appropriate for
modeling  reservoirs.   Moreover,  a  lake may  have a  unique  morphometry  or  chemical
composition that renders inappropriate the  equations presented  above.  For this reason,  a
"stratification options" screen is available (through the "site" screen or "water-volume" screen)
that allows a user to specify the following characteristics of a stratified system:

   •   a constant or time-varying thermocline depth;
   •   options as to how to route inflow and outflow water; and
   •   the timing of stratification.

Water volumes for each segment are calculated as a function of the overall system volume and
the thermocline depth (see  (10)).  Because of this,  if a time-varying thermocline depth is
specified, water from one segment must usually be transferred into the other segment, along with
the state variables within that water.  In this manner, specifying a time-varying thermocline depth
has the potential to promote mixing between layers. Alternatively, using the linked-mode model,
two stratified segments may be specified with water volumes that are calculated independently
from the thermocline depth; see section 3.8 for more details about stratification in linked-mode.

By default, AQUATOX routes inflow and outflow to  and from both segments as weighted by
volume. For example, if the hypolimnion has twice as  much volume as the epilimnion, twice as
                                        54

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3


much inflow water will be routed to the hypolimnion as to the epilimnion (and twice as much
outflow water will be routed from the hypolimnion).  The user has the option to route all inflow
and outflow waters to and from either segment.  In this case, all of the nutrients, chemicals, and
other loadings within the inflow water will be routed directly to the specified segment and will
not be  transferred to the other  segment  except through turbulent  diffusion  or overturn.
Atmospheric and point-source loadings are assumed to be  routed to the epilimnion in all cases
(unless a linked-mode model is used in which case more flexibility is present).

Additionally, if a user has information about the timing of stratification, this may be specified on
the "stratification-options"  entry screen.  This can be used to specify winter stratification, for
example,  or precise periods  of stratification  for  each  year modeled.   If only  one year of
stratification dates are entered and multiple years are modeled, all years are assumed to stratify
and overturn on the dates  specified in the user input (regardless of the year specified).

3.5 Temperature

Temperature is  an important controlling factor  in  the model.   Virtually all processes are
temperature-dependent.   They include stratification; biotic  processes such  as decomposition,
photosynthesis,  consumption, respiration, reproduction,  and  mortality;  and  chemical  fate
processes such as microbial degradation, volatilization, hydrolysis,  and bioaccumulation.  On the
other hand, temperature rarely fluctuates rapidly in aquatic systems. Default water temperature
loadings  for  the epilimnion  and  hypolimnion  are  represented  through  a  simple   sine
approximation for seasonal variations (Ward, 1963) based on user-supplied observed means and
ranges (Figure 45):

        r       +     r    **     / i n  TempRcmge
        Temperature = TempMean + (-1.0	—
                                            2                                         (24)
                    • (sin(0.0174533 • (0.987 • (Day + PhaseShift) - 30))))]
where:
       Temperature  =     average daily water temperature (deg. C);
       TempMean   =     mean annual temperature (deg. C);
       TempRange  =     annual temperature range (deg. C),
      Day          =     day of year (d); and
      PhaseShift   =     time lag in heating (=  90 d).

Observed temperature loadings should be entered  if responses to short-term variations are of
interest.  This is especially important if the timing of the onset of stratification is critical, because
stratification is a function of the  difference in hypolimnetic and epilimnetic temperatures  (see
Figure 42).  It also is important in streams subject to releases from reservoirs and other point-
source temperature impacts.
                                        55

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                         CHAPTER 3
3.6 Light

Light  is  important   as   the  controlling   factor  for
photosynthesis and photolysis.  The default incident light
function  formulated for AQUATOX is a variation on the
temperature equation, but without the lag term:
                                         Light: Simplifying Assumptions

                                          • Ice  cover  is assumed when the
                                           average  water temperature drops
                                           below 3 degrees centigrade.
                                          • Photoperiod is  approximated by
                                           Julian date (day of year)
                                          • Average daily light is the program
                                           default, although hourly light may
                                           be simulated
Solar = LightMean +
Ll8htRan8e .
                                      01 74533 . Day . L 76) . FracLi
                                                               Light
                    Fracught =1.0- 0.98(Canopy)
where:
       Solar         =
       LightMean    =
       LightRange   =
       Day
       Frac Light
       Canopy       =
       average daily incident light intensity (ly/d);
       mean annual light intensity (ly/d);
       annual range in light intensity (ly/d);
       day of year (d, adjusted for hemisphere);
       fraction of site that is shaded; and
       user input fraction of site that is tree shaded.
                                                                                      (25)
The derived values are given  as  average light  intensity  in  Langleys  per  day (Ly/d  = 10
      r\
kcal/m -d).  An observed time-series of light also can be supplied by the user; this is especially
important if the  effects of daily weather conditions are  of interest.    If the average  water
temperature drops below 3 deg.C, the model assumes the presence of ice cover and decreases
transmitted light to 15% of incident radiation. (This has changed from 33% in Release 2.2.) This
reduction, due to the reflectivity and transmissivity of ice and snow, is  an average of widely
varying values  summarized by Wetzel  (2001).  For estuaries,  average water temperature must
fall below -1.8 deg.C before the model assumes ice cover due to the influence of salinity.

Shade can be an  important limitation  to light, especially in riparian systems.  A user input
"fraction of site subject to shade from a canopy" parameter can be entered either as a constant or
as a time-series within the "Site" input screen.  This parameter can be left as zero for no shading
effects on light.  Transmission of light through a riparian (stream-side) canopy is a combination
of diffuse and direct transmission (Canham et al. 1990).  The average of four forest types from
closed hemlock to open spruce (and cypress) forests is 2% of incident radiation (Canham et al.
1990). Detailed studies in a Midwestern mixed  deciduous  forest confirm this value for the
summer months,  although transmission increased to 40%  in winter (Oliphant et al. 2006). A
value of 2% transmission for a closed  canopy is used in AQUATOX.  If the density of canopy
                                        56

-------
 AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 3
 varies during the year, then a time-series  should be provided, keeping in mind that the 2%
 transmission will still apply to the fraction of canopy that is indicated.

 Photoperiod  is an integral part of the photosynthesis  formulation.  It is approximated using the
 Julian date following the approach of Stewart (1975) (Figure 46):
           12 + A- cos (380 •
                         Photoperiod = •
                                                       365
                                                             248)
                                                   24
                                                         (26)
 where:
        Photoperiod   =

        A
        Day
fraction of the day with daylight (unitless); converted from hours
by dividing by 24;
hours of daylight minus 12 (d); and
day of year (d, converted to radians).
 A is the difference between the number of hours of daylight at the summer solstice at a given
 latitude and the vernal equinox, and is given by a linear regression developed by Groden (1977):
 where:
        Latitude
        Sign
  A = 0.1414 • Latitude - Sign • 2.413

latitude (deg., decimal), negative in southern hemisphere; and
1.0 in northern hemisphere, -1.0 in southern hemisphere.
                                                                                     (27)
Figure 45. Annual Temperature
                  Figure 46. Photoperiod as a Function of Date
    TEMPERATURE IN A MIDWESTERN POND
              77     153    229    305
          39     115     191    267    343
                    JULIAN DAY
                                                 : 0.65
                                                  0.55
                                                Q
                                                "5 0.45
                                                tz
                                                I 0.4
                                                ro
                                                LL 0.35
                         1    53   105   157  209  261   313   365
                          27   79   131   183   235   287  339
                                     Julian Date

                             — Latitude 40 N — Latitude 40 S
                                         57

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                              CHAPTER 3
Hourly Light

When the model is run with an hourly time-step, solar radiation is calculated as variable during
the course of each day.  The following equation is used to distribute the average daily incident
light intensity over the portion of the day with daylight hours.
         Solar,
                          Solar,,,
                        Photoperiod
                      sin
                             brae Day Passed
      1-Photoperiod
             2
                                       Photoperiod
                                                    (28)
                                  • Frac
                                                                       Light
where:
       Solar
            houriy
       Photoperiod
       FracDayPassed
       Frac Light
solar radiation at the given time-step (ly/d);
average daily incident light intensity (ly/d), see (25);
fraction of the day with daylight (unitless); see (26);
fraction of the day that has passed (unitless)
fraction of site that is un-shaded, (frac., 1.0-user input shade);
A user may enter a constant or time-series shade variable in the site window ("Fraction of Site
that is Shaded"). When this input is utilized then the Frac Light variable is calculated.
               Figure 47: Average light per day is distributed during daylight hours
                        in a semi-sinusoidal pattern based on photoperiod.
                 1200
                 1000 -

                  800 -

                  600 -

                  400 -

                  200 -

                    0
                      0
 10
20         30
   Hours
50
                                        58

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                      CHAPTER 3
3.7 Wind

Wind is an important driving variable because it determines
the stability of blue-green algal blooms, affects reaeration
or oxygen  exchange, and controls volatilization of some
organic chemicals.  Wind  also can affect the  depth of
stratification  for estuaries.   Wind is usually measured at
meteorological stations at a height of 10 m and is expressed
as m/s. If site data are  not available, default variable wind speeds are represented through a
Fourier series of sine  and cosine terms; the mean and twelve additional harmonics  seem to
effectively capture the variation (Figure 48):
      Wind: Simplifying Assumptions

       • If site data are not available a
        Fourier series is used to represent
        wind loadings
Win d =
                                           365
sinCoeff „ . Sm
where:
       Wind        =    wind speed; amplitude of the Fourier series (m/s);
       CosCoeffo    =    cosine coefficient for the 0-order harmonic, which is the mean wind
                         speed (default = 3 m/s);
       CosCoeffn    =    cosine coefficient for the n* -order harmonic;
       Day         =    day of year (d);
       SinCoeffn    =    sine coefficient for the nth-order harmonic;
       Freqn        =    selected frequency for the nth- order harmonic.

This default loading is based  on an annual cycle  of data taken from the Buffalo, NY airport.
Therefore,  it has a 365-day repeat, representative of seasonal variations in wind.  Frequencies
were selected to ensure that the standard deviation of the Fourier series and the data were closely
matched. The frequency of wind-speeds of less than three meters per second were also precisely
matched to observed data as well as the periodicity of wind-events.   The Fourier approach is
quite useful because the mean can be specified by the user and the variability will be imposed by
the function.

If ice cover is predicted, wind is set to 0. A user also may input a site-specific time series,  which
may be important where the timing of a cyanobacteria bloom or reaeration is of interest.
                                        59

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                              CHAPTER 3
            Figure 48.  Default wind loadings for Onondaga Lake with mean = 4.17 m/s.
               ONONDAGA LAKE, NY (CONTROL)  Run on 04-24-08 9:07 PM
                              (Epilimnion Segment)
                                                                               Wind (m/s)
        1/12/1989
                  5/12/1989
                             9/9/1989
                                       1/7/1990
                                                 5/7/1990
                                                            9/4/1990
                                                                      1/2/1991
3.8 Multi-Segment Model

AQUATOX Release 3  includes the capability to link AQUATOX segments together, tracking
the flow of water and the passage of state variables from segment to segment.  Some general
guidelines for using this model follow:
                                                             Multi-Segment Model: Simplifying
                                                             Assumptions

                                                              • All  linked  segments  have  an
                                                                identical set of state variables
                                                              • Each segment is well mixed
                                                              • Linkages between segments may be
                                                                unidirectional or bidirectional
                                                              • Dynamic  stratification  does  not
                                                                apply; stratified pairs of segments
                                                                must be specified by the user
   .   All linked segments must have an identical set of
       state variables. (State variables that do not occur in
       one segment may be set to zero there.)
   •   Parameters  pertaining  to  animal,  plant,   and
       chemical state variables (i.e. "underlying data") are
       considered global to the  entire linked  system. If
       the user changes  one  of these  parameters in one
       segment,   this  parameter  changes  within  all
       segments.
       On  the  other  hand,  "site"  parameters,  initial
       conditions, and boundary conditions  are unique to
       each segment.
   •   State variables  can  pass from  segment  to segment through active upstream and
       downstream migration, passive drift, diffusion, and bedload.
   •   Mass balance of all state variables is maintained throughout a multi-segment simulation.

There are two  types of linkages that may be specified between individual segments, "cascade
links" and "feedback links." A cascade link is unidirectional; there is no potential for water or
                                         60

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                  CHAPTER 3
state variable flow back upstream.  Segments that are linked together by cascade linkages are
solved separately from one another moving from upstream to downstream. This is particularly
useful when modeling faster flowing rivers and streams.

A feedback link allows for water or state variables to flow in both directions.  For bookkeeping
purposes, water flows are required to be unidirectional (i.e. entered water flows over a feedback
link must not be negative).  However, two feedback links may be specified simultaneously (in
opposite directions) to allow for bidirectional water flows. Feedback links may also be subject to
diffusion; a diffusion coefficient, characteristic  length,  and cross section must be entered for
diffusion to be calculated,  see (32).  Segments  that are linked together by feedback links are
solved simultaneously.  There may only be one contiguous set of segments linked together by
feedback linkages within a simulation (i.e. the model will not solve a "feedback" set of segments
followed by downstream cascade segments followed by more feedback segments below that.)

Figure 49 gives an example of a simulation in which cascade segments and feedback segments
are both included.  In this case, AQUATOX solves the  simulation from the top down,  solving
each segment 1-4,  6, and 6b  individually before  moving on to solve the feedback segments
simultaneously.  Finally, segments 11-14 are be solved  individually using the results from the
simultaneous segment run.

             Figure 49: An example of feedback and cascade segments linked together.
                                                 \
                                                  x  )  Feedback Seg.
Cascade Seg.

Feedback Link

Cascade Link
                                      61

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3
Stratification and the Multi-Segment Model

Dynamic stratification as described in section 3.4 does not apply to the multi-segment model.
Instead, a user may specify two linked segments as a stratified pair.  In this case, the segments
must be linked together with a feedback linkage.  A "stratification" screen within each segment's
main interface allows a user to specify whether a segment  is  part of a stratified pair and, if so,
whether it is the epilimnion or the hypolimnion segment.

When two segments are set up as stratified together, the thermocline area is  defined by the user-
entered cross section between.  Annual  cycles of stratification and overturn may be specified
using the time varying water flows and dispersion coefficients.  As was the  case in the dynamic
stratification model, fish automatically migrate to the epilimnion in the case of hypoxia in the
lower segment.  Sinking phytoplankton  and  suspended detritus in the  epilimnion segment fall
into the designated hypolimnion segment. The light climate of the bottom segment is limited to
that light which penetrates the segment defined as the epilimnion.

When  the  linked system  has  enough  specified throughflow between the  epilimnion and
hypolimnion segment, it is considered to be "well mixed."  This is defined as when the average
daily water  flow between segments  is  greater than 30%  of the total water  volume in  both
segments. In this case, fish are assumed to have an equal preference to both segments and they
migrate to equality in a biomass basis.  (This allows fish to return to the hypolimnion if it had
earlier  been  vacated due to anoxia.) Another implication of a well-mixed stratified system (in
linked  mode) is  that a weighted  average of light climate is used  when calculating plant
productivity.  The calculation of LightLimit for plants (38)  is based on a thickness-weighted
average of algal biomass  and  sediment throughout the entire  thickness  of the  system.   This
prevents unreasonable  model results due to the light climate in  a very thin epilimnion, for
example. Because the  system is well-mixed, suspended algae should instead be subject to the
light climate throughout the water column.
State Variable Movement in the Multi-Segment Model

To maintain mass balance, all state variables that are subject to washout or passive drift are also
added to any downstream linked segments.  The calculation for this process is as follows:

             Washin=    y   Washoutupanam • VolumeUpstream • FracWash,,^
                         / J                -TT  J                                      \   S
                      upstream link             ' ^"    Dowstream Segment

In the case  of toxicants that are absorbed to or contained within  a drifting state variable, the
following equation is used:

                                                   • VolumeUpstream • FracWas^^
                                                                                    '   '
                                           T/ /
                upstream links                    ' OlUmt Dow ,stream Segment
                                       62

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 3


where:

       Washin           =     inflow load from upstream segment (unit/Ldownstream'd);
       Washout upstream   =     washout from upstream segment (unit/LUpStream'd), see (16);
       Volume segment     =     volume of given segment (m3);
       FracWashThiSLink  =     fraction  of  upstream  segment's  outflow  that goes to  this
                              particular downstream segment (unitless);
       WashinToxCamer   =     inflow load  of toxicant sorbed  to a carrier from  an upstream
                              Segment (Hg/Ldownstream'd);
       Washout earner     =     washout of toxicant carrier from  upstream (mg/Lupsireanrd);
       PPBcarrier        =     concentration of toxicant in carrier upstream (|j,g/kg),  see (310);
       le-6             =     units conversion (kg/mg)

This Washin term is added to all derivatives for state variables that are suspended in the water
column and subject to  drift or "washout."


Dissolved state variables are subject to diffusion across feedback links.

               r»vcr  •           DiffCoeff• Area i                        \
              Diffusion      =   JJ    JJ	— (Conc0      -Cone     )           (32)
                                  CharLength

where:
      DiffusionThisSeg  =  gain of state variable due to diffusive transport over the feedback link
                        between two segments, (unit/d);
      DiffCoeff      =  dispersion coefficient of feedback link,  (m2 /d);
      Area          =  surface area of the feedback link (m );
      CharLength    =  characteristic mixing length of the feedback link, (m);
                     =  concentration of state variable in the relevant segment, (unit/m3);
                                       63

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                 CHAPTER 4
                                        4.  BIOTA

The biota consists of two main groups, plants and animals;
each is represented by a set of process-level equations.  In
turn, plants are differentiated into algae and macrophytes,
represented  by   slight variations  in  the   differential
equations.    Algae  may   be  either  phytoplankton  or
periphyton. Phytoplankton are  subject  to sinking and
washout, while periphyton  are subject to substrate limitation and scour by currents. Bryophytes
and freely-floating macrophytes are  modeled  as   special classes  of macrophytes, limited  by
nutrients in the water column.  These  differences are treated at the process level in the equations
(Table 5).  All are subject to habitat availability, but to differing degrees.

                   Table 5. Significant Differentiating Processes for Plants
Biota: Simplifying Assumptions

 • Biomass  is  simulated but  not
   numbers of individual organisms
 • Responses  are   simulated   as
   averages for the entire group
Plant Type
Phytoplankton
Periphyton
Benthic
Macrophytes
Rooted-Floating
Macrophytes
Free-Floating
Macrophytes
Bryophytes
Nutrient
Lim.
J
J


J
a
Current
Lim.

J




Sinking
J





Washout
J



J

Sloughing

J




Breakage


J
J
J
a
Habitat
J
J
J
J
J
a
Animals  are  subdivided  into  invertebrates  and  fish;  the  invertebrates  may be  pelagic
invertebrates, benthic insects or other benthic  invertebrates.  These groups are represented by
different parameter values and by variations in the equations. Insects are subject to emergence
and  therefore are  lost from  the system, but  benthic  invertebrates  are not.   Fish may be
represented by both juveniles and adults,  which are connected by promotion.  One fish species
can be designated as multi-year with up to 15 age classes connected by promotion. Differences
are shown in Table 6. In addition,  a bioaccumulative endpoint  such  as bald eagle, dolphin, or
mink that feeds on  aquatic compartments  can be simulated; it is  defined by feeding preferences,
biomagnification factor, and clearance rate.
                                        64

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                                             CHAPTER 4
                     Table 6.  Significant Differentiating Processes for Animals
Animal Type
Pelagic Invert.
Benthic Invert.
Benthic Insect
Fish
Washout
J



Drift

J
J

Entrainment

J
J
a
Emergence


J

Promotion/
Recruitment



a
Multi-year



a
4.1  Algae
                                                     o                                 9
                                 -expressed as g/m  for phytoplankton, but as g/m  for periphyton-
The change in algal biomass-
is  a  function  of  the  loading  (especially  phytoplankton   from  upstream),  photosynthesis,
respiration, excretion or photorespiration, nonpredatory mortality, grazing or predatory mortality,
sloughing,  and  washout.  As noted  above, phytoplankton also are  subject  to  sinking.   If the
system is stratified, turbulent diffusion also affects the biomass of phytoplankton.
  Plants: Simplifying Assumptions
   • Photosynthesis is modeled as a maximum observed rate multiplied by reduction factors.  The reduction factors are
     assumed to be independent of one another.
     Intracellular storage of nutrients is not modeled; constant stoichiometry within species is assumed
     For each individual nutrient, saturation kinetics is assumed
     Algae exhibit a nonlinear, adaptive response to temperature changes
     Low temperatures are assumed not to affect algal mortality
     The ratio between biovolume and biomass is assumed to be constant for a given growth form
     Constant chlorophyll a to biomass ratios are assumed within algae groups

  Phytoplankton-speciflc
   • Phytoplankton other than cyanobacteria are assumed to be mixed throughout the well-mixed layer unless specified as
     "surface floating."
   • In the event of ice cover, all phytoplankton will occur in the top 2 m
   • Sinking of phytoplankton is modeled as a function of physiological state
   • Phytoplankton are subject to downstream drift as a simple function of discharge
   • To model phytoplankton (and zooplankton) residence time, an implicit  assumption may be made that upstream reaches
     included in the "Total River Length " have identical environmental conditions as the reach being modeled

  Cyanobacteria-specific
   • By default cyanobacteria are specified as "surface floating" in which case they are assumed to be located in the top 0.1 m
     unless limited by lack of nutrients or sufficient wind occurs  in which  case they  are located within the top 3 m. This
     default assumption (that cyanobacteria float) can be changed by the user.
   • The averaging depth for "surface floating" plants is three meters to more  closely correspond to monitoring data.
   • Cyanobacteria are not severely limited by nitrogen due to facultative nitrogen fixation (if N less than 1A KN)

  Periphyton-specific
   • Periphyton are limited by slow currents that do not replenish nutrients and carry away senescent biomass
   • Periphyton are assumed to adapt to the ambient conditions of a particular channel
   • Periphyton are defined as including associated detritus; non-living biomass is modeled implicitly

  Macrophyte-speciflc
   • Macrophytes occupy the littoral zone
   • Rooted macrophytes are not limited by nutrients but are assumed to take up necessary nutrients from bottom sediments
                                                65

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 4
             dBiomass
                      Phyto
                  dt
                          = Loading + Photosynthesis - Respiration - Excretion
                        -Mortality - Predation ± Sinking ± Floating
                        - Washout + Washin ± TurbDiff + Diffusion
                                                                    Slough
                                                               Seg
                                                              (33)
             dBiomass
                  ~dt
Pen  = Loading + Photosynthesis - Respiration - Excretion

     -Mortality - Predation + Sedperi - Slough
(34)
where:
       dBiomass/dt

       Loading
       Photosynthesis
       Respiration
       Excretion
       Mortality
       Predation
       Washout
       Washin

       Sinking

       Floating
       TurbDiff
       Diffusionseg

       Slough

       Sedpen
=    change in biomass of phytoplankton and periphyton with respect to
              T          r\
     time (g/m -d and g/m -d);
=    boundary-condition loading of algal group (g/m3-d and g/m2-d);
                              T          r\
=    rate of photosynthesis (g/m -d and g/m -d), see (35);
                       T          r\
=    respiratory loss (g/m  -d and g/m -d), see (63);
=    excretion or photorespiration (g/m3-d and g/m2-d), see (64);
                               T          r\
=    nonpredatory mortality (g/m -d and g/m -d), see (66);
=    herbivory (g/m3-d and g/m2-d), see (99);
=    loss due to being carried downstream (g/m3-d), see (129);
=    loadings  from upstream segments (linked segment version only,
     g/m3-d), see (30);
=    loss  or gain due to sinking between layers and sedimentation to
     bottom (g/m3-d), see (69);
=    loss  from the  hypolimnion or gain  to the epilimnion due to the
     floatation of "surface-floating" phytoplankton.  100% of "surface-
     floating" phytoplankton that arrive in the hypolimnion through
     loadings or water flows are set to immediately float.
     turbulent diffusion (g/m3-d), see (22) and  (23);
=    gain or loss  due to diffusive transport over  the feedback link
     between two segments, (g/m3-d), see (32);
=    Scour loss of Periphyton or addition to linked Phytoplankton, see
     (75); and
=    Sedimentation  of  Phytoplankton   to   Periphyton,  see  (83).
Figure 50 and Figure 51 are examples of the predicted changes in biomass and the processes that
contribute to these changes in a eutrophic lake. Note that photosynthesis and predation dominate
the diatom rates, with respiration much less important during the growing season.
                                       66

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                    CHAPTER 4
Figure 50. Predicted algal biomass in Lake Onondaga, New York
        ONONDAGA LAKE, NY (PERTURBED) Run on 04-23-08 2:59 PM
                          (Epilimnion Segment)
         1/12/1989  5/12/1989   9/9/1989   1/7/1990   5/7/1990   9/4/1990   1/2/1991
         • Cyclotella nan (mg/L dry)
         •Greens (mg/L dry)
         - Phyt, Blue-Gre (mg/L dry)
         • Cryptomonad (mg/L dry)
Figure 51. Predicted process rates for diatoms in Lake Onondaga, New York
   ONONDAGA LAKE, NY (PERTURBED) Run on 04-23-08 2:59 PM
                     (Epilimnion Segment)
            3/11/1989
                       9/9/1989
                                  3/10/1990
                                              9/8/1990
• Cyclotella nan Photosyn (Percent)
• Cyclotella nan Respir (Percent)
• Cyclotella nan Excret (Percent)
- Cyclotella nan Other Mort (Percent)
• Cyclotella nan Predation (Percent)
 Cyclotella nan Washout (Percent)
• Cyclotella nan Sediment (Percent)
 Cyclotella nan TurbDiff (Percent)
 Cyclotella nan SinkToHypo (Percent)
                                             67

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Photosynthesis is modeled as a maximum observed rate multiplied by reduction factors for the
effects of toxicants, habitat, and suboptimal light, temperature, current, and nutrients:

           Photosynthesis = PMax • PProdLimit • Biomass • HabitatLimit • SaltEffect        (35)

The limitation of primary production in phytoplankton is:

                    PProdLimit = LtLimit • NutrLimit • TCorr • FracPhoto                (36)

Periphyton have an additional limitation based on available substrate, which includes the littoral
bottom and the available surfaces of macrophytes.  The macrophyte surface area conversion is
based on the observation of 24 m2 periphyton/m2 bottom (Wetzel, 1996) and assumes that the
observation was made with 200 g/m macrophytes.

                PProdLimit = LtLimit • NutrLimit • VLimit • TCorr • FracPhoto
                                                                                    (37)
                     • ( FracLittoral + SurfAreaConv • BiomassMacroPhytes)
where:
    Pmax           =   maximum photosynthetic rate (1/d);
    LtLimit         =   light limitation (unitless), see (38);
    NutrLimit       =   nutrient limitation (unitless), see (55);
    Vlimit          =   current limitation for periphyton (unitless), see (56);
    TCorr          =   limitation due to suboptimal temperature (unitless), see (59);
    HabitatLimit    =   in  streams, habitat  limitation  based  on  plant  habitat preferences
                        (unitless), see (13).
    SaltEffect       =   effect of salinity on photosynthesis (unitless);
    FracPhoto      =   reduction factor for effect  of toxicant  on  photosynthesis   (unitless),
                        see (421);
    FracLittoral    =   fraction of area that is within euphotic zone (unitless) see (11);
    SurfAreaConv   =   surface area conversion for periphyton growing on macrophytes (0.12
                        m2/g);
    BiomassMacm    =   total biomass of macrophytes in system (g/m ); and
    Biomassperi      =   biomass of periphytic algae (g/m ).
Under optimal conditions, a reduction factor has a value of 1; otherwise, it has a fractional value.
Use of a multiplicative construct implies that the factors are independent.  Several authors (for
example, Collins, 1980; Straskraba and Gnauck, 1983) have shown that there  are interactions
among the factors.   However, we feel the data are insufficient  to generalize to all algae;
therefore, the simpler multiplicative construct is used, as in many other models (Chen and Orlob,
1975; Lehman et al., 1975; J0rgensen, 1976; Di Toro et al., 1977; Kremer and Nixon, 1978; Park
et al., 1985; Ambrose et al., 1991).  Default parameter values for the various processes are taken
primarily from compilations (for example, J0rgensen, 1979; Collins and Wlosinski, 1983; Bowie
et al., 1985); they may be modified as needed.
                                       68

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Light Limitation

Because it is  required for photosynthesis, light is a very  important limiting variable.   It is
especially important in controlling competition among plants with differing light requirements.
Similar to many other models (for example, Di Toro et al., 1971; Park et al., 1974, 1975, 1979,
1980; Lehman et al., 1975; Canale et al., 1975, 1976; Thomann et al., 1975, 1979; Scavia et al.,
1976;  Bierman  et  al.,  1980; O'Connor  et  al.,  1981),  AQUATOX uses the Steele  (1962)
formulation for  light limitation.  Light is specified as  average daily radiation.  The average
radiation is multiplied by the photoperiod, or the fraction of the day with  sunlight, based on a
simplification  of Steele's (1962) equation proposed by Di Toro et al. (1971).   The equation is
slightly different when the model is run with a daily versus an hourly time-step:
                          e • Photoperiod • (LtAtDepthn .,  -LtAtTopn ., )• PeriphytExt
        LtLimit Dailv = 0.85	^-^	^^	—	
                                      Extinct • (Depth Bottom - DepthTop )


               ,,,.  .,     =   g • (LtAtDepthHourly - LtAtTopHourly)- PeriphytExt
               LiLimii Hourl                                                            I-*")
                                     Extinct •( Depth Bottom-DepthTop)

where:
       LtLimitTimeStep =     light limitation (unitless);
       e            =     the base of natural logarithms (2.71828, unitless);
       Photoperiod =     fraction of day with daylight (unitless), see (26);
       Extinct      =     total light extinction (1/m), see (40), (41);
       DepthBottom   =     maximum depth or depth of bottom of layer if stratified (m);  if
                           periphyton or macrophyte then limited to euphotic depth;
       Depthrop     =     depth of top of layer (m);
       LtAtTop      =     limitation of algal growth  due to light, (unitless) see (44), (45);
       LtAtDepth    =     limitation due to insufficient light, (unitless), see (43);
       PeriphytExt  =     extinction due  to   periphyton;   only  affects  periphyton  and
                           macrophytes (unitless), see (42).

Because the equation overestimates by  15 percent the cumulative effect of light limitation over a
24-hour day, a correction factor of 0.85 is applied  to the daily formulation (Kremer and Nixon,
1978).  When AQUATOX is run with an hourly time-step, the correction factor of 0.85 is not
relevant, nor the inclusion of photoperiod.

Light limitation does not apply to free-floating macrophytes as these are assumed to be located at
the surface of the water.

Even when the model is run with an hourly time-step, two algal equations utilize the daily light
limit equation (38) as most appropriate. First, when calculating algal mortality, the stress factor
for suboptimal light and nutrients (68) is expecting the input of daily light limitation (i.e. the
plants do not all die each night). Secondly, when calculating the sloughing of benthic algae (75)
                                        69

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4


the calculation of suboptimal light is calibrated to daily light limitation, not the instantaneous
absence or presence of light (i.e. sloughing is not more likely to occur when it is dark).

Extinction of light is based on several additive terms: the baseline extinction coefficient for water
(which may include suspended sediment if it is  not modeled explicitly), the so-called "self-
shading" of plants, attenuation due to suspended parti culate organic matter (POM) and inorganic
sediment, and attenuation due to dissolved organic matter (DOM):

             Extinct = WaterExtinction + PhytoExtinction + ECoeffDOM • DOM
                         + ECoejfPOM-l
-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
                                       T  cv
                                 _TimeStep  V __ -Extinct VSeg • DepthBottom
         LtAtDepthTimeStep = e(      LigMSat-LightCorr      '*                  }     (43)


Light limitation at the surface of the water body is computed by:
                          LtAtTopTimeStgp = e LightSat-LightCorr                       (44)

and light limitation at the top of the hypolimnion is computed by:

                                  ^      LjghtT,meStep      -Extinct Epl -DepthTop
                LtAtTopTimeStep = e LightSat-LightCorr                              (45)


where:
       LtAtTop      =     limitation of algal growth due to light, (unitless multiplier, 0 being
                           no limitation, 1 being 100% limitation)
       LtAtDepth    =     limitation due to insufficient light, (unitless, see LtAtTop)
       Extinct       =     overall extinction of light in relevant vertical segment (1/m), (40)
       Light ''TimeStep  =     photosynthetically active radiation (ly/d), (46);
       LightCorr    =     Correction factor,  1.0 for a daily time-step, 1.25 for an  hourly
                           time-step.   LightSat  is  increased  by  25%  to  account  for
                           instantaneous solar radiation as opposed to daily averages;
       LightSat     =     light saturation level for photosynthesis (ly/d).

Phytoplankton not specified as "surface floating" are assumed to be mixed throughout the  well
mixed layer, although subject to sinking.  However, healthy cyanobacteria (and some other algal
species) tend to float.  Therefore, if the phytoplankton is specified as "surface floating" and the
nutrient limitation  is  greater  than 0.25 (Equation (55)) and  the wind is less than 3 m/s  then
DepthBottom for surface floating algae is set to 0.1 m to account for buoyancy.  Otherwise it is
set to  3 m to represent downward transport by Langmuir circulation.  When calculating  self-
shading for surface-floating algae the model accounts for more intense self shading in the upper
layer of the water  column due to the floating concentration of algae there.  The Extinct term in
equation (43) is multiplied by the segment thickness and divided by the thickness over which the
floating algae occur so that the more intense self-shading effects of these algae concentrated at
the top of the system are properly accounted for.  Rather than average the biomass of "surface
floating" plants over the entire water column, the biomass is normalized to the top 3 m to more
closely correspond with monitoring data.

Under the  ice, all  phytoplankton are represented as occurring in the top 2 m (cf LeCren and
Lowe-McConnell,  1980). As  discussed in Section 3.6, light is decreased to 15% of incident
radiation  if ice  cover  is  predicted.   Approximately  half the incident solar radiation  is
photosynthetically active (Edmondson, 1956):

                                LightTmeStep = SolarTmeStep • 0.5                            (46)
                                        71

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
where:
       Solar
            TimeStep
daily light intensity on a daily (25) or hourly (28) basis (ly/d).
The light-limitation function represents both  limitation for suboptimal light intensity and
photoinhibition at high light intensities (Figure 52).  When considered  over the course of the
year, photoinhibition can occur in very clear, shallow systems during summer mid-day  hours
(Figure 54), but it often is not a factor when considered over 24 hours (Figure 55).

To help understand the occurrence of photoinhibition as opposed to insufficient light, two new
output "photosynthetic limitation variables"  are available—"LowLt_LEVI" and "HighLt_LEVI."
These output variables are same as the overall light limitation factor (Lt_LEVI) but are modified
to indicate photoinhibition  as opposed to insufficient light.   When low-light limitation causes
light conditions to be  sub-optimal then the  "high-light  limitation"  is  set to  zero.  When
photoinhibition is occurring then the "low-light limitation"  is set to  zero. To determine this
difference AQUATOX differentiates the equations used to produce the curves in Figures 50 and
51 (see (38) and (39)) and and determines whether the current light is greater than or less than
the maximum value.

It is also worth noting that in simulations with a one-day time step, the light limitation factor
(Lt_LIM) represents a daily light limitation and is therefore subject to the photoperiod.  In other
words, if the sun is shining only 50% of the day, the maximum the LtLimit can be is 0.5. This is
because Lt_LEVI is a limitation  on the maximum  daily photosynthesis  rate for a plant which
would be based on 24-hours of light exposure.
The extinction coefficient for pure water varies considerably in the  photosynthetically-active
400-700 nm range (Wetzel, 1975, p. 55); a value of 0.016 (1/m) correspond to the extinction  of
green light.   In many models  dissolved  organic  matter  and suspended  sediment  are not
considered separately,  so a  much larger extinction coefficient is used  for "water"  than  in
AQUATOX.  The attenuation coefficients have  units of l/m-(g/m3) because they represent the
amount of extinction caused by a given concentration  (Table 7).
Table 7. Light Extinction and Attenuation Coefficients
WaterExtinction
ECoeffPhyto diatom
ECoeffPhytoblue.ereen
ECoeffDOM
ECoeffPOM
ECoeffSed
0.02 1/m
0.14 l/m-(g/m3)
0.099 l/m-(g/m3)
0.03 l/m-(g/m3)
0.12 l/m-(g/m3)
0.17 l/m-(g/m3)
Wetzel, 1975
calibrated
Megard et al., 1979 (calc.)
Effleretal., 1985 (calc.)
Verduin, 1982
Straskraba and Gnauck,
1985
All coefficients may be user-supplied in the plant or site underlying data.
                                        72

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                  CHAPTER 4
 Figure  52. Instantaneous  Light  Response
 Function
          Diatoms in 0.5-m Deep Pond
  j= 0.9
    0.88
       200  250  300  350  400  450  500
                   Light (ly/d)
 Figure 54. Mid-day Light Limitation
          Diatoms in 0.5-m Deep Pond
     500
     200
        3   53  103 153 203 253 303 353
                 Julian Date
      0.88
            — Light    —Limitation
                    Figure 53.  Daily Light Response Function
                            Diatoms in 0.5-m Deep Pond
                                                   0.55
                       0.25
                         200   250   300   350  400
                                  Average Light (ly/d)
                             450
                    Figure 55.  Daily Light Limitation
                             Diatoms in 0.5-m Deep Pond
                                                                                  0.55
200
                              53 103 153 203 253 303 353
                                    Julian Date
0.25
                               — Light     Limitation
The Secchi depth, the depth at which a Secchi disk disappears from view, is a commonly used
indication of turbidity. It is computed as (Straskraba and Gnauck, 1985):

                                    Secchi =	—	                                (47)
                                             Extinction
where:
       Secchi
Secchi depth (m).
This relationship also could be used to back-calculate an overall Extinction coefficient if only the
Secchi depth is known for a site.

It should be noted that although Secchi depth can be  computed for the hypolimnion  segment,
based on the suspended material, it is  a relatively meaningless value for the hypolimnion and
generally should be ignored.  Light extinction in the hypolimnion is calculated based on the light
that has first filtered through the epilimnion as shown in equation (45).
                                         73

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                             CHAPTER 4
As a verification of the extinction computations, the calculated and observed Secchi depths were
compared for Lake George, New York.  The  Secchi depth is estimated to be 8.3 m in Lake
George, based on site data for the various components (Figure 56).  This compares favorably
with observed values of 7.5 to 11 (Clifford, 1982).

                 Figure 56. Contributions to light extinction in Lake George NY.
                     POM (26.13%)
                                             Sediment (0.00%)
                                               rWater (6.97%)
                                             §S|W/-Phytoplankton (1.59%)
                                                         (65.32%)
Adaptive Light

Saturating  light can  be specified  as  a constant for each  plant taxonomic group  (classic
AQUATOX approach) or it can be adaptive based on Kremer and Nixon (1978) and similar to
the approach  used in EFDC.   The adaptive light  saturation  is the  weighted average  of
photosynthetically active solar radiation (PAR) at the optimal depth for growth of a given plant
group, using an approximation based on the user-specified light saturation and site solar radiation
and turbidity at the beginning of the simulation:
LightSatCalc =
                                                       st7}+ 0. l(LightHist3)
(48)
                             LightHistn=PAR-e
                                               (-Extinct
                                                                      (49)
where:
       LightSatCalc
       LightHistn

       PAR
       Solar
       Extinct
              adaptive light saturation (Ly/d)
              photosynthetically active radiation at optimum depth for plant
              growth n days prior to simulation date (Ly/d)
              photosynthetically active radiation, Solar * 0.5 (Ly/d)
              incident solar radiation (Ly/d)
              total light extinction computed dynamically (40).
                                       74

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


If the  LightSatCalc  is greater or less than the user-entered  maximum and minimum light
saturation coefficients ("Plant underlying data" screen) then the LightSatCalc is set to the user-
entered maximum or minimum.  This LightSatCalc variable is then used in the LtAtDepth and
LtAtTop calculations (43)-(45).

                                    \n(LightSat I MaxDailyLight)
                         ^Pt Plant =         ;T~~"                                  (5U)
                                          - ExtinctImtcond

where:
       ZOpt piant      =      optimum depth for a given plant (a constant approximated at the
                             beginning of the simulation in meters);
       LightSat       =      user entered light saturation coefficient (Ly/d);
       MaxDailyLight  =      maximum   daily-averaged  incident  solar  radiation for  one
                             calendar year forward from the start date (Ly/d);
       ExtinctMtCond   =      initial condition total light extinction (unitless);
Nutrient Limitation

There are several  ways that nutrient limitation  has  been represented in models.  Algae  are
capable of taking up and storing sufficient nutrients to carry them through several generations,
and models have been developed to represent this.  However, if the timing of algal blooms is not
critical, intracellular storage of nutrients can be ignored, constant stoichiometry can be assumed,
and the model is much simpler.  Therefore, based  on the efficacy of this simplifying assumption,
nutrient limitation  by external nutrient concentrations is used in AQUATOX, as in many other
models  (for example, Chen,  1970; Parker, 1972; Lassen and Nielsen, 1972; Larsen et al., 1974;
Park et al., 1974; Chen and  Orlob, 1975; Patten et al., 1975; Environmental Laboratory, 1982;
Ambrose et al.,  1991).

For an individual nutrient, saturation kinetics is assumed, using the Michaelis-Menten or Monod
equation (Figure 57); this approach is founded on numerous studies (cf. Hutchinson, 1967):

                                          Phosphorus                               .,...
                               PLmnt =	                            (51)
                                        Phosphorus + KP

                                            Nitrogen
                                NLimit = •
                                         Nitrogen + KN

                                             Carbon
                                CLimit =	
                                        Carbon + KCO2                             (53)
where:
       PLimit       =      limitation due to phosphorus (unitless);
       Phosphorus   =      available soluble phosphorus (gP/m3);
       KP          =      half-saturation constant for phosphorus (gP/m3);
                                       75

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
       NLimit
       Nitrogen
       KN
       CLimit
       Carbon
       KC02
 limitation due to nitrogen (unitless);
 available soluble nitrogen (gN/m3);
 half-saturation constant for nitrogen (gN/m3);
 limitation due to inorganic carbon (unitless);
 available dissolved inorganic carbon (gC/m3); and
 half-saturation constant for carbon (gC/m3).

	Figure 57. Nutrient limitation	
                       MICHAELIS-MENTEN RELATIONSHIP
                                      DIATOMS
                        0.00 0.01 0.03  0.04 0.05 0.07 0.08  0.09
                                     PHOSPHATE (mg/L)
Nitrogen fixation in cyanobacteria is handled by setting NLimit to 1.0 if Nitrogen is less than half
the KN value.  Otherwise, it is assumed that nitrogen fixation is not operable, and NLimit is
computed as for the other algae.  AQUATOX also provides an option to trigger nitrogen fixation
as a function of an input parameter,  the ratio of inorganic N to inorganic P, which may be
selected and specified in the "Study Setup" screen.  When the ratio falls below the threshold,
nitrogen fixation is assumed to occur; the default threshold N:P is 7.

Concentrations must be expressed in terms of the chemical element.  Because carbon dioxide is
computed  internally, the concentration of carbon is  corrected for the molar  weight of the
element:
                                 Carbon = C2CO2-CO2
                                                        (54)
where:
       C2CO2
       C02
 ratio of carbon to carbon dioxide (0.27); and
 inorganic carbon (g/m3).
Like many models (for example, Larsen et al., 1973; Baca and Arnett, 1976; Scavia et al., 1976;
Smith, 1978; Bierman  et al., 1980; Park et al., 1980; Johanson et al.,  1980; Grenney  and
Kraszewski, 1981;  Ambrose et  al.,  1991), AQUATOX uses the minimum limiting nutrient,
whereby the Michaelis-Menten equation  is evaluated for each nutrient, and the factor  for the
nutrient that is most limiting at a particular time is used:
                                       76

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


                         NutrLimit = min(PLimit, NLimit, CLimit)                      (55)
where:
       NutrLimit    =      reduction due to limiting nutrient (unitless).

Alternative  formulations used  in  other models  include  multiplicative  and harmonic-mean
constructs,  but the minimum limiting nutrient construct is well-founded in laboratory studies
with individual species.

Current Limitation

Because they are fixed in  space,  periphyton also are limited by slow  currents  that do not
replenish nutrients and carry away senescent biomass. Based on the work of Mclntire (1973)
and Colby  and Mclntire (1978), a factor relating photosynthesis to current velocity is used for
periphyton:
                   VLimit - minr/, RedStrllWater +  V^Coeff-Velocity
                                                1+ VelCoeff-Velocity
where:
       VLimit       =      limitation or enhancement due to current velocity (unitless);
       RedStillWater =      user-entered reduction in  photosynthesis in absence  of current
                           (unitless);
       VelCoeff     =      empirical proportionality coefficient for velocity (0.057, unitless);
                           and
       Velocity      =      flow rate (converted to m/s), see (14).

VLimit has  a minimum  value  for photosynthesis in  the absence of currents  and increases
asymptotically to a  maximum value for optimal current velocity (Figure 58).  In high currents
scour can limit periphyton; see (75).  The value of RedStillWater depends on the circumstances
under which the maximum photosynthesis rate was measured; if PMax was measured in still
water then RedStillWater = 1, otherwise a value of 0.2 is appropriate (Colby and Mclntire, 1978).
                                       77

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
               Figure 58. Effect of current velocity on periphyton photosynthesis.
Of.
O 1
H ]
i
H
z n ft
£ °-8
111
O
z
*t n e
z
ui
K
O
-» n ^
REDUCTION
0 C
ro 4
o




/
/
/


/
/
f


I

/
/






/






^







-~~"






















































20 40 60 80 100 120
VELOCITY (cm/s)
Adjustment for Suboptimal Temperature

AQUATOX uses a general but complex formulation to represent the effects of temperature.  All
organisms  exhibit a  nonlinear,  adaptive  response  to temperature  changes  (the  so-called
Stroganov  function).  Process  rates other than  algal  respiration increase  as the ambient
temperature increases  until the optimal temperature for the organism is reached; beyond that
optimum,  process rates decrease until the  lethal  temperature is reached.    This  effect is
represented by a complex algorithm developed by O'Neill et al. (1972) and modified slightly for
application to aquatic systems (Park et al., 1974). An intermediate variable VTis computed first;
it is the ratio of the difference between the maximum temperature at which a process will occur
and the ambient temperature over the difference  between the maximum temperature and the
optimal temperature for the process:
                    VT = -
where:
       Temperature
       TMax
       TOpt
       Acclimation
    (TMax + Acclimation) - Temperature
(TMax + Acclimation) - (TOpt + Acclimation)
 ambient water temperature (deg. C);
 maximum temperature at which process will occur (deg. C);
 optimal temperature for process to occur (deg. C); and
 temperature acclimation (deg. C), as described below.
(57)
Acclimation to both increasing and decreasing temperature is accounted for with a modification
developed by Kitchell et al. (1972):
                       Acclimation = XM-[1- e(~KT' ABS(T^P^^-TR^J                   (58)
 where:
                                      78

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       XM     =    maximum acclimation allowed (2.0 deg. C);
       KT     =    coefficient for  decreasing acclimation  as temperature  approaches Tref
                    (value is 0.5 and unitless);
       ABS    =    function to obtain absolute value; and
       TRef    =    "adaptation" temperature below which there is no acclimation (deg. C).
The mathematical sign of the variable Acclimation is negative if the ambient temperature  is
below the temperature at which there is no acclimation; otherwise, it is positive.

If the variable VTis less than zero, in other words, if the ambient temperature exceeds (TMax +
Acclimation), then the suboptimal factor for temperature is set equal to zero and the process
stops. Otherwise, the suboptimal factor for temperature is calculated as (Park et al., 1974):
                                       = VTXT-e(Kr'(1-VT))                            (59)
where:
                                                       _
                                            400
where:

                WT = \n(Q10) • ((TMax + Acclimation) - (TOpt + Acclimation))            (61)
and,

               YT = \n(Q10) • ((TMax + Acclimation) - (TOpt + Acclimation) + 2)           (62)

where:

       Q10   =     slope or rate of change per 10°C temperature change (unitless).

This well-founded, robust algorithm for TCorr is used in AQUATOX to obtain reduction factors
for suboptimal temperatures for all biologic processes in animals and plants, with the exception
of decomposition and plant respiration. By varying the parameters, organisms with both narrow
and broad temperature tolerances can be represented (Figure 59, Figure 60).
                                       79

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                          CHAPTER 4
Figure 59. Temperature response of cyanobacteria  Figure 60.  Temperature response of diatoms
           STROGANOV FUNCTION
               BLUE-GREENS
                              TOpt
             10     20     30
                TEMPERATURE (C)
                                  40
                                                      STROGANOV FUNCTION
                                                             DIATOMS
 10      20      30
   TEMPERATURE (C)
40
Algal Respiration

Endogenous or dark respiration is the metabolic process whereby oxygen is taken up by plants
for the  production of energy for maintenance and carbon dioxide  is released  (Collins and
Wlosinski, 1983). Although it is normally a small loss rate for the organisms, it has been shown
to be exponential with temperature (Aruga, 1965). Riley (1963, see also Groden, 1977) derived
an equation representing this relationship.  Based on data presented by Collins (1980), maximum
respiration is constrained to 60% of photosynthesis. Laboratory experiments in support of the
CLEANER model confirmed the empirical relationship and provided additional evidence of the
correct parameter values (Collins, 1980), as demonstrated by Figure 61:
                     Respiration = Resp20 • 1.045
                                              (Temperature-20
• Biomass
                                                                                  (63)
where:
       Respiration   =
       Resp20
       1.045
       Temperature  =
       Biomass      =
                          dark respiration (g/m3-d);
                          user input respiration rate at 20°C (g/g-d);
                          exponential temperature coefficient (/°C);
                          ambient water temperature (°C); and
                    =     plant biomass (g/m ).

This construct also applies to macrophytes.
                                       80

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                     CHAPTER 4
                       Figure 61. Respiration (Data From Collins, 1980)
                                 DARK RESPIRATION
                                   10
                                       20        30
                                 TEMPERATURE(C)
40
Photorespiration

Algal excretion, also referred to as photorespiration, is the release of photosynthate (dissolved
organic material) that occurs in the presence of light.  Environmental conditions that inhibit cell
division but  still allow photoassimilation  result  in release  of organic compounds.  This is
especially true for both low and high levels of light (Fogg et al., 1965; Watt, 1966; Nalewajko,
1966; Collins, 1980). AQUATOX uses an equation modified from  one by Desormeau (1978)
that is the inverse of the light limitation:
where:
               Excretion = KResp • LightStress • Photosynthesis

Excretion     =     release of photosynthate (g/m3-d);
KResp        =     coefficient   of   proportionality   between   excretion
                    photosynthesis at optimal light levels (unitless); and
                    photosynthesis (g/m3-d), see (35),
                                                                                    (64)
                                                                                     and
       Photosynthesis =

and where:
where:
       LtLimit
                          LightStress = 1- LtLimit

                    light limitation for a given plant (unitless), see (38).
                                                                                    (65)
Excretion is a continuous function  (Figure 62) and has a tendency to overestimate excretion
slightly at light levels close to light  saturation where experimental evidence suggests a constant
relationship (Collins, 1980).  The construct for photorespiration also applies to macrophytes.
                                        81

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                          CHAPTER 4
                     Figure 62. Excretion as a fraction of photosynthesis
                    EFFECT OF LIGHT ON PHOTORESPIRATION
                   w             DIATOMS IN POND
                   ftO.09
                         200   250
                      300    350   400
                         LIGHT (ly/d)
450    500
Algal Mortality

Nonpredatory algal mortality can occur as a response to toxic chemicals (discussed in Chapter
8) and as a response to unfavorable environmental conditions.  Phytoplankton under stress may
suffer greatly increased mortality due to autolysis and parasitism (Harris, 1986). Therefore, most
phytoplankton decay occurs in the water column rather than in the sediments (DePinto,  1979).
The rapid remineralization of nutrients in the water column may result in a succession of blooms
(Harris,  1986).  Sudden changes in the abiotic environment may cause the algal population to
crash; stressful changes include nutrient depletion, unfavorable temperature, and damage by light
(LeCren and Lowe-McConnell, 1980). These are represented by a mortality term in AQUATOX
that includes toxicity,  high temperature (Scavia and Park, 1976), and combined nutrient and
light limitation (Collins and Park, 1989):
where:
      Mortality
      Poisoned
      KMort
      Biomass
Mortality = (KMort + Excess T + Stress) • Biomass + Poisoned

    =      nonpredatory mortality (g/m3-d);
    =      mortality rate due to toxicant (g/m3-d), see (417);
    =      intrinsic  mortality rate (g/g-d); and
    =      plant biomass (g/m3),
                                                                                  (66)
and where:
and:
                              Excess! = •
                                          (Temperature - TMax)
                                               2
                       „     _ 7 _  -EMort • (1 - (NutrLimit • LtLimit))
                                                                  (67)
                                                                                  (68)
                                      82

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 4
where:
       Excess!  =   factor for high temperatures (g/g-d);
       TMax     =   maximum temperature tolerated (° C);
       Stress     =   factor for suboptimal light and nutrients (g/g-d),
       Emort     =   approximate maximum  fraction  killed  per  day with  total limitation
                     (g/g-d);
       NutrLimit =   reduction due to limiting nutrient (unitless), see (55)
       LtLimit   =   light limitation (unitless), see (38).

Exponential functions are used so that increasing stress leads  to rapid increases in mortality,
especially with high temperature where mortality is 50% per day at the TMax (Figure 61), and, to
a much lesser degree, with suboptimal nutrients and light (Figure 64). This simulated process is
responsible in part for maintaining realistically high levels of detritus in the simulated water
body. Low temperatures are assumed not to affect algal mortality.
 Figure 63. Mortality due to high temperatures
   raO.8
  T3
  "£
  | 0.6
   ro
  HI
                                /
                                  TMax
      24
          26
              28
                  30   32   34   36
                   Temperature
                                   38
                                       40
                        Figure 64. Mortality due to light limitation
                                                           ALGAL MORTALITY
                                                                DIATOMS
                                                  0.03
                                   250    300     350
                                           LIGHT (ly/d)
                                                      400
                                                             450
Sinking

Sinking of phytoplankton,  either between layers or to the bottom sediments, is modeled as a
function  of physiological state,  similar to  mortality.  Phytoplankton that are not stressed are
considered to sink at given rates, which are based on field observations and implicitly account
for the effects of averaged water movements (cf Scavia, 1980).  Sinking also is represented as
being impeded by turbulence associated with higher discharge (but only when discharge exceeds
mean discharge):
                     Sink =
   KSed MeanDischarge
   Depth    Discharge
• SedAccel • Biomass
(69)
where:
       Sink
=  phytoplankton loss due to settling (g/m3-d);
                                        83

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 4
       KSed
       Depth
       MeanDischarge
       Discharge
       Biomass
 =  intrinsic settling rate (m/d);
 =  depth of water or, if stratified, thickness of layer (m);
 =  mean annual discharge (m3/d);
 =  daily discharge (m3/d), see Table 3; and
 =  phytoplankton biomass (g/m3).
The model is able to mimic high sedimentation loss associated with the crashes of phytoplankton
blooms, as discussed by Harris  (1986).  As  the  phytoplankton are stressed by toxicants  and
suboptimal light, nutrients, and  temperature,  the  model computes an exponential increase in
sinking (Figure 65), as observed by Smayda (1974), and formulated by Collins and Park (1989):
                  0  , A   ,	  ESed • (!- LtLimit • NutrLimit • TCorr • FracPhoto)
                  SedAccel — e
                                                             (70)
where:
       SedAccel
       ESed
       LtLimit
       NutrLimit
       FracPhoto

       TCorr
    increase in sinking due to physiological stress (unitless);
    exponential settling coefficient (unitless);
    light limitation (unitless), see (38);
    nutrient limitation (unitless), see (55); and
    reduction factor for effect of toxicant on photosynthesis (unitless),
    see (421);
    temperature limitation (unitless), see (59).

Figure 65. Sinking as a function of nutrient stress
                                    SINKING IN POND
                             DIATOMS, DEPTH = 3 m, BIOMASS = 1
                        (O
                            0.00  0.02  0.03   0.05  0.06  0.08  0.10
                                       PHOSPHATE (g/cu m)
Washout and Sloughing

Phytoplankton are subject to downstream drift. In streams and in lakes and reservoirs with low
retention times this may be a significant factor in reducing or even precluding phytoplankton
populations (LeCren and Lowe-McConnell, 1980).  The process is modeled as a simple function
of discharge:
                                           Discharge
                                 I phytoplankton    -,T  T
                                             Volume
                                                     • • Biomass
                                                             (71)
                                        84

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
where:
       Washoutphytopiankton    =     loss due to downstream drift (g/m3-d),
       Discharge           =     daily discharge (m3/d);
       Volume              =     volume of site (m3); and
       Biomass             =     biomass of phytopiankton (g/m3).

Periphyton often  exhibit a pattern of buildup and then a  sharp decline in  biomass due to
sloughing.  Based on extensive experimental data from Walker Branch, Tennessee (Rosemond,
1993), a complex sloughing formulation, extending the approach of Asaeda and Son (2000), was
implemented. This function was able to represent a wide range of conditions better (Figure 66
and Figure 67).

                         Washout penphyton = Slough + DislodgePenjox                      (72)
where:
       Washoutpenphyton   =    loss due to sloughing (g/m3-d);
       Slough          =    loss due to natural causes (g/m3-d), see (75); and
       DislodgePeri, TOX    =   loss due to toxicant-induced sloughing (g/m3-d), see (427).
                                       85

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                        CHAPTER 4
Figure 66. Comparison of predicted biomass of periphyton, constituent algae, and observed biomass of
periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of both N and P and removal
of grazers in Spring, 1989.
                             55  T^
             a
             §
                                                                            I
                       -Diatoms(g/m2) —•— Oth alg(g/m2)
Periphyton
                                                                Observed
                                       86

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
Figure 67. Predicted rates for diatoms in Walker Branch, Tennessee, with addition of both N and P and
removal of grazers in Spring, 1989. Note the importance of periodic sloughing. Rates expressed as g/m2 d.
     16
     U
     12
     10
   2
   a>
   £
   re
       1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 2526 27 28 29 3031 32 33 34 35 36 37 38 3940 41 42 43 44 4546 4748
                     0Photosynthesis n Respiration D Excretion C Mortality IPredation • Sloughing
Natural sloughing is a function of senescence due to suboptimal conditions and the drag force of
currents acting on exposed biomass. Drag increases as both biomass and velocity increase:
              DragForce = Rho • DragCoeff • Vel2 • (BioVol • UnitArea)2'3 -IE-6
                                                          (73)
where:
       DragForce
       Rho
       DragCoeff
       Vel
       BioVol
       UnitArea
       1E-6
drag force (kg m/s );
density of water (kg/m3);
drag coefficient (2.53E-4, unitless);
velocity (converted to m/s) see (14);
biovolume of algae (mm3/mm2);
unit area (mm2);
conversion factor (m2/mm2).
Biovolume is not modeled  directly by AQUATOX,  so a simplifying assumption is that the
empirical  relationship between biomass and biovolume is constant for a given growth  form,
based on observed data from Rosemond (1993):
                                        87

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
                                        Biomass  „, ,
                              BiovolDia = -TTT^T'ZMean
                                        2.08E-9                                  (?4)
                                         Biomass  _, ,
                              Biovolm  =         • ZMean
where:
                                                               T    /^
       Biovol^ia     =     biovolume of non-filamentous algae (mm /mm );
       Biovolpn      =     biovolume of filamentous algae (mm3/mm2);
       Biomass      =     biomass of given algal group (g/m2);
       ZMean       =     mean depth (m).

Suboptimal light, nutrients, and temperature cause senescence of cells that bind the periphyton
and keep them attached to the substrate.   This effect is represented by a factor, Suboptimal.,
which  is computed in  modeling the effects of environmental conditions on photosynthesis.
Suboptimal decreases the critical force  necessary to cause sloughing.  If the drag force exceeds
the critical force for a given algal group  modified by the Suboptimal factor and an adaptation
factor, then sloughing occurs:

                   If DragForce > Suboptimal0rg • FCrit0rg • Adaptation
                   then Slough = Biomass • FracSloughed
                   else Slough = 0                                                  /^

where:
       Suboptimalorg   =   factor for  Suboptimal  nutrient,  light, and  temperature effect  on
                          senescence of given periphyton group (unitless);
       FCritorg        =   critical  force necessary  to  dislodge  given  periphyton  group (kg
                          m/s );
       Adaptation      =   factor to adjust for mean discharge of site  compared to reference
                          site (unitless);
       Slough         =   biomass lost by sloughing (g/m3);
       FracSloughed   =   fraction of biomass lost at one time, editable.

                  Suboptimal0  = NutrLimit0  • LtLimit0   • TCorr0 •  20
                                                                                  (76)
                  If Suboptimal0rg  > 1  then Suboptimal0rg = 1
where:
       NutrLimit     =     nutrient  limitation for given algal group (unitless) computed  by
                          AQUATOX;  see (55);
       LtLimitorg     =     light limitation for given  algal group  (unitless)  computed  by
                          AQUATOX;  see (38); and
       TCorr        =     temperature limitation for a given algal group (unitless) computed
                          by AQUATOX; see (59).
       20           =     factor to desensitize construct.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
The  sloughing construct was  tested  and  calibrated  (U.S.  E.P.A.,  2001)  with data from
experiments with artificial and woodland streams in Tennessee (Rosemond, 1993, Figure 66).
However, in modeling periphyton at several  sites, it was observed that sloughing appears to be
triggered at greatly differing mean velocities.  The working hypothesis is that periphyton adapt to
the ambient conditions of a particular channel. Therefore,  a factor is included to adjust for the
mean discharge of a given site compared to the reference site in Tennessee.  It is still necessary
to calibrate FCrit for each site  to account for  intangible differences in channel and flow
conditions, analogous to the calibration of shear  stress by  sediment modelers, but the range of
calibration needed is reduced by the Adaptation factor:

                                                Vel2
                                 Adaptation =	
                                              0.006634                              (77)
where:
       Vel           =      velocity for given site (m/s), see (14);
       0.006634      =      mean velocity  for reference experimental stream (m/s).
Detrital Accumulation in Periphyton

In phytoplankton,  mortality results in  immediate production of detritus,  and that transfer is
modeled. However, for purposes  of modeling, periphyton are defined as including associated
detritus.   The accumulation  of non-living biomass is modeled implicitly by not simulating
mortality due to suboptimal conditions. Rather, in the simulation biomass builds up, causing
increased self-shading, which in turn makes the periphyton more vulnerable to sudden loss due
to sloughing. The fact that part of the biomass is non-living is ignored as a simplification of the
model.
Chlorophyll a

Chlorophyll a is not simulated directly.  However, because chlorophyll a is commonly measured
in aquatic systems and because water quality managers are accustomed to thinking of it as an
index of water  quality, the model converts phytoplankton biomass estimates into approximate
values for chlorophyll a.  The ratio of carbon to chlorophyll a exhibits a wide range of values
depending on the nutrient status of the algae (Harris, 1986); cyanobacteria often have higher
values (cf  Megard et al., 1979).  AQUATOX uses a value of 45 ugC/ug  chlorophyll a for
cyanobacteria and  a value of 28 for other phytoplankton as reported in the documentation for
WASP (Ambrose et al., 1991).  The values are more representative for blooms than for static
conditions, but managers are usually most interested in the maxima. The results are presented as
total chlorophyll a in ug/L; therefore, the computation is:


                       omassBlar          ,  \':— 'Biomass Diatom ^-'Biomassoth' _ o_  1 000
                                                                      '
                            45                       28
where:
                                       89

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       ChlA         =     biomass as chlorophyll a (ug/L);
       Biomass      =     biomass of given alga (mg/L);
       CToOrg      =     ratio of carbon to biomass (0.526, unitless); and
       1000         =     conversion factor for mg to ug (unitless).

Periphytic chlorophyll a is computed as a conversion from the ash-free dry weight (AFDW) of
periphyton; because periphyton can collect inorganic sediments, it is important to measure and
model it as AFDW.  The conversion factor is based on the observed average ratio of chlorophyll
a to AFDW for the  Cahaba River near Birmingham, Alabama (unpub.  data)  and  also based on
data published in Biggs (1996)and Rosemond (1993).

                                Perichlor = AFDW- 5.0                             (79)

where:
                                                      r\
       PeriChlor    =     periphytic chlorophyll a (mg/m );
       AFDW       =     ash free dry weight (g/m ).
Moss chlorophyll a is output for all plants designated with the plant type "Bryophytes." In this
case, ash free dry weight is multiplied by 8.91 to get the estimate of chlorophyll a in mg/m2
(Stream Bryophyte Group, 1999, p. 160).  Total benthic chlorophyll a is also output in units of
mg/m2 (the sum of periphyton and moss chlorophyll a).
Phytoplankton and Zooplankton Residence Time

Phytoplankton and zooplankton can quickly wash out of a short reach, but they may be able to
grow over an extensive reach of a river, including its tributaries. Somehow the volume of water
occupied by the phytoplankton needs to be taken into consideration.  To solve this problem,
AQUATOX takes into account the "Total Length" of the river being simulated, as opposed to the
length  of the river reach, or "SiteLength"  so that phytoplankton and zooplankton production
upstream can be estimated.  This parameter can be directly entered on the "Site Data" screen or
estimated from the watershed area based on Leopold et al. (1964).

                       TotLength = \.609-lA-(WaterShed-0386)06                   (80)

where:
       TotLength  =   total river length (km);
       Watershed  =   land surface area contributing to flow out of the reach (square km);
       1.609      =   km per mile;
       0.386      =   square miles per square km.

If Enhanced Phytoplankton Retention is not chosen (or the total length or watershed area is
entered as zero,) the phytoplankton and zooplankton residence time equations are not used and
Equations (71) and (129) are used to calculate washout.   In this case, the phytoplankton
residence time is equal to the retention time of the system.
                                       90

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Otherwise, to simulate the inflow of plankton from upstream reaches plankton upstream loadings
are estimated as follows:
                                              [      Washoutbio
                                               TotLength I SiteLength
Loadingupstream = Washoutblota - \ ^-T	TTT^f	 \              (81)
where:
       Loadingupstream   =  loading of plankton due to upstream production (mg/L);
       Washout biota     =  washout of plankton from the current reach (mg/L);
       TotLength       =  total river length (km);
       SiteLength       =  length of the modeled reach (km).

An integral assumption in this approach is that upstream reaches included in the total river length
have identical environmental conditions as the reach being modeled and that plankton production
in each mile up-stream will be identical  to plankton production in the given reach.  Residence
time for plankton within the total river length is estimated as follows:

                                  _   Volume  (TotLength |
                            residence
                                    Discharge I SiteLength
where:
       Residence          =  residence time for floating biota within the total river length (d);
       Volume          =  volume of modeled segment reach (m3); see (2);
       Discharge       =  discharge of water from modeled reach (m3/d); see Table 3;
       TotLength       =  total river length (km);
       SiteLength       =  length of the modeled reach (km).

Periphyton-Phytoplankton Link

Periphyton may slough or be physically scoured, contributing to the suspended (sestonic) algae;
this may be reflected in the  chlorophyll a observed in the water column. Periphyton may be
linked to  a phytoplankton compartment so that sestonic  chlorophyll a reflects the results of
periphyton sloughing.  One-third of periphyton is assumed to become phytoplankton and two
thirds is assumed to become suspended  detritus in a sloughing event.  The default is linkage to
detritus with a warning.

Additionally, when phytoplankton undergoes sedimentation it will now be incorporated into the
linked periphyton layer if such a linkage exists. If multiple periphyton species are linked to a
single phytoplankton species, biomass is distributed  to periphyton weighted by the mass of each
periphyton compartment.  (A single periphyton compartment  cannot be  linked  to multiple
phytoplankton compartments.)
                                       91

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4
                                                    Periphyton A
                                                                                   io-t\
                                                                                   (83)
where:
       SedperiphytonA     =  sedimentation that goes to periphyton compartment A;
       Sinkphyto         =  total sedimentation of linked phytoplankton compartment, see (69);
       Mass periphyton A    =  mass of periphyton compartment A;
       Mass AH Linked Peri   =  mass of all periphyton compartments linked to the
                          relevant phytoplankton compartment.

If no linkage is present,  settling phytoplankton are assumed to contribute to sedimented detritus.

4.2 Macrophytes

Submersed aquatic vegetation or macrophytes can be an important component of shallow aquatic
ecosystems. It is not unusual for the majority of the biomass in an ecosystem to be in the form of
macrophytes  during  the  growing  season.    Seasonal  macrophyte   growth,  death,  and
decomposition can affect nutrient cycling, and detritus and oxygen concentrations.  By forming
dense cover, they can modify habitat and provide protection from predation for invertebrates and
smaller fish (Howick et al., 1993); this  function is represented in AQUATOX (see Figure 73).
Macrophytes  also  provide  direct and indirect food sources for many  species of waterfowl,
including swans, ducks, and coots (Jupp  and Spence, 1977b).

AQUATOX represents rooted macrophytes as occupying the littoral zone, that area of the bottom
surface that occurs within the euphotic zone (see (11) for computation).  Similar to periphyton,
the macrophyte  compartment has units of g/m . In nature, macrophytes can be greatly reduced if
phytoplankton blooms or higher levels of detritus increase the turbidity of the water (cf.  Jupp and
Spence, 1977a).   Because the depth of the euphotic zone is  computed as a  function of the
extinction  coefficient (ZEuphotic =  4.605/Extinci).,  the  area predicted  to be occupied by
macrophytes can increase or decrease depending on the clarity of the water.

The macrophyte equations are  based  on submodels developed for the International Biological
Program (Titus  et al., 1972; Park et al., 1974) and CLEANER models (Park et al., 1980) and for
the Corps of Engineers'  CE-QUAL-R1 model (Collins et al.,  1985):

               dBiomass      ,     „,       ,                  „
               - = Loading + Photosynthesis - Respiration - Excretion

                             - Mortality - Predation - Breakage                        (84)
                           + Washout Preeploat - Washinpreeploat
and:

               Photosynthesis  = PMax • LtLimit • TCorr • Biomass • FracLittoral
                                                                                   (85)
                           • NutrLimit • FracPhoto • HabitatLimit
where:
                                       92

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
       dBiomass/dt
       Loading
       Photosynthesis
       Respiration
       Excretion
       Mortality
       Predation
       Breakage
       PMax
       LtLimit
       TCorr
       HabitatLimit

       FracLittoral
       NutrLimit

       FracPhoto
       Washout
               FreeFloat
       Washin
              FreeFloat
change in biomass with respect to time (g/m -d);
loading of macrophyte, usually used as a "seed" (g/m2-d);
                         r\
rate of photosynthesis (g/m -d);
                   r\
respiratory loss (g/m -d), see (63);
excretion or photorespiration(g/m2-d), see (64);
                          r\
nonpredatory mortality (g/m  -d), see (87);
herbivory (g/m2-d), see (99);
loss due to breakage (g/m2-d), see (88);
maximum photosynthetic rate (1/d);
light limitation (unitless), see (38);
correction for suboptimal temperature (unitless), see (59);
in streams, habitat  limitation based on plant habitat preferences
(unitless), see (13);
fraction of bottom that is in the euphotic zone (unitless) see (11);
nutrient limitation for bryophytes  or freely-floating  macrophytes
(unitless), see (55);
reduction factor for effect of toxicant on photosynthesis (unitless),
see (421);
washout of freely floating macrophytes, see (86); and
loadings from linked upstream segments (g/m3-d), see (30);
They share  many  of the constructs  with the  algal submodel described  above.   Temperature
limitation is modeled similarly,  but  with different parameter values.  Light limitation also is
handled similarly, using the Steele (1962) formulation; the application of this equation has been
verified with laboratory data (Collins et al., 1985). Periphyton are epiphytic in the presence of
macrophytes;  by  growing on the  leaves they  contribute  to  the light  extinction for  the
macrophytes (Sand-Jensen,  1977).   Extinction due to  periphyton biomass is computed  in
AQUATOX, by inclusion in LtLimit.  For rooted macrophytes, nutrient limitation is not modeled
at this time because  macrophytes can obtain  most of their nutrients from bottom sediments
(Bristow and Whitcombe, 1971; Nichols and Keeney, 1976; Barko and Smart, 1980). Bryophytes
and  freely floating macrophytes  assimilate  nutrients  from water  and  are  subject to nutrient
limitation.

Release 3 includes free-floating macrophytes.   These macrophytes are assumed to be floating at
the  upper layer  of  the water  column and  therefore  are  not  subject to light  limitation.
Furthermore,  free-floating  macrophytes are  not subject  to the FracLittoral limitation  to
macrophyte photosynthesis (85).  On the other hand the washing of macrophytes out  of the
system is affected by the carrying capacity for the species:
                Washoutfn
                        freefloat
                                  KCap / ZMean - State \  Discharge
                                      KCap/ZMean
                               Volume
                                                         (86)
where:
                                        93

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                  CHAPTER 4
               freefloat
       State
       KCap
       ZMean
       Discharge
       Volume
loss due to being carried downstream (g/m3 -d),
concentration of dissolved or floating state variable (g/m3),
carrying capacity (g/m2);
mean depth from site underyling data (m);
discharge (m3/d), see Table 3; and
volume of site (m3), see (2);
Simulation of macrophyte respiration and excretion utilize the same equations as algae; excretion
in rooted macrophytes results in "nutrient pumping" because the nutrients are assumed to come
from the sediments but are excreted to the water column1. Non-predatory mortality is modeled
similarly to algae as a function of suboptimal temperature (but not light).  However, mortality is
a function of low as well as high temperatures, and winter die-back is represented as a result of
this control; the response is the inverse of the temperature limitation (Figure 68):
                Mortality = [KMort + Poisoned + (l-e
                                                    -EMort'(l-TCorr)
                                   )] • Biomass
(87)
where:
        KMort
        Poisoned
        EMort
intrinsic mortality rate (g/g-d);
mortality rate due to toxicant (g/g-d) (417), and
maximum mortality due to suboptimal temperature (g/g-d).
Sloughing of dead leaves can be a significant loss (LeCren and Lowe-McConnell,  1980); it is
simulated as an implicit result of mortality (Figure 68).
                        Figure 68. Mortality as a function of temperature
                                 MACROPHYTE MORTALITY
                                    10     20     30     40
                                        TEMPERATURE (C)
Macrophytes are subject to breakage due to higher water velocities; this breakage of live material
is different from the sloughing of dead leaves.  Although breakage is a function of shoot length
1 Because nutrients are not usually explicitly modeled in bottom sediments, macrophyte root uptake can result in
loss of mass balance, particularly in shallow ponds. The optional sediment diagenesis model does include nutrients
but linkage to macrophytes through root uptake has not yet been specified and implemented.  However, the total
mass of nutrients taken into the water column through macrophyte uptake can be tracked as a model output (N and P
"Root Uptake" in kg).
                                        94

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                         CHAPTER 4
and growth form as well as currents (Bartell et al., 2000; Hudon et al., 2000), a simpler construct
was developed for AQUATOX (Figure 69):
                                     Velocity - VelMax
where:
       Breakage
       Velocity
       VelMax
       Gradual
       UnitTime
       Biomass
                         Breakage =
                                    Gradual • UnitTime
                                   • Biomass
(88)
  =     macrophyte breakage (g/m2 -d);
  =     current velocity (cm/s) see (14);
  =     velocity at which total breakage occurs (cm/s);
  =     velocity scaling factor (20 cm/s);
  =     unit time for simulation (1 d);
  =     macrophyte biomass (g/m2).

Figure 69.  Breakage of macrophytes as a function of current
           velocity; VelMax set to 300 cm/s.
                   CD
                                         Velocity (cm/s)
The Breakage formulation also applies to freely floating macrophytes and may be considered
entrainment in periods of high flow. As such, VelMax, should be set to a relatively high value for
these organisms.

Bryophytes (mosses  and liverworts) are a special  class  of macrophytes that attach to hard
substrates,  are  stimulated by and take up nutrients directly from the water, are resistant to
breakage, and decompose very slowly (Stream Bryophyte  Group, 1999).  Nutrient limitation is
enabled when the "Bryophytes" plant type is selected, just as it is for algae. The model assumes
that when a bryophyte breaks or dies the result is 75% particulate and 25% dissolved refractory
detritus; in contrast,  other macrophytes are assumed to yield  62% labile detritus.   All other
differences  between bryophytes and other macrophytes in AQUATOX are based on differences
in parameter values.  These include low saturating light levels,  low optimum temperature, very
low mortality rates, moderate resistance to breakage, and resistance to herbivory (Arscott et al.,
1998; Stream Bryophyte Group, 1999).  Because in the field it is difficult to separate bryophyte
chlorophyll from that of periphyton, it is computed so that  the two can be combined and related
to field values:
                          MossChlor = l<(BryoConv • BiomassBtyo)
                                                                 (89)
where:
      MossChlor  =   bryophytic chlorophyll a (mg/m );
                                       95

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4

                                                                                    9     9
       BryoConv   =   conversion from bryophyte AFDW to chlorophyll a (8.9 mg/m : g/m );
                    =   biomass of given bryophyte (AFDW in g/m2).
Currents and wave agitation can both stimulate and retard macrophyte growth.  These effects
will be modeled in a future version.  Similar to the effect on periphyton, water movement can
stimulate photosynthesis in macrophytes (Westlake, 1967); the same function could be used for
macrophytes as  for periphyton,  although with different parameter values. Jupp  and  Spence
(1977b) have shown that wave agitation can severely limit macrophytes; time-varying breakage
eventually will be modeled when wave action is simulated.

4.3 Animals
  Animals: Simplifying Assumptions
   • Ingestion is represented by a maximum consumption rate adjusted for conditions of food, temperature, sublethal toxicant
    effects, and habitat preferences
    Reproduction is implicit in the increase in biomass
    Macrophytes can provide refuge from predation
    AQUATOX is a food-web model including prey switching based on prey availability
    Specific dynamic action (the metabolic "cost" of digesting and assimilating prey) is represented as proportional to food
    assimilated
    Unless spawning dates are entered by the user, spawning occurs as a function of water temperature
    Zooplankton and fish will migrate vertically from an anoxic hypolimnion to the epilimnion
    Promotion from one size class of fish to the next is estimated as a fraction of total biomass growth
Zooplankton,  benthic invertebrates, benthic insects,  and fish are modeled, with  only slight
differences  in formulations, with  a generalized animal submodel  that  is  parameterized to
represent different groups:

           dBiomass       ,  _,             ^  .         „            „  ,
          	= Load + Consumption - Defecation - Respiration -Pishing
              dt
                    - Excretion - Mortality - Predation - GameteLoss ± DiffusionSeg
             - Washout + Washin ± Migration - Promotion + Recruit - Entrainment         (90)


              GrowthRate = Consumption - Defecation - Respiration - Excretion
where:
       dBiomass/dt  =     change in biomass of animal with respect to time (g/m3-d);
       Load         =     biomass  loading, usually from upstream, or calculated from user-
                           supplied fish stocking data (g/m3-d);
       Consumption =     consumption of food (g/m3-d), see (98);
       Defecation    =     defecation of unassimilated food (g/m3-d), see (97);
       Respiration   =     respiration (g/m3-d), see (100);
       Fishing       =     loss of organism due to fishing pressure (g/m3-d),  user input fraction
                           fished multiplied by the biomass.
       Excretion     =     excretion (g/m3-d), see (111);
       Mortality     =     nonpredatory mortality (g/m3-d), see (112);
       Predation     =     mortality from being preyed upon (g/m3-d), see (99);
                                         96

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 4
       GameteLoss   =
       Washout      =

       Washin       =
       Diffusionseg   =

       Migration     =
       Promotion    =
       Recruit       =
       Entrainment   =

       GrowthRate   =
loss of gametes during spawning (g/m -d), see (126);
loss due to being carried downstream by washout and drift (g/m3-d),
see (129) and (130);
loadings from linked upstream segments (g/m3-d), see (30);
gain or loss due to diffusive transport over the feedback link between
two segments, pelagic inverts, only (g/m3-d), see (32);
loss (or gain) due to vertical migration (g/m3-d), see (133);
promotion to next size class or emergence (g/m3-d),  see (136);
recruitment from previous size class (g/m3-d), see (128);
entrainment  and  downstream  transport by floodwaters  (g/m3-d)
(132).
estimated growth rate as a function of  derivative  terms, output  in
units of percentage per day when animal's "rates output" is turned
on.
The change in biomass (Figure 70) is a function of a number of processes (Figure 71) that are
subject to  environmental factors, including  biotic interactions.  Similar to the way algae are
treated, parameters for different  species of invertebrates and fish are loaded and  available for
editing by  means  of the entry screens.  Biomass of zoobenthos and fish is expressed as g/m2
instead of g/m3.

Figure 70.  Predicted changes in biomass in a stream	
       18.7

       17.0
       15.3
       13.6
       11.9
       10.2
        3.4
        0.0
                     Cahaba River AL (CONTROL)
                       Run on 03-8-08 10:13 AM
                                          Obs Corbicula (g/m2dry)
                                         • Caddisfly,Tric (g/m2 dry)
                                         - Corbicula (g/m2 dry)
                                          Mussel (g/m2 dry)
                                          Obs Snails (g/m2 dry)
                                          Mayfly (Baetis (g/m2 dry)
                                          Gastropod (g/m2 dry)
                                          Obs Mayfly (g/m2 dry)
                                          Obs Caddis fly (g/m2 dry)
          2/26/2000  8/26/2000  2/24/2001   8/25/2001  2/23/2002  8/24/2002
                                         97

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                        CHAPTER 4
Figure 71.  Predicted Process Rates for the Invasive Clam Corbicula, Expressed as Percent of Biomass;
Yellow Spikes are Entrainment During Storm Events; Consumption Depends on Sloughing Periphyton..
       30.0
                 Cahaba River AL (CONTROL)
                   Run on 03-8-08 10:13 AM
                     12/5/2000
                                   12/5/2001
                                                 12/5/2002
                                                       - Corbicula Load (Percent)
                                                       • Corbicula Consumption (Percent)
                                                       • Corbicula Defecation (Percent)
                                                       • Corbicula Respiration (Percent)
                                                       Corbicula Excretion (Percent)
                                                       Corbicula Scour_Entrain (Percent)
                                                       • Corbicula Predation (Percent)
                                                       Corbicula Mortality (Percent)
                                                       Corbicula GameteLoss (Percent)
Consumption, Defecation, Predation, and Fishing

Several formulations have been  used in various models to represent consumption of prey,
reflecting the fact that there are different modes of feeding and that experimental evidence can be
fit by any one of several equations (Mullin  et al., 1975; Scavia,  1979; Straskraba and Gnauck,
1985).

Ingestion is represented in AQUATOX by a maximum consumption rate, adjusted for ambient
food, temperature, oxygen, sediment, and salinity conditions, and reduced for sublethal toxicant
effects and limitations due to habitat preferences of a given predator:
       Ingestion preypred = CMaxpred • SatFeeding • TCorr pred • FoodDilution
  • HabitatLimit • ToxReduction • HarmSS • SaltEffect • O2EffectFrac • Biomasspred
                                                                                       (91)
where:
       Ingestion prey, pred
       Biomass
       CMax
       SatFeeding
       TCorr
       FoodDilution
                   ingestion of given prey by given predator (g/m -d);
                   concentration of organism (g/m3-d);
                   maximum feeding rate for predator (g/g-d);
                   saturation-feeding kinetic factor, see (93);
                   reduction factor for suboptimal temperature (unitless), see Figure 59;
                   factor to account for  dilution  of  available  food  by suspended
                   sediment (unitless), see (120);
ToxReduction    = reduction due to effects of toxicant (see (424), unitless); and
                                         98

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       HarmSS         = reduction due to suspended sediment effects (see (116), unitless);
       SaltEffect        = effect of salinity on ingestion rate (unitless), see (440);
       O2EffectFrac    = effect of reduced oxygen on ingestion (unitless), see (205); and
       HabitatLimit     = in streams, habitat limitation based on predator habitat preferences
                         (unitless), see (13).

The maximum consumption rate is sensitive to body size, so an alternative to specifying CMax
for fish is to compute  it using an allometric equation and  parameters  from  the Wisconsin
Bioenergetics Model (Hewett and Johnson, 1992; Hanson et al., 1997):

                               CMax = CA • MeanWeighf8                           (92)
where:
       CA          =    maximum consumption for a 1-g fish at optimal temperature (g/g-d);
       MeanWeight  =    mean weight for a given fish species (g);
       CB          =    slope of the allometric function for a given fish species.

Many  animals adjust their search  or filtration in accordance with the  concentration of prey;
therefore,  a saturation-kinetic term is used (Park et al., 1974, 1980; Scavia and Park, 1976):

                                       Preference     , • Food
                 SatFeeding =	^^	             (93)
                              Zprey( Preference prey pred • Food) + FHalfSat pred

where:
       Preference     =    preference of predator for prey (unitless);
       Food         =    available biomass  of given prey (g/m3);
       FHalfSat      =    half-saturation constant for feeding by a predator (g/m3).

The food actually available to a predator may be reduced in two ways:

                         Food = (Biomassprey - BMinpred) • Refuge                     (94)
where:
       BMin        =      minimum prey biomass needed to begin feeding (g/m3); and
       Refuge       =      reduction factor for prey hiding in macrophytes (unitless).

Search or filtration may virtually cease below a minimum prey biomass (BMin)  to conserve
energy (Figure 72), so that a minimum food level is incorporated (Parsons et al., 1969;  Steele,
1974; Park et al., 1974; Scavia and Park, 1976; Scavia et al., 1976; Steele and Mullin, 1977).
However,  some filter feeders such as cladocerans (for example, Daphnid) must constantly filter
because the filtratory appendages also serve for respiration; therefore, in these animals there is
no minimum feeding level and BMin is set to 0.
                                       99

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
                         Figure 72. Saturation-kinetic consumption
                                  BASS CONSUMPTION
                                 BASS BIOMASS = 1 g/cu m
                           0.03 ^
                         : 0.025
                              0  2.65  5.3 7.95 10.613.2515.918.55
                                     PREY BIOMASS (g/cu m)
Macrophytes can provide refuge from predation; this is represented by a factor related to the
macrophyte biomass that is original with AQUATOX (Figure 73):
                                          BiomassMa.
where:
      Hal/Sat
      BlOmaSSMacro  =
 Refuge = 1 - -
           BiomassMacm + Hal/Sat

half-saturation constant (20 g/m3), and
biomass of macrophyte (g/m3).

  Figure 73. Refuge from predation
                                                                                  (95)
                                  100       200       300
                                  MACROPHYTE BIOMASS
                                      400
AQUATOX is a food-web model with multiple potential food sources.  Passive size-selective
filtering (Mullin,  1963;  Lam and Frost,  1976)  and active raptorial  selection  (Burns,  1969;
Berman and Richman, 1974; Bogdan and McNaught, 1975; Brandl and Fernando, 1975) occur
among aquatic organisms.  Relative preferences are  represented in AQUATOX by a matrix of
preference parameters first  proposed  by O'Neill (1969) and used in several aquatic models
                                      100

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4


(Bloomfield et al.,  1973; Park et al.,  1974;  Canale et  al., 1976;  Scavia  et al., 1976).  Higher
values indicate increased preference by a given predator for a particular prey compared to the
preferences for all possible prey. In other words, the availability of the prey is weighted by the
preference factor.

The  preference factors  are normalized so that if a potential food source is not modeled or is
below the BMin value, the other preference factors  are modified accordingly,  representing
adaptive preferences:

                                                 Pref
                               n  /•                 J prey,pred                           snf\
                               Prefere"ce *">>•*"* =                                      (96)
where:
       Preference prey,pred   =     normalized  preference  of given  predator  for given  prey
                                (unitless);
       Pref preyt pred         =     initial  preference  value from  the animal  parameter screen
                                (unitless); and
       SumPref           =     sum of preference values for all food sources that are present
                                above  the minimum  biomass  level  for feeding  during a
                                particular time step (unitless).

Similarly, different prey types have different potentials  for assimilation by different predators.
The fraction  of ingested prey that  is egested  as  feces or discarded (and which is treated as a
source of detritus by the model, see (153) and (154)), is  indicated by a matrix of egestion
coefficients with the same structure as the preference matrix, so that defecation is computed as
(Parketal., 1974):

       Defecation pred = Zprey ( EgestCoeff prey pred • Ingestion prey pred + IncrEgest • Ingest NoTm )   (97)

where:
       Defecation pred        =      total defecation for given predator (g/m3-d);
       Inge stion prey, pred      =      ingestion of given prey by given predator (g/m3-d) (91);
       EgestCoeffpK^ pred     =      fraction of ingested  prey that is egested (unitless); and
       IncrEgest            =      increased egestion due to toxicant (see Eq. (425), unitless);
       IngestNoTox           =      ingestion excluding toxic effects, calculated as  Ingestion
                                   divided by ToxReduction (see Eq. (424), g/m3-d).

Consumption of prey  for a  predator is also  considered predation or grazing  for  the prey.
Therefore,  AQUATOX represents consumption as a source term for the predator and as a loss
term for the prey:

                          Consumption pred = ^prey ( Ingestion prey pred)                      (98)
                           Predation prey = Zpred (Inge stion prey pred)                       (99)
where
       Consumption prsd      =      total consumption rate by predator (g/m3-d); and
       Predation prey         =      total predation on given  prey (g/m3-d).


                                        101

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Fishing pressure is represented simply as a fraction of biomass removed each day.  A potential
future model enhancement could allow for temporally variable fishing pressures to better reflect
harvesting seasons.  Fish may also be stocked within a modeled system by entering time series in
grams per day or grams per meter squared per day.

Respiration

Respiration can be considered as having three components (Cui and Xie, 2000), subject to the
effects of salinity:

         Respiration pred = (StdResppred + ActiveResp pred + SpecDynAction pred) • SaltEffect   (100)
where:
       Respirationpred     =    respiratory loss of given predator (g/m3-d);
       StdResppred        =    basal  respiratory  loss  modified by temperature (g/m3-d);  see
                              (101);
       ActiveResppred     =    respiratory loss associated with swimming (g/m3-d), see (104);
       SpecDynActionpred  =    metabolic cost of processing food (g/m3-d), see (110); and
       SaltEffect         =    effect of salinity on respiration (unitless), see (440).

Standard respiration is a rate at resting in which the  organism is expending energy without
consumption.  Active respiration is modeled only in fish  and  only when allometric (weight-
dependent) equations are used, so  standard  respiration  can  be  considered as a  composite
"routine" respiration for invertebrates and in the simpler implementation for fish.  The so-called
specific dynamic action is the metabolic cost of digesting and assimilating prey.  AQUATOX
simulates standard respiration as a basal  rate modified by a temperature dependence and, in fish,
a density dependence (see Kitchell et al., 1974):

               StdResppred = BasalResp pred • TCorrpred • Biomass pred • DemityDep          (101)
where:
       BasalResp^d  =     basal  respiration  rate at optimal  temperature  for  given predator
                           (g/g-d); parameter input by user as "Respiration Rate" or computed
                           as a function of the weight of the animal (see below);
       TCorrpred      =     Stroganov temperature function (unitless), see Figure 59;
       Biomasspred    =     concentration of predator (g/m3); and
       DensityDep    =     density-dependent respiration  factor used in computing standard
                           respiration, applicable only to fish (unitless). See (109)

As an alternative formulation, respiration in fish can be modeled as a function of the weight of
the fish using an allometric equation (Hewett and Johnson, 1992;  Hanson et al., 1997):

      StdResppred = BasalResppred • MeanWeight predRBfred. TFnpred • Biomass pred • DensityDep (102)

where:
       MeanWeightpred      =      mean weight for a given fish (g);
                                       102

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4


       RBpred               =      slope of the allometric function for a given fish;
       TFnprsd              =      temperature function (unitless).

The allometric functions are based on the well known Wisconsin Bioenergetics Model and, for
convenience, use the published parameter values  for that model (Hewett and Johnson, 1992;
Hanson et al.,  1997).  However, the basal respiration rate in that model is expressed  as  g of
oxygen per  g organic matter of fish per day, and this has  to be converted to  organic matter
respired:

                                BasalResp pred = RApred • 1. 5                          (103)
where:
       RApred      =      basal respiration rate, characterized as the intercept of the allometric
                         mass function in the  Wisconsin Bioenergetics Model documentation
                         (g O2/g organic matter -d);
       1.5         =      conversion factor (g organic matter/g ©2).

Swimming activity may be large and variable (Hanson et al., 1997) and is subject to calibration
for a particular site, considering currents and other factors:

                         ActiveResp pred = Activity pred • Biomasspred                     (104)
where:
       Activity pred   =      activity factor (g/g-d).

Activity  can be a complex function of temperature.   The Wisconsin  Bioenergetics Model
(Hewett and Johnson, 1992; Hanson et al., 1997) provides two alternatives. Equation Set 1  uses
an exponential temperature function:
                                         = e(RQ'Temp)                                (105)
where:
       RQ   =     the Qio or rate of change per lOdeg. C for respiration (1/deg. C);
       Temp =     ambient temperature (deg. C).

This is coupled with a complex  function for swimming speed  as  an  allometric function of
temperature (Hewett and Johnson,  1992; Hanson et al., 1997):
                                  Activity pred =

                     lfTemp>RTL Then  Vel = RKl • MeanWeightRK4                (106)
                       Else Vel = ACT • MeanWeightRK4 • e(BACT ' Temp}

where:
       RTO     =   coefficient for swimming speed dependence on metabolism (s/cm);
       RTL     =   temperature below which swimming activity is an exponential function of
                    temperature (deg. C);
       Vel      =   swimming velocity (cm/s);
                                       103

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       RK1     =   intercept for swimming speed above the threshold temperature (cm/s);
       RK4     =   weight-dependent coefficient for swimming speed;
       ACT     =   intercept for swimming speed for a 1 g fish at deg. C (cm/s); and
       BACT    =   coefficient for swimming at low temperatures (1/deg. C),

Equation Set 2 uses the Stroganov function used elsewhere in AQUATOX:

                                      TFn = TCorr                                 (107)
and activity is a constant:

                                     Activity = ACT                                (108)
where:
       TCorr  =     reduction factor for suboptimal temperature (unitless), see (59);
       ACT   =     activity factor, which is not the same as ACT'm Equation Set 1 (g/g-d).
Respiration in fish increases with crowding due to competition for spawning sites, interference in
feeding,  and other factors.   This  adverse intraspecific  interaction  helps  to  constrain the
population to the carrying capacity; as the biomass approaches the carrying capacity for a given
species the respiration is increased proportionately (Kitchell et al., 1974):

                           rx   •  ^     ,  IncrResp • Biomass                      ^««^
                           DensityDep = 1 +	                      (109)
                                             T^/""^   / r7~\ if
                                             KCap I ZMean
where:
      IncrResp     =      increase in respiration at carrying capacity (0.5);
      KCap        =      carrying capacity (g/m2);
      ZMean       =      mean depth from site underyling data (m).
With the IncrResp value of 0.5, respiration is increased by 50% at carrying capacity (Kitchell et
al., 1974), as shown in  Figure 74.  This density-dependence is used only for fish, and not for
invertebrates.
                                       104

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                      CHAPTER 4
               Figure 74. Density-dependent factor for increase in respiration as fish
                 biomass approaches the carrying capacity (10.0 in this example).
           0)
              1.6
              1C
              -!••->
           C  1.4
           Q
           >•  1 2
           ^  .L.Z
           'i/»
           §  1.1
           Q
                                   4           6

                                  Biomass (g/m3)
                                                                            10
As a simplification, specific dynamic action is represented as proportional to food assimilated
(Hewett and Johnson, 1992; see also Kitchell et al., 1974; Park et al., 1974):
              SpecDynAction  , = KResp  , • (Consumption  , - Defecation   )
                                                                             (110)
where:
       KResppred
                =   proportion of assimilated energy lost to specific dynamic action
                    (unitless); parameter input by user as "Specific Dynamic Action;"
Consumptionpred =   ingestion (g/m3-d) see (98); and
Defecationpred   =   egestion of unassimilated food (g/m3-d), see (97).
Excretion

As respiration occurs, biomass is lost and nitrogen and phosphorus are excreted directly to the
water (Home and Goldman 1994); see  (169) and (183).  Ganf and Blazka (1974) have reported
that this process is important to  the dynamics  of the Lake George, Uganda, ecosystem.  Their
data were converted by Scavia  and  Park (1976)  to  obtain a proportionality constant relating
excretion to respiration:
where:
       Excretionpred
       KExcrpred
       Respirationpred
                         Excretionpred = KExcrpred • Respiration
                                                            pred
                                                                             (HI)
                           excretion rate (g/m -d);
                           proportionality constant for excretion:respiration (unitless);
                           respiration rate (g/m3-d), see (100).
                                       105

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


Excretion is approximately 17 percent of respiration,  which is not an important biomass loss
term for animals, but it is important in nutrient recycling.   All biomass lost  due to  animal
excretion is assumed to convert to dissolved labile detritus, see (151).

Nonpredatory Mortality

Nonpredatory  mortality is a  result of both  environmental  conditions  and the  toxicity  of
pollutants:

                Mortality pred = Dpred • Biomass pred + Poisoned pred +-Mort'Ammoma
                                                                                   (112)

where:
       MortalityPred   =     nonpredatory mortality (g/m3-d);
       Dpred          =     environmental mortality  rate; the maximum value of (113) and
                           (114), is used (1/d);
                     =     biomass of given animal (g/m3);
       Poisoned      =     mortality due to toxic effects (g/m3-d), see (417);
       MortAmmonia    =     ammonia mortality, (g/m3-d), see (179);
       Mortiow02     =     low oxygen mortality, (g/m3-d), see (203); and
       MortsedEffects   =     mortality from suspended sediments, (g/m3-d), see (115)
       Mort salinity     =     mortality from salinity , (g/m3-d), see (112)

Under normal conditions a baseline mortality rate is used:

                                    D pred = KMortpred                               (H3)
where:
       KMortpred    =      normal nonpredatory mortality rate (1/d).

An exponential function is used for temperatures above the maximum (Figure 75):

                                               Temperature - TMaxpred
                            Dpred = KMortpred +	~	                        (114)

where:
       Temperature  =      ambient water temperature (°C); and
                           maximum temperature tolerated (°C).
                                       106

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
                       Figure 75. Mortality as a function of temperature

i
fY
LU
CL
Q
LU
_l
*:
o
i-
g
(r
LL.

1
OP
OR
OA
A
09

-
MORTALITY OF BASS
,


/
TMax I
3 6 9121518212427303336394:
TEMPERATURE






>
The lower lethal temperature is often 0°C (Leidy and Jenkins, 1976), so it is ignored at this time.

Suspended Sediment Effects
The approach  used  to quantify  lethal and
sublethal  effects of  suspended  sediments  is
based  on  logarithmic  models  described by
Newcombe (for example, Newcombe 2003).
                    Summary of Sediment Effects:
                      Mortality
                      Reduction in feeding
                      Dilution of food by sediment particles
                      Stimulation of invertebrate drift
                      Loss  of  spawning  and  protective  habitat  in
                      interstices
They take the form of:

               LethalSS = SlopeSS • \n(SS) + InterceptSS + SlopeTime • \n(TExp)
                                                        (115)
where:
       LethalSS

       SlopeSS
       SS
       InterceptSS
       SlopeTime
       TExp
cumulative fraction killed by given exposure to a given suspended
sediment concentration (fraction/d)
slope for sediment response (unitless)
suspended inorganic sediment concentration (mg/L)
intercept for suspended sediment response (unitless)
slope for duration of exposure (unitless)
duration of exposure (d)
Unfortunately, there is a dearth of quantitative data on response to sediments.  Therefore, the
responses are grouped according  to sensitivity, and parameters for surrogate species are used.
The user can specify different parameter values; the values given below are provided as defaults.

For sublethal effects, avoidance behavior is noted at SS of about 100 mg/L (Doisy and Rabeni
2004); however, this could only be used as a cue for migration in the model and has been ignored
                                       107

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
at this time.

Reduction in feeding occurs in game fish due to visual impairment (Crowe and Hay 2004).  SS
of 25 mg/L seems to be threshold for response (Rowe et al. 2003). The general equation (115) is
used to represent a decrease in food due to turbidity, but without the exposure factor because the
response is instantaneous:
                        HarmSS = SlopeSS • \n(SS) + InterceptSS
                                                       (116)
where:
       HarmSS
       SlopeSS
       SS
       InterceptSS   =
reduction factor for impairment of visual predation (unitless)
slope for suspended sediment response (-0.36, unitless)
suspended inorganic  sediment concentration (mg/L).  If TSS is
modeled see (244) otherwise, the  sum of inorganic  sediments in
the water column (e.g. Sand+Silt+Clay);
intercept for suspended sediment reponse (2.11, unitless)
The equation is parameterized using data for coho salmon with 1-hr exposure (Berry et al. 2003).
It was verified with numerous other qualitative observations for salmon, Arctic grayling, and
trout (Berry et al. 2003).  This equation is used for all visual-feeding fish, especially game fish.
The user has the option of turning on this factor

             Figure 76. Reduction in feeding by coho salmon (Oncorhynchus kisutch)
                   due to suspended sediments. Data from (Berry et al. 2003).
                            Reduced Feeding in Salmon
                       0
      100
   200
SS (mg/L)
300
400
For modeling lethal effects, mortality can occur in fish over a range of suspended sediments.
Because of the lack of suitable quantitative data,  these responses  are divided into sensitivity
categories specific to this model and differing from Clarke and Wilber (2000) with parameters
                                      108

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


for surrogate species that can be considered representative for groups of organisms.  The factor
also can be turned off for those organisms that are completely insensitive.

Tolerant

This category represents those species having a 24-hr LCio > 5000 mg/L SS. Generally, these
are benthic species exposed to the flocculent zone and bottom sediments. The general equation
(115) is parameterized to accommodate the 24-hr lethality observations and is extended to other
times of exposure by fitting to observed 48-hr lethal responses:

                       LethalSS = l.62-\n(SS) - 14.2 + 3.5-\n(TExp)                  (117)

where:
       LethalSS     =     cumulative fraction killed by given exposure to a given suspended
                          sediment concentration (fraction/d)
       TExp        =     time of exposure to given level of suspended sediment (d)
       SS           =     minimum  suspended  inorganic   sediment  concentration  over
                          exposure time (mg/L).  If TSS is modeled see (244) otherwise, the
                          sum  of  inorganic  sediments  in  the  water  column  (e.g.
                          Sand+Silt+Clay).


The parameters are based on the benthic  estuarine fish spot (Leiostomus xanthurus\ using data
compiled in Berry et al. (2003).

Due to lack  of data beyond 48 hours, this equation is applied using one- and two-day exposure
times only. The maximum effect is chosen from these two equation results.
                                      109

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
            Figure 77. Lethality of suspended sediments to spot (Leiostomus xanthurus),
                a tolerant species, based on data compilation of Berry et al. (2003).
Fraction Killed
Spot Mortality
•i nn
Osn
OGf\
.DU
OAf\
Oon
.zu
r\ r\r\
s
/
/
/
1 /
I

VJ.VJVJ III
0 2000 4000 6000 8000 10000 12000
SS (mg/L)

* SS 24-hr • SS 48-hr • SS 24-hr Est
A SS 48-hr Est 	 Log. (SS 24-hr) 	 Log. (SS 48-hr)
Sensitive

This category represents those species having 250 mg/L < 24-hr LCio <5000 mg/L SS.  Small
estuarine species seem to be highly sensitive to suspended sediment (Figure 82).  The general
parameters are based on a composite fit to  data for  bay  anchovy, menhaden, and Atlantic
silversides taken from a compilation by Berry et al. (2003). The equation is:
                       LethalSS = 0.34 • \n(SS) -1.85 + 0.1- \n(TExp)
       (118)
This equation is applied using one- and two-day exposure times along with effects from  one,
two, and three weeks exposure. The maximum effect is chosen from these multiple calculations.

Figure 78 illustrates the response curve for white perch. The equation exhibits good extension to
juvenile rainbow trout with a 28-d exposure to SS (Figure 79) and Chinook salmon with a 1.5-d
exposure (Figure 80).  In both  cases  the equation  is slightly  over-protective, but that is
considered appropriate.
                                       110

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                          CHAPTER 4
          Figure 78.  Lethality of suspended sediments to white perch (Morone americana),
               a sensitive species, based on data compilation of (Berry et al. 2003).
                             White Perch Mortality, 24-hr
                                            2000
                                          SS (mg/L)
                                 4000
        Figure 79. Lethality of suspended sediments to juvenile rainbow trout (Oncorhynchus
            mykiss) using parameters for sensitive species. Data from Berry et al. (2003).
                       Juvenile Rainbow Trout Mortality,  19-d,
                     1.00
                     0.80
                 1  0.60
                     0.00

                    -0.20
200     400      600      800     1000
                                          SS(mg/L)
                                  4  Obs  •  Est - Log. (Obs)
                                      Ill

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                          CHAPTER 4
          Figure 80. Lethality of suspended sediments to Chinook salmon (Oncorhynchus
         tshawytschd) using parameters for sensitive species. Data from Berry et al. (2003).
                           Chinook Salmon Mortality, 36-hr
                              10000    20000   30000   40000    50000
                                          SS (mg/L)
                                   Obs1.5d
Est 1.5 cl
•Trend
Highly Sensitive

This category represents those species having a 24-hr LCio < 250 mg/L SS.   Small estuarine
species seem to be highly sensitive to  suspended sediment (Figure 82).  The general parameters
are based on a composite fit to data for bay anchovy, menhaden, and Atlantic  silversides taken
from a compilation by (Berry et al. 2003).  The equation is:
                      LethalSS = 0.328 • \n(SS) - 1.375 + 0.1 • \n(TExp)
                                 (119)
This equation is applied using one and two day exposure times along with effects from one two
and three weeks exposure. The maximum effect is chosen from these multiple calculations.
                                      112

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
              Figure 81. Three-dimensional plot of equation for highly sensitive fish
Although not verified with observed data from longer exposure periods, the equation appears to
be robust; it yields reasonable predictions  of mortality for a range of SS concentrations and
exposure periods.
                                      113

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
             Figure 82. Response of bay anchovy to SS. Data from (Berry et al. 2003)
Composite of Highly Sensitive Spp., 24-hr
1 nnoA _,
Qno/
yu/o
sno/
OU/O
7fW
(U/o
Rno/'
^no/
OU/o
/i no/
4U/0
/^no/'
ono/'
•ino/
lU/o
no/

_^^* — ***
^ 	
^^^
./.
/
/
/ y = 0.3278Ln(x) - 1.3751
/ R2 = 0.7125
*/

U /o n i i i i
0 200 400 600 800 1000 1200
TSS
Sediment Effects on Filter Feeders

Sediments can clog filter-feeding apparatuses in invertebrates and some fish. A 25% reduction
in feeding in Daphnia occurs with  SS of 6 NTU (-22 mg/L) (Henley, 2000); rotifers are not
affected (Rowe et al. 2003).  Equation (116) can be parameterized to reflect the  Daphnia
response (SlopeSS = -0.46 and InterceptSS = 2.2, Figure 83).

             Figure 83. Reduction in feeding by Daphnia due to suspended sediments.
                  Points represent LC75 and supposed LC50, and LC5 values.
HarmSS Reduction
Reduced Feeding in Daphnia
<\
\
On
00
0-7
. 1
OC
Oc
OA
.**
Oo
.O
00
0<\
. \
\

\
\^
N.
Xy^
^S.
^•x.
^^^-V^^^
^>
0 20 40 60 80 100 12
SS (mg/L)
20
                                      114

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                       CHAPTER 4
Increased turbidity can inhibit feeding by mussels; 600 to 750 mg SS/L reduced clearance rates
in several mussel species (Henley et al. 2000). This can be used to parameterize Equation (116)
(SlopeSS=-OA7 and!nterceptSS=3.1, Figure 84).

            Figure 84. Reduction in clearance of sediment by freshwater mussels due to
          suspended sediments.  Points represent supposed LC95, LC50, and LC10 values.
                o
                '•8
                3
                •o
                V)
                V)
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
                           Reduced Clearance in Mussels
                                200
                        400
                     SS (mg/L)
600
800
Reduced pumping was observed at SS > 1000 mg/L in the Eastern oyster (Berry et al. 2003).
This too can be used to parameterize Equation (116).  (SlopeSS = -0.61 and InterceptSS = 4.72,
Figure 85).
                                      115

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                  CHAPTER 4
              Figure 85. Reduced pumping in Eastern oysters (Crassostrea virginicd).
                     Points represent supposed LC90, LC50, and LC5 values.
HarmSS Reduction
Reduced Pumping in Oysters
On
.y
00
0-7
. 1
OC
Oc
.0
OA -
00
.O
On
0<\
. \
n
\
\
\
\
\
X
\
\v
^x^
_5

0 500 1000 1500 2000 2500
SS (mg/L)
A related factor, which is treated separately in the model, is the degree to which there is dilution
of food by inorganic particles, offset by selective sorting of particles and feeding (Henley, 2000).
Mytilus edulis, the blue mussel, and Crassostrea virginica,  the Eastern oyster,  actively sort
particles; their food intake should not be affected by SS until very high levels that clog the filter
feeding mechanism  are reached.  In contrast,  there is limited selective feeding among many
clear-water  clams, including the surf clam Spisula solidissima,  the Iceland scallop Chlamys
islandica, and  probably many of  the endangered freshwater mussels  (Henley,  2000).   The
dilution of available food for both filter feeders and grazers decreases as a proportionate function
of sediment corrected for the degree to which there is selective feeding (Figure 86):
                  FoodDilution   =
                                                   Food
                                    Food + Sed • Proportion • (1 - Sorting)
                                                         (120)
where:
       FoodDilution  =

       Food         =

       Sed

       Sorting       =
       Proportion    =
factor  to  account for dilution of  available food  by suspended
sediment (unitless)
preferred food for filter feeders (mg/L) and for grazers (g/m2) (see
(94))
suspended sediment  for filter  feeders  (mg/L)  and  deposited
sediment for benthic grazers (g/m2)
degree to which there is selective feeding (unitless)
proportionality constant,  set to 0.01 for snails and grazers and set
to 1.0 for all other organisms,  (unitless)
                                        116

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
To account for the fact that snails and grazers feed on periphyton above the depositional surface,
a proportionality constant is utilized for those organisms.

The intermediate variable  Sed depends on the computation of suspended sediment for filter
feeders and the computation of deposited sediment for benthic grazers. If the optional sediment
transport submodel (Section 6.1) is used then:
             Sed  =  (Conc(Silt) + Conc(Clay))       or
             Sed  =  (Deposit(Silt) + Deposit (Clay)) • Vol I Surf Area • 1000 • 1.0
where:
       Sed
       Concsed
       1000
       Deposited
       Volume
       SurfArea
       1.0
                                                       (121)
suspended sediment for filter feeders  (taxa = 'Susp. Feeder'  or
'Clam' in units of mg/L  or g/m3) and deposited sediment for
benthic grazers (taxa = 'Sed Feeder' or 'Snail' or 'Grazer' in units
of g/m2);
concentration of suspended silt or clay (mg/L) (224);
conversion factor for kg to g;
amount of sediment deposited (kg/m3 day) (230);
water volume, (m3);
surface area, (m2); and
days' accumulation of sediment (day)
If the sediment transport submodel  is not used and TSS is used as  a driving variable then
suspended sediment is computed for filter feeders. Additionally, when TSS is used as a driving
variable, deposited sediment (Sed) is calculated using the relationship shown in Figure 89.
                        Sed
                           Suspended
         =  InorgSed
                        SedDeposlted = 0.270\n(InorgSed60day)-0.072
                                                                                 (122)
where:
       Sed
       InorgSed     =

       InorgSed6oday  =
food dilution equation input (120), (mg/L or g/m2);
suspended inorganic  sediment computed from TSS  (mg/L) (see
(244));
60 day average of suspended inorganic sediment computed from
TSS (mg/L) (see (244))
                                      117

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                          CHAPTER 4
        Figure 86. The FoodDilution factor as a function of TSS with Food kept constant at 10
                           mg/L and with Sorting set to 0 and 0.5.
                 1.00
                 0.00
                               Sediment Dilution Factor
                            10
                                     •Sorting = 0'
•Sorting = 0.5
Continued high levels of SS can cause mortality in oysters as shown in Figure 87. However, this
can  be  interpreted  as  the  natural  consequence  of  reduced filtration  as predicted  by
parameterization of (115).  Therefore, oyster mortality due to SS is not simulated separately.

               Figure 87. Response of oysters to SS. Data from (Berry et al. 2003).
Fraction Killed
Eastern Oyster Mortality, 12-d Exposure
•i nn -*>
1 .UU
Onn
.yu
Oon
.OU
07PI
. /U
Oen
.OU
Ocn
.OU
OA n
.•HI
Oon
.ou
Oon
.<£U
O-i n
. 1 U
n nn
V
^^"^
^r
y^
/*
/
/'
/
/
T

\j.\j\j i i
0 1000 2000 3000 4000
SS (mg/L)
                                       118

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


Sediment Effects on Grazers

Sediment reduces preference of New Zealand mud snails and mayflies for periphyton (Suren
2005), which is ignored by the model.  More important, the food quality of periphyton declines
linearly with increasing fine sediment content (Broekhuizen et al. 2001). This is represented as
food dilution by (120).

Riffle areas are degraded or lost by deposition of fine sediment, including sand (Crowe and Hay
2004).  A  12-17% increase in fines in riffles areas resulted in 27-55%  decrease in mayfly
abundance; this did not affect chironomids and simulids, and riffle beetles actually increased
(Crowe and Hay 2004).  Drift rates doubled from 2.3%/d to 5.2%/d with a  16% increase in fine
interstitial sediments; chironomids and caddisflies were affected (Suren and Jowett 2001). This is
represented by a function in which the deposition rate  is compared to a trigger value beyond
which there is accelerated drift:
                               Drift =  Dislodge • Biomass                          (123)

where:
       Drift        =       loss of zoobenthos due to downstream drift (g/m3-d); and
       Dislodge    =       fraction of biomass subject to drift per day (unitless).

Nocturnal drift is a natural phenomenon:

                             Dislodge = AvgDrift • AccelDrift                        (124)

where:
       AvgDrift     =      fraction of biomass subject to normal drift per day (unitless).

                              AccelDrift  =  ev»p°>*-™**°r)                         (125)

where:
       AccelDrift    =      factor for increasing invertebrate drift due to sediment deposition
                           (unitless);
       Deposit      =      total rate of inorganic sediment deposition (kg/m day), (125b);
       Trigger      =      deposition rate at which drift is accelerated (kg/m2 day).
                                       119

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 4
                     Figure 88. AccelDrift as a function of depth-corrected
                           sediment deposition with Trigger = 0.2.
                        Drift as a Function of Sedimentation
                    0.000
                               0.2
            0.4     0.6     0.8
             Deposit (kg/m2 d)
1.2
The model computes daily sediment deposition rate based on suspended sediment using the
following relationship:
where:
      Deposit
      SS
                               Deposit  =  2.70-\n(SS)
                                                     (125b)
total rate of inorganic sediment deposition (kg/m  day), (125b);
suspended inorganic sediment concentration (mg/L).  If TSS is
modeled see (244) otherwise,  the sum of inorganic  sediments in
the water column (i.e. Sand+Silt+Clay);
                                      120

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
                       Figure 89. Relationship of one-day sedimentation
                     to average TSS; data from Larkin and Slaney (1996)..
Deposited Sediment (kg/m2 d)

1 on
1 nn
n an



n nn

y=0.2705ln(x)- 0.0723 1'1B
R2~0971') — "^^^^^^"^
^ " * 0.92
S^"^
If 0.56
/ 0.55
/
4 0.06
0 20 40 60 80
Mean Suspended Sediment (mg/L)
Ephemeroptera, Pteroptera, and Trichoptera (mayfly, stonefly, and caddisfly or EPT) diversity
declines when the fines (<0.25 mm) exceed 0.8% (Kaller and Hartman 2004). This change in
composition should result from proper parameterization of Equation (125).

Interstitial Sediments

Salmonid  reproduction is adversely affected  by deposition of fines,  with  27%  fines being a
threshold  (Nelson and Platts, unpublished report, cited  by Rowe et  al. 2003). "Multiple age
classes of both salmonids and sculpins were uncommon where average instream surface  fines
were greater than 30%, and nearly absent above 40%" (Rowe et al. 2003). Both the eggs and the
yolk-fry or alevins are sensitive to sedimentation of fines, including sand.  Sedimentation in
spawning  gravels  can be related to average suspended sediment  (TSS)  concentrations (Larkin
and Slaney 1996). The relationship is logarithmic for average TSS over a 60-day period (Figure
90).
                                      121

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
                 Figure 90. Relationship of 60-day sedimentation to average TSS;
                            data from (Larkin and Slaney 1996).

:5"
o
ID
ro 4UU.UU
1
— onn nn
c
0)
E
"o *)nn nn
LO
•c
01
•S 1 nn nn
o
Q.
&
n nn -
y= 108. lln(x)- 28.898 *71.86
R2 = 0.9712 /

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
             Figure 91. Relationship of 60-day percent embeddedness to average TSS;
                             data from (Larkin and Slaney 1996).


V)
V)
01

01
•c
~B
01
.a
1
se






30%

25%


ino/


15%

__^ 	
^J^r—
.f^

/ ^ 26%




f
f
A 1fi% y=0.0777ln(x)- 0.0208
7 16o/ R2 = 0.9712

5/o -j-
0% 2%
0 50 100 150 200
Mean Suspended Sediment (mg/L)
Gamete Loss and Recruitment

Eggs  and sperm can be a significant fraction of adult biomass;  in bluegills these can be  13
percent and 5  percent, respectively (Toetz,  1967),  giving  an average of 9  percent if the
proportion of sexes is equal.  Because only a small fraction of these gametes results in viable
young when shed at the time of spawning, the remaining fraction is lost to detritus in the model.

There are two options for determining the date or dates on which spawning will take  place.  A
user can specify up to three dates on which spawning will take place.  Alternatively, one may use
a construct that was modified from a formulation by Kitchell et al. (1974).  As a simplification,
rather than requiring species-specific spawning temperatures,  it assumes that spawning occurs
when the temperature first enters the range from six tenths of the optimum temperature to 1° less
than the optimal temperature. This is based on a comparison of the optimal temperatures with the
species-specific spawning temperatures reported by Kitchell  et al. (1974).  Depending on the
range of temperatures, this simplifying assumption usually will result in one or two spawnings
per year in  a  temperate  ecosystem  when a simple sinusoidal temperature function is used.
However, the user also can specify a maximum number of spawnings.

The loss rate for gametes is  estimated for  both fish and invertebrates  as a function of user-
specified intrinsic gamete mortality and increased mortality due to effects of organic toxicants,
low oxygen, and salinity on adults (FracAdults being a function of carrying capacity).
                                       123

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 4
                      If (0.6 • TOpt) < Temperature < (TOpt -1.0) then
        GameteLoss = (GMort + IncrMort + O2EffectFrac ) • FracAdults • PctGamete
                                  • SaltMort • Biomass
                                                           (126)
where:
       Temperature    =
       TOpt
       GameteLoss    =
       GMort
       IncrMort       =

       O2EffectFrac   =
       PctGamete
       FracAdults
       SaltMort
       Biomass
            else GameteLoss = 0

    ambient water temperature (°C);
    optimum temperature (°C);
    loss rate for gametes (g/m3-d);
    gamete mortality (1/d);
    increased gamete and embryo mortality due to toxicant (see (426),
    1/d);
    calculated fraction of gametes lost at a given oxygen concentration
    and exposure time (1/d), see (205);
=   fraction of adult predator biomass that is in gametes (unitless); and
=   fraction of biomass that is adult (unitless);
=   effect of salinity on gamete loss rate (unitless), see (440); and
=   biomass of predator (g/m3).
As the biomass of a population reaches its carrying capacity, reproduction is usually reduced due
to stress; this results in a population that is primarily adults. Therefore, the proportion of adults
and the fraction of biomass in gametes are assumed to be at a maximum when the biomass is at
the carrying capacity (Figure 92):

                      Figure 92. Correction for population-age structure
                                         BASS
                                 PctGamete = 0.09, GMort = 0.1
                            0.1  0.7 1.3 1.9 2.5  3.1  3.7 4.3 4.9 5.5
                                          BIOMASS
                                      124

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


                           7-   AJ  1+   , n  (  Capacity   ~]
                           FracAdults = 7.0-	—
                                           {KCap/ZMean)
                                                                                (127)
     if Biomass > KCap I ZMean then Capacity = 0 else Capacity = KCap I ZMean - Biomass
where:
       KCap        =     carrying capacity, the maximum sustainable biomass (g/m2);
       ZMean      =     mean depth from site underyling data (m).


Spawning in large fish  results in an increase in the biomass of small fish  if both small and large
size classes are of the same species. Gametes are lost from the large fish,  and the small fish gain
the viable gametes through recruitment:

            Recruit = (1- (GMort + IncrMort)) • FracAdults • PctGamete • Biomass        (128)

where:
       Recruit      =     biomass gained from successful spawning (g/m3-d).


Washout and Drift

Downstream transport  is an important loss term for invertebrates. Zooplankton are subject to
transport downstream similar to phytoplankton:
where:
                            Washout =   1SC arge • Biomass                        (129)
                                        Volume

       Washout      =     loss of zooplankton due to downstream transport (g/m3-d);
       Discharge     =     discharge (m3/d), see Table 3;
       Volume       =     volume of site (m3), see (2); and
       Biomass      =     biomass of invertebrate (g/m3).

Likewise,  zoobenthos exhibit drift,  which  is detachment  followed by washout,  and it is
represented by a construct that is original with AQUATOX:

                       WashoutZoohenthos = Drift = Dislodge • Biomass                  (130)
where:
       Drift      =     loss of zoobenthos due to downstream drift (g/m3-d); and
       Dislodge   =     fraction of biomass subject to drift per day (unitless), see (131)  and
                       (132)

Nocturnal drift is a natural phenomenon:

                                 Dislodge = AvgDrift                             (131)
                                      125

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
where:
      AvgDrift
fraction of biomass subject to normal drift per day (unitless).
Animals  also are subject to  entrainment and downstream transport in flood waters.  In fact,
annual variations in fish populations in streams are due largely to variations in flow, with almost
100% loss during large floods in Shenandoah National Park (NFS, 1997). A simple exponential
loss function was developed for AQUATOX:
where:
       Entrainment  =
       Biomass      =
       MaxRate      =
       Vel
       VelMax
       Gradual      =
                                 Vel-VelMax
Entrainment = Biomass • MaxRate • e Oradual

entrainment and downstream transport (g/m3-d);
biomass of given animal (g/m3);
maximum loss per day (1/d);
velocity of water (cm/s), (14);
velocity at which there is total loss of biomass (cm/s); and
slope of exponential, set to 25 (cm/s).
                                                                                 (132)
                Figure 93. Entrainment of animals as a function of stream velocity
                                 with VelMax of 400 cm/s
1 1

-a
a
.= n a -
S °
B
O
?? n j
2
^ 02


/
7
7
/
_^/

0 100 200 300 400 500
Velocity (cm/s)
Entrainment is not applied to pelagic invertebrates as these organisms already passively wash out
of a system during a flood event (129).

Vertical Migration

When presented with unfavorable conditions, most animals will attempt to migrate to an adjacent
area with more favorable conditions.  The current version of AQUATOX, following the example
of CLEANER (Park  et al., 1980), assumes that zooplankton and fish will exhibit avoidance
behavior  by migrating vertically from an anoxic hypolimnion to the  epilimnion.  AQUATOX
assumes that EC50growth is the best indicator of when the species has become so intolerant of the
oxygen climate that it is going to migrate. This also allows more tolerant species to spend more
time in the hypolimnion and less tolerant species to migrate earlier.  The assumption is that
anoxic conditions will persist until overturn.
                                      126

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 4
The construct calculates the absolute mass of the given group of organisms in the hypolimnion,
then divides  by the volume of the epilimnion  to  obtain the biomass being  added to the
epilimnion:
                               If VSeg = Hypo and Anoxic
                                                                                 (133)
                       Migration =
                                    HypVolume •  Biomass pred,
                                  pred, hypo
                                            Epi Volume
where:
       VSeg
       Hypo
       Anoxic
       Migration
       HypVolume
       Epi Volume
       Biomasspred:hypo
=   vertical segment;
=   hypolimnion;
=   boolean variable for anoxic conditions when O2 < EC50gmwth',
=   rate of mi grati on (g/m3 • d);
=   volume of hypolimnion (m3), see Figure 36;
=   volume of epilimnion (m3), see Figure 36; and
=   biomass of given predator in hypolimnion (g/m3).
In the estuarine model, fish will also migrate vertically based on salinity cues (see Section 10.5).
In the multi-segment version of AQUATOX, fish will vertically migrate to achieve equality on a
biomass basis if the system becomes well mixed (see Section 3.).

Migration Across Segments

To simulate seasonal migration patterns animals may be set up  to move from one segment to
another during a multi-segment model run.  Animals may migrate to or from a segment on any
date  of the  year to represent an appropriate seasonal  pattern; however, reaches must be linked
together with "feedback links" for migration to be enabled.  The user must specify the  date on
which  migration occurs,  the fraction of the state variable's concentration expected to migrate,
and the segment(s) involved.   The calculation of  state variable movement to and from each
segment must be normalized to the volume of water in the destination segment:
                     MigrationFromSeg = ConeSourceSeg • FracMoving
                                                           (134)
              Migration
                               Cone SourceSeg-Volume SourceSeg • FracMoving
                        ToSeg
                                            Volume
                                                           (135)
                                                   'Destination
where:
      Migration FromSeg
      Migration ToSeg
      L- OnC Segment
      Volume Sesment
      FracMoving
    loss of state variable in source segment (mg/LSourceSeg'd);
    gain of state variable in destination segment (mg/
    concentration of state variable in given segment (mg/L);
    volume of given segment (m3);
    user input fraction of animals migrating on given date (unitless);
                                      127

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Anadromous Migration Model

A new option in AQUATOX Release 3.1 is to model the migration of fish into and out of the
main study area in order to approximate anadromous migration behavior. Anadromous fish live
most of their adult life in saltwater, but they return to freshwater to spawn, and juveniles grow
for a few months to a few years before going to saltwater; during their time in freshwater they
may be exposed to and bioaccumulate organic  toxicants. Chinook salmon and Pacific lamprey
are  two species  in the animal database that can be used with this model.   The anadromous
migration component is a fairly simple model that holds off-site fish in what is assumed to be a
clean "holding tank."  No additional exposure  of the  fish to the toxicant is predicted to occur
while off-site, but growth dilution and depuration of toxicant is assumed to occur.

To get to this model, a size-class fish must be modeled and then an "Anadromous" button will
appear in the fish loading options screen.  Inputs for this model are shown below

   •   Day-of-year of migration (integer)
   •   Fraction of biomass migrating (fraction)
   •   Day-of-year of adult return (integer)
   •   Years spent off site (integer)
   •   Mortality fraction (fraction)

Based  on these  parameters and the weight of the juvenile and adult organisms, the biomass
returning to the freshwater study area may be calculated as follows.
                                                               MeanWeightA
                                                              ' MeanWeightJmemlt
Biomass LoadmgMult = Biomass Departmgjmemle (1 - MortalityFrac} —	——r^-  (135b)
The chemical concentration in these returning fish can also be estimated, given the depuration
rate for the chemical in the fish:

                                                        t d s MeanWeightJuvenile
          ^-OnC Loading, Adult = ^OnC Departing,Juvenile U ~~ DepUTatlOn)    ——
                                                             MeanWeightAdult

where:
      Biomass      =    predicted loading or migrating biomass (g/m3);
      MortalityFrac =    user-input assumed  fraction of juveniles that do not survive to return
                         (frac);
      MeanWeight  =    user-input mean weight of the juvenile or adult size-class organism;
      ConCorganism   =    concentration of chemical in departing or returning fish (|ig/kg ww);
      Depuration   =    user-input  depuration  rate for the  given chemical in  the  given
                         organism (I/day);

A spreadsheet version of this  sub-model  is  available in "Anadromous_Model.xlsx" and is
installed in the STUDIES directory when Release 3.1 is installed.
                                       128

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Promotion and Emergence

Although AQUATOX is an ecosystem model, promotion to the next size class is important in
representing the emergence of aquatic insects, and therefore loss of biomass from the system,
and in predicting bioaccumulation of hydrophobic organic compounds in larger fish.  The model
assumes that promotion is determined by the rate of growth.  Growth is considered to be the sum
of consumption and the loss terms other than mortality and migration; a fraction of the growth
goes into promotion to the next size class (cf Park et al., 1980):

Promotion = KPropred' (Consumption - Defecation - Respiration - Excretion - GameteLoss)  (136)
where:
      Promotion    =   rate of promotion to the next size class or insect emergence (137)
                        (g/m3-d);
      KPro         =   fraction of growth  that  goes  to promotion  or emergence (0.5,
                        unitless);
      Consumption  =   rate of consumption (g/m3-d), see (98);
      Defecation    =   rate of defecation (g/m3-d), see (97);
      Respiration   =   rate of respiration (g/m3-d), see (100);
      Excretion     =   rate of excretion (g/m3-d), see (111); and
      GameteLoss  =   loss rate for gametes (g/m3-d), see (126).

This is  a simplification  of a  complex response that  depends  on the mean weight  of  the
individuals.  However, simulation of mean weight would require modeling both biomass and
numbers of individuals (Park et al., 1979,  1980), and that is beyond the scope of this model at
present.  Promotion of multi-age fish is straightforward; each age class is promoted to the next
age class on the first spawning date each year. The oldest age class merely increments biomass
from the previous age class to any remaining biomass in the class.  Of course, any associated
toxicant is transferred to the next class as well.  Recruitment to the youngest age class is the
fraction of gametes that are not subject to mortality at spawning.  Note that the user specifies the
age at which spawning begins on the "multi-age fish" screen.

Insect emergence can be an important factor in  the dynamics of an aquatic ecosystem.  Often
there is synchrony in the emergence; in AQUATOX this is assumed to be  cued to temperature
with additional forcing as twice the promotion that would ordinarily  be computed,  and is
represented by:
              If Temperature > (0.8 • TOpt) and Temperature < (TOpt -1.0) then
                                                                                 (137)
                             Emergelnsect = 2 • Promotion
where:
      Emergelnsect        =      insect emergence (mg/L-d);
      Temperature        =      ambient water temperature (°C); and
      TOpt               =      optimum temperature (°C);
                                      129

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4


Because emergence is a function of the organism's growth rate, if the temperature passes through
the optimal temperature interval while the growth rate of the organism is zero or below zero,
emergence of insects does not occur.

4.4 Aquatic-Dependent Vertebrates

Herring gulls and other shorebirds were added to AQUATOX Release 3 as a bioaccumulative
endpoint — not as a dynamic variable but as a post-processed variable reflecting dietary exposure
to a contaminant.  In fact, the endpoint can be used to simulate bioaccumulation for any aquatic
feeding organism,  such as bald eagles, mink, and dolphins,  provided that the organism  feeds
exclusively on biotic compartments modeled within AQUATOX.   The user can specify a
biomagnification factor (BMP) and  the preferences  for various  food sources so that  alternate
exposures can be computed.  Dietary preferences are input as fraction of total food consumed by
the modeled species and are normalized to 100% when the model is run.

The concentration  of each chemical  is based on the chemical concentration in prey at a given
time-step.


                     PPBBirdTmicant = ±(PrefPrey • BMFTm • PPBPreyJm)                 (138)
                                     z=l

where:
                       =  estimated  concentration  of this toxicant in bird or other organism
                       =  biomagnification factor for this chemical in bird or other organism
                          (unitless);
       PPBpreyjox       =  concentration of this chemical in prey (ug/kg), see (310).

Uptake of toxicant is assumed to be instantaneous, but depuration of the chemical is governed by
the user-input clearance rate.  If the concentration of chemical is declining in shorebirds (due to
the concentrations of the chemical  declining in prey), the lowest the chemical concentration in
birds can fall to at any time is calculated as follows:

                      PPBBirdLowestJOX = PPBBird^ (1 - Clear, J AT                (139)

where:
       PPBBirdiowestjox =  lowest cone, of this toxicant in gulls or other organism  at this time-
                          step (ug/kg);
       PPBBirdTox,t-i    =  concentration of this toxicant in in the previous time-step (ug/kg);
       Clear Tox         =  clearance rate for the given toxicant, (I/day)

4.5 Steinhaus Similarity Index

Within the  differences graph portion of the  output interface, a user may select to write a set of
Steinhaus similarity indices in Microsoft Excel format.  The Steinhaus index  (Legendre and
Legendre  1998) measures the concordance in values (usually numbers  of individuals, but
                                       130

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4

biomass in this application) between two samples for each species. Typically it is computed from
monitoring data from  perturbed and  unperturbed, or reference,  sites.  When calculated by
AQUATOX it is a measure of the difference between the control and perturbed simulations.  A
Steinhaus index of 1.0 indicates that all species have identical biomass in both simulations (i.e.,
the perturbed  and control simulations);  an index of 0.0 indicates a  complete dissimilarity
between the two simulations.

The equation for the Steinhaus index is as follows:

                             "    /
                          2- 2_,mm(Biomass    t l , Biomass    t b
                     S=^	      "
                                         contml + Biomass ,. perturbed)
                            '•=1                                                     (140)

where:
       S               =  Steinhaus similarity index at time t;
       Biomass i_controi   =  biomass of species i, control scenario at time t;
       Biomass perturbed =  biomass of species i, perturbed scenario at time t.

A time-series of indices is written for each day of the simulation representing the similarity on
that date.  Separate indices are  written out for plants,  all animals, invertebrates only, and fish
only.
4.6 Biological Metrics

Ecological indicators are defined as primarily biological and are measurable characteristics of the
structure, composition, and function of ecological systems (Niemi  and McDonald 2004). The
term "indicator" as used by Niemi and McDonald is a rather broad one,  and includes two terms
often used  within  the biocriteria  program,   "metric" and "index".  A biological metric is a
numerical value that represents a quantitative community parameter, such as species diversity,
or percent EPT (see below).  A multimetric index is  a number that integrates several metrics to
express a site's condition or health,  such as an IBI (Index of Biological Integrity). AQUATOX
has the ability to calculate numerous metrics, some of which can be compared to similar metrics
derived from monitoring data. However,  there are limitations in the application of many such
metrics that reflect the differing capabilities of simulation models as opposed to field studies.
Models can predict continuing  complex responses  to changing conditions,  while field
measurements usually  represent  snapshots of  existing  conditions  with  limited  empirical
predictive power.  Aquatic models have limited taxonomic resolution and usually represent
biomass; most metrics  and  indices  applied  in  the field are based  on  detailed taxonomic
identifications and involve counting the numbers of individual organisms per sample. Therefore,
only a subset of possible indicators can be implemented with AQUATOX; however, given the
biologic realism of the model, the list is much more extensive than for other models.
                                       131

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


Biotic metrics and indices have been widely used  for several decades, stimulated in part by
inclusion in rapid bioassessment protocols (RBP) by  the US EPA (Plafkin et al. 1989).  Most are
applicable to streams and wadeable rivers (Barbour et al. 1999), though there is a suite of indices
(the trophic state indices) that were developed as a measure of eutrophication in lakes. Metrics
can be  calculated  for  algae, which  indicate  short-term impacts; macroinvertebrates, which
integrate short-term impacts on localized  areas; and  fish,  which are indicators of long-term
impacts over broad reaches (Barbour et al. 1999).

Ecological indicator measures fall into several well  defined categories.  Those metrics that are
presently calculated in AQUATOX are shown below in boldface; the others enumerated here can
be calculated offline using exported Excel output files:

   •   Composition—many  metrics  related to community  composition  are suitable  for
       simulation with  AQUATOX by selecting the appropriate "Benthic  metric designation"
       category on the underlying data screen; they include:
          o  % EPT (the following three combined) (Barbour et al.  1999)
                 •  % Ephemeroptera (mayfly larvae) (Maloney and Feminella 2006)
                 •  % Plecoptera (stonefly larvae) (Barbour et al. 1999)
                 •  %Trichoptera (caddisfly  larvae) (Barbour et al. 1999)
          o  % chironomids (midge larvae) (Barbour et al. 1999)
          o  % oligochaetes (aquatic worms) (Barbour et al.  1999)
          o  % Corbicula (invasive Asian clam) (Barbour et al. 1999)
          o  % Eunotia (interstitial  diatom characteristic of low-nutrient conditions) (Lowe et
             al. 2006)
          o  %   cyanobacteria  (cyanobacteria  characteristic   of  high-nutrient,  turbid
             conditions) (Trimbee and Prepas 1987) .

   •   Trophic—these include metrics that can be calculated from AQUATOX output:
          o  Periphytic chlorophyll a (Barbour et  al. 1999)
          o  Sestonic chlorophyll a (Barbour et al. 1999)
          o  % predators (can apply to both macroinvertebrates and fish) (Barbour et al. 1999)
          o  % omnivores (best applied to fish in AQUATOX) (Barbour et al. 1999)
          o  % forage or insectivorous fish (Barbour et al. 1999)

   •   Trophic  state—surrogates for  lake and reservoir algal biomass adjusted to  a common
       scale (Gibson et  al. 2000):
          o  TSI(TN) (total nitrogen)
          o  TSI(SD) (Secchi depth)
          o  TSI(CHL) (chlorophyll a)
          o  TSI(TP) (total phosphorus)

   •   Ecosystem bioenergetic—whole ecosystem metrics:
          o  Gross primary productivity, GPP (g O2/m2 d) (Odum 1971), more meaningful
             if expressed as an annual measure (g O2/m2 yr) (Wetzel 2001)
          o  Community respiration, R (g  O2/m   d)  (Odum 1971), more  meaningful if
             expressed as an annual measure (g O2/m2 yr) (Wetzel 2001)
                                      132

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


          o  P/R (ratio of GPP to community respiration) (Odum 1971)
          o  Turnover time (P/B, ratio of GPP to biomass in days) (Odum 1971)

In addition to those listed above, there are several ecological  indicators that are not suitable for
simulation modeling in general or for AQUATOX in particular:

   •   Richness—these are  based  on numbers of observed taxonomic  groups and are not
       suitable for simulation modeling;
   •   Tolerance/intolerance—based on number of tolerant or intolerant species and therefore
       unsuitable for modeling;
   •   Life cycle—percent of organisms with short or long life cycles, not easily modeled with
       AQUATOX.

The trophic state indices are applicable to lakes and reservoirs. They are lognormal-transformed
values that attempt to convert environmental variables to a  common value representing algal
biomass (Gibson et al. 2000):

       Secchi Depth (m):        TSI (SD)  = 60 - 14.41 ln(SD)
       Chlorophyll a (ug/L):     TSI (CHL) = 9.81 In(CHL) + 30.6
       Total Phosphorus (mg/L): TSI (TP)  = 14.42 ln(TP)  + 4.15
       Total Nitrogen (mg/L):   TSI(TN)   = 54.45 + 14.43 ln(TN)

The user can specify  over what time period the  indices are averaged.  This  enables better
comparison with field-derived TSIs,  which are generally calculated from samples taken during
the growing season. Obviously, chlorophyll a is the best representation of algal biomass, and that
metric should generally be used in determining the trophic state of a lake or reservoir (Table 8).
However, comparing the TSIs is also informative (Table 9).

The bioenergetic metrics are widely used by ecologists and have practical value as indicators of
accumulating organic matter (Odum  1971) and response to watershed disturbance (Dale and
Maloney 2004).
                                      133

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                        CHAPTER 4
           Table 8. Changes in Temperate Lake Attributes According to Trophic State
                 (Gibson et al. 2000, adapted from Carlson and Simpson 1996).
TSI
Value
<30



30-40


40-50






50-60




60-70






70-K)


>BO



SO
(mj
>a



B-4


4-2






2-1




0.5-1






0.25-
0.5

<0.25



TP
1 NTU

Iron, manganese,
taste, and odor
problems worsen
















Recreation




















Weeds, algal
scums, and
tow
transparency
discourage
swimming
and boating







Fisheries

Salmonid
fisheries
dominate

Salmonid
fisheries in
deep lakes
Hypolimnetic
anoxia
results in
loss of
salmon ids.
Walleye may
predominate
Warm-water
fisheries
only. Bass
may be
dominant










Rough fish
dominate.
summer fish
kills possible
     Table 9. Conditions Associated with Various Trophic State Index Variable Relationships
                                      (Gibson et al. 2000).
        Relationship Between TSI Variables
                   Conditions
       TSI (CHL) = TSI(CHL) = TSI(SD)
       TSI(CHL)>TSI(SD)

       TSI(TP)=TSI(SD)>TSI(CHL)
       TSI(SD) =TSI(CHL) >TSI{TP)

       TSI(TP) >TSI(CHL) =TSl(SD)
Algae dominate light attenuation
Large particulates, such as Aphanizomenon flakes.
dominate
Nonalgal particulates or color dominate light attenuation
Phosphorus limits aigal blomass (TN,'TP ratio greater than
33:1)
Zooplankton grazing, nitrogen, or some factor other than
phosphorus limits algal biomass
                                         134

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


Invertebrate Biotic Indices

As noted above, some invertebrate biotic indices can be readily computed by AQUATOX with
one caveat:  they are based on relative biomass  rather than numerical density (number  of
individuals representing a taxonomic group).  The simplifying assumption is that weights  of
individuals are  roughly comparable.  Of course, this is  not  actually the case, but individual
weights vary greatly depending on the growth stage, so use of biomass has less error.

Computation of benthic invertebrate  indices by AQUATOX  requires that  the taxonomic
affiliations be designated as:

   •   oligochaete (worm)
   •   chironomid (midge) and other fly larvae
   •   mayfly
   •   stonefly
   •   caddisfly
   •   beetle
   •   mussel
   •   other bivalve
   •   amphipod
   •   gastropod
   •   other.
With  these  designations AQUATOX  can compute %  EPT (Ephemeroptera  or mayflies,
Plecoptera or stoneflies, and Trichoptera or caddisflies) as a percentage of the total biomass of
benthic invertebrates. Mussels are excluded from the computation in AQUATOX because the
potential biomass of a single individual may exceed that of all other invertebrates. The EPT are
usually the most sensitive aquatic insect orders, so this index is often useful. A  detailed study of
Fort Benning, Georgia streams showed  that %Ephemeroptera, %Plecoptera, and %Trichoptera
were significantly inversely  correlated with the degree of disturbance in the watershed (Maloney
and Feminella 2006, Mulholland et al. 2007). The index has been used in evaluating remediation
(Purcell et  al.  2002).   The user is cautioned  to  ensure  that the  benthic metric  chosen is
appropriate for the region; e.g. one wouldn't expect Plecoptera to be prevalent in Florida, due to
their temperature preferences.

Chironomids  (midge larvae) are generally tolerant (Maloney and Feminella 2006), so the %
Chironomids index is useful as an indicator of disturbance (Figure 94).  Another index that might
indicate disturbance is the computed value of % Oligochaetes; however, that metric has exhibited
mixed results in Georgia (Maloney and Feminella 2006).
                                      135

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
          CHAPTER 4
   Figure 94. Example of % Chironomid index computed for Upatoi Creek, Fort Benning, Georgia;
   	observed values are courtesy of George Williams	
     100
     12/6/1999
              Upatoi Creek Ft Benning G A (Control)
                                                     100
 Obs%Chiro (percent)

• Pet Chironomid (%)
                     12/5/2001
                                      12/5/2003
                                      136

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
               CHAPTER 4
                               5.  REMINERALIZATION
5.1 Detritus

For the purposes of AQUATOX, the term  "detritus" is used to
include  all   non-living  organic  material  and   associated
decomposers (bacteria  and fungi).  As such, it includes both
particulate and dissolved material in the sense of Wetzel (1975),
but  it  also  includes  the microflora  and  is  analogous  to
"biodetritus" of Odum  and de  la Cruz  (1963) .   Detritus is
modeled as eight compartments:  refractory (resistant) dissolved,
suspended, sedimented,  and buried  detritus; and labile  (readily
decomposed)  dissolved,  suspended,  sedimented,   and buried
detritus (Figure 95).  This degree of disaggregation is considered
necessary to provide more realistic simulations of the detrital
food  web;  the bioavailability  of  toxicants, with orders-of-
magnitude differences in partitioning;  and biochemical oxygen demand, which depends largely
on the decomposition rates. Buried  detritus is considered to be taken out of active participation
in the functioning of the ecosystem.  In general, dissolved organic material is about ten times that
of suspended particulate matter in lakes and streams (Saunders, 1980), and refractory compounds
usually predominate; however, the proportions are modeled dynamically.

                        Figure 95. Detritus compartments in AQUATOX
Detritus: Simplifying
Assumptions

 • Refractory detritus does not
   decompose directly but is
   converted to labile detritus
   through colonization
 • Detrital sedimentation is
   modeled with simplifying
   assumptions (unless the
   sediment submodel for streams
   is included)
 • Biomass of bacteria is not
   explicitly modeled
detr.
fm.
detr. ^
fm.

Refractory
Dissolved

Refractory
Suspended
colonisation
~"^£X
colonization r-^
U--*
ingestion r~^_
L~-> *

Labile
Dissolved

Labile
Suspended
A sedimentation A sedimeLt;
-.,.-.,, \7 \7
scour Y scour y
detr.
ex
* c
Refractory
Sediments
A bujlal
pojsure Y
Refractory
Buried
colonization ,^_
ingestion r^_
ex
Labile
Sediments
buijlal
Doipure Y
Labile
Buried
^ detr.
fm.
decomp.r-^
U--- ^
detr.
ingestion p^^
decomp.r-^ +
ition
detr.
ingestion ^_^
decomp.p^ ^

onnection to detritivores + connection to nutrients
                                        137

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4




The concentrations of detritus in these eight compartments are the result of several competing

processes:



  dSuspRefrDetr        ,      ^   „    _ ,          TTr  ,      TTr  ,
  	—	  = Loading + Detrrm - Colonization - Washout + Washm
        dt                                                                        (141)

            - Sedimentation - Ingestion + Scour ± Sinking ± TurbDiff ± DiffusionSeg




         dSuspLabDetr       ,      ^  „    _ ,          ^
         	 = Loading + Detrrm + Colonization - Decomposition
               dt

                        - Washout + Washin- Sedimentation - Ingestion + Scour        (142)


                        + Sinking ± TurbDiff + Diffusion Seg




           dDissRefrDetr        ,      ^   „    _  ,           TTr  ,      TTr  ,
           	  = Loading + Detrrm - Colonization - Washout + Washin
                dt                                                               (143)

                          + TurbDiff ± Diffusion Seg



         dDissLabDetr       ,      ^  „   ^              TTr  ,      TTr  ,
         	 = Loading + Detrrm - Decomposition - Washout + Washin
               dt              6                 *                              (144)

                         +  TurbDiff + Diffusionseg




              dSedRefrDetr    T    ,.     ^   „    0  ,.      .    „
              	 = Loading + DetrFm + Sedimentation + Exposure
                   dt               *                             P              (145)

                         - Colonization - Ingestion - Scour - Burial



           dSedLabileDetr   T    ,.      ^   „    0  ,.      .     _  ,  .
           	 = Loading + DetrFm + Sedimentation + Colonization
                 dt                                                              (146)

                  - Ingestion - Decomposition - Scour + Exposure - Burial



               dBuriedRefrDetr    0  ,                 ,  0      „                 ,. ._
               	 = Sedimentation + Burial - Scour - Exposure          (147)
                      dt



              dBuriedLabileDetr   0  ,                 ,   0      „
              	 = Sedimentation + Burial - Scour - Exposure
                      dt                                                          (148)

where:

       dSuspRefrDetr/dt       =   change in concentration of suspended refractory detritus

                                 with respect to time (g/m3-d);

       dSuspLabileDetr/dt     =   change in concentration  of  suspended labile detritus with

                                 respect to time (g/m3-d);

       dDissRefrDetr/dt       =   change in concentration  of dissolved refractory detritus

                                 with respect to time (g/m3-d);
                                      138

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                           CHAPTER 4
       dDissLabDetr/dt

       dSedRefrDetr/dt

       dSedLabileDetr/dt

       dBuriedRefrDetr/dt

       dBuriedLabileDetr/dt

       Loading

       DetrFm
       Colonization

       Decomposition
       Sedimentation


       Scour


       Exposure

       Burial

       Washout
       Washin
       DiffusionSeg

       Ingestion

       Sinking

       TurbDiff
change  in concentration of dissolved labile  detritus with
respect to time (g/m3-d);
change  in concentration of sedimented  refractory detritus
with respect to time (g/m3-d);
change  in concentration of sedimented labile detritus with
respect to time (g/m3-d);
change  in concentration of buried  refractory detritus with
respect to time (g/m3-d);
change  in concentration of buried  labile detritus with
respect to time (g/m3-d);
loading of given detritus from nonpoint and point sources,
or from upstream (g/m3-d);
detrital formation (g/m3-d);
colonization of refractory detritus by decomposers (g/m3-d),
see (155);
loss due to microbial decomposition (g/m3-d), see (159);
transfer from suspended detritus to sedimented detritus by
sinking (g/m3-d); in streams with the inorganic  sediment
model attached see (235),  for all other systems see (165);
resuspension from  sedimented detritus (g/m3-d); in streams
with the inorganic  sediment model attached see (233),  for
all other systems see (165) (resuspension);
transfer from buried to sedimented by scour of overlying
sediments (g/m3-d);
transfer from  sedimented to deeply buried (g/m3-d),  see
(167b);
loss due to being carried downstream (g/m3-d), see (16);
loadings from upstream segments (g/m -d), see (30);
gain or loss due to  diffusive transport  over the  feedback
link between two segments, (g/m3-d), see (32);
loss  due  to ingestion by detritivores  and filter feeders
(g/m3.d), see (91);
detrital  sinking from epilimnion and to hypolimnion under
stratified conditions, see (165); and
transfer  between  epilimnion  and hypolimnion  due  to
turbulent diffusion (g/m3-d),  see (22) and (23).
As a simplification, refractory detritus is considered not to decompose directly, but rather to be
converted to labile detritus through microbial colonization.  Labile detritus is then available for
both decomposition and ingestion by detritivores (organisms that feed on detritus).  Because
detritivores digest microbes  and defecate the remaining organic  material,  detritus  has to be
conditioned through microbial colonization before it is suitable food. Therefore, the assimilation
efficiency of detritivores for refractory material is usually set  to  0.0, and the  assimilation
efficiency for labile material is increased accordingly.
                                       139

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 4
Sedimentation and scour (resuspension) are opposite processes. In shallow systems there may be
no long-term sedimentation (Wetzel et al., 1972), while in deep systems there may be little
resuspension. In the classic AQUATOX model, sedimentation is a function of flow, ice cover
and,  in very shallow water, wind based on simplifying assumptions.  Scour and exposure of
organic matter are applicable only in streams where they are keyed to the  behavior of clay and
silt.  Scour as an explicit function of wave and current action is not implemented, however, the
capability  to link to hydrodynamic models is  provided.  See chapter 6 for a discussion  of the
various inorganic sediment models and their implications to organic sediments.

Within AQUATOX, the user must specify the percentage particulate and percentage refractory
for each source of organic matter. Table 10 presents some guidance on populating these variables
based on Allan (1995), Hessen and Tranvik (1998), and Wetzel (2001).  These percentages can
be specified as constant variables or by using a time-series.

  Table 10. Suggested detrital boundary conditions based on literature and in the absence of data
Ecosystem
Oligotrophic lakes
Eutrophic lakes
Forested streams
Rivers
Blackwater stream
Particulate
%
10%
15%
20%
30%
5%
Refractory
%
90%
86%
60%
60%
95%
OM cone.
(mg/L)
4
24
5
14
26
AQUATOX simulates detritus as organic matter (dry weight); however, the user can input data
as organic carbon or carbonaceous biochemical oxygen demand (CBOD) and the model will
make the necessary conversions.  Organic matter is assumed to be 1.90 •  organic carbon as
derived  from  stoichiometry (Winberg  1971).    The conversion from BOD  includes  the
simplifying  assumption that any BOD data input  into the  model  are primarily based  on
carbonaceous oxygen demand:
                           OM = CBOD •
                  fCBOD5_CBODL
                  [   O2Biomass
(148b)
where:
                      CBOD5_CBODV =
                                                   1
                                         100% - PercentRefrTme
                                                          (148c)
and:
       OM
       CBOD
=   organic matter input as required by AQUATOX (g OM/m3);
=   carbonaceous biochemical demand 5-day from  user input (g
    /m3);
                                      140

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       CBOD5 CBODU  =  CBODs  to ultimate  carbonaceous  BOD  conversion factor,  also
                           defined as CBODu:CBOD5 ratio;
       PercentRefr Time   =  user-defined percent refractory matter for  given source of organic
                           matter, may be a time series;  and
       O2Biomass       =  ratio  O2 to organic matter (OM).  (remineralization parameter, the
                           default is 0.575 based on Winberg (1971));

AQUATOX has always  assumed that user-input BODs loadings are primarily  composed of
carbonaceous oxygen demand but this assumption has been made more explicit in Release 3.1.
The equations above are used by AQUATOX when converting initial conditions and loadings in
CBODs to organic matter, when estimating CBODs from organic matter for simulation output,
and when linking HSPF BOD data.  Equations (148b) and  (148c) are  new  to AQUATOX
Release 3.1 and a warning message is displayed if an older study that utilizes BOD loadings is
imported into the current version.


Detrital Formation

Detritus is formed in several ways:  through mortality, gamete loss, sinking of phytoplankton,
excretion and defecation:
                                 •etr = ^brota (Mort 2 detr, brota ' Mortality hwj                (149)


DetrFmD,SSRefrDetr = ^b,ota (Mort 2detr,b,ota' Mortality>biota) + ^blota(Excr2detr,blota • Excretion)      (150)


DetrFmDlssLMeDetr = Zfaoto (Mort 2detr,b,ota • Mortality hwta) + iWofa (Excr 2detr,bwta • Excretion)      (151)


DetrFmsmpLabaeDetr = Zb.ota (Mort 2 detr, biota • Mortality^) + l^ammah GameteLoss             (152)


DetrFmSedLabUeDetr = Spred (Def 2detr, pred ' Defecation^) + Compartment ('Sinking'compartment)         (153)


                     DetrFmSedRefrDetr = S pred (Def 2 detr, pred ' DefeCatlOHpred)
                                                                                    \    /
                    + I*compartment(Sedimentationc0mpartment • PlantSinkToDetr)
where:
       DetrFm         =   formation of detritus (g/m3-d);
       Mort2detr, biota    =   fraction  of given  dead  organism that goes  to given  detritus
                           (unitless);
       Excr2detr, biota    =   fraction  of excretion that goes to given  detritus (unitless),  see
                           Table 11;
       Mortality biota    =   death rate for organism (g/m3-d), see (66), (87) and (112);
       Excretion       =   excretion rate for organism (g/m3-d), see (64) and (111) for plants
                           and animals, respectively;
       GameteLoss     =   loss rate for gametes (g/m3-d), see (126);
                                       141

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
             tr, biota     ~
       Defecation VK&   =
       Sedimentation   =
       PlantSinkToDetr =
fraction of defecation that goes to given detritus (unitless);
defecation rate for organism (g/m3-d), see (97);
loss of phytoplankton to bottom sediments (g/m3-d), see (69); and
labile and refractory portions of phytoplankton (unitless, 0.92 and
0.08 respectively).
A fraction  of mortality, including  sloughing of leaves from macrophytes, is assumed to go to
refractory detritus; a much larger fraction goes to labile detritus. Excreted material goes to both
refractory and labile detritus, while gametes  are considered to be labile.   Half the defecated
material is  assumed to be labile because of the conditioning due to ingestion and  subsequent
inoculation with bacteria in the gut (LeCren  and Lowe-McConnell,  1980); fecal  pellets sink
rapidly  (Smayda,   1971),  so  defecation  is  treated  as  if it  were  directly to  sediments.
Phytoplankton that sink to the bottom (that are not linked to periphyton compartments) are
considered  to become detritus;  most are consumed quickly by zoobenthos (LeCren  and Lowe-
McConnell, 1980) and are not available to be resuspended.

                         Table 11. Mortality and Excretion to Detritus

Dissolved Labile Detritus
Dissolved Refractory Detritus
Suspended Labile Detritus
Suspended Refractory Detritus
Algal
Mortality
0.27
0.03
0.65
0.05
Macrophyte
Mortality
0.24
0.01
0.38
0.37
Bryophyte
Mortality
0.00
0.25
0.00
0.75
Animal
Mortality
0.27
0.03
0.56
0.14

Dissolved Labile Detritus
Dissolved Refractory Detritus
Algal
Excretion
0.9
0.1
Macrophyte
Excretion
0.8
0.2
Bryophyte
Excretion
0.8
0.2
Animal
Excretion
1.0
0.0
Colonization

Refractory detritus is converted to labile detritus through microbial colonization. When bacteria
and  fungi  colonize  dissolved  refractory  organic matter,  they are in effect  turning it into
particulate matter.   Detritus is usually refractory because  it has a deficiency  of nitrogen
compared to microbial biomass.  In order for microbes to colonize refractory detritus, they have
to take up additional nitrogen from  the water (Saunders et al., 1980).  Thus, colonization is
nitrogen-limited, as well as being limited by suboptimal temperature,  pH, and dissolved oxygen:
                  Colonization = ColonizeMax • DecTCorr • NLimit • pHCorr
                                 • DOCorrection • RefrDetr
                                                        (155)
where:
       Colonization
rate of conversion of refractory to labile detritus (g/m -d);
                                       142

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 4
       ColonizeMax
       Nlimit
       DecTCorr
       pHCorr
       DOCorrection

       RefrDetr
maximum colonization rate under ideal conditions (g/g-d);
limitation due to suboptimal nitrogen levels (unitless), see (157);
the effect of temperature (unitless), see (156);
limitation due to suboptimal pH level (unitless), see (162);
limitation due to  suboptimal oxygen level (unitless), see (160);
and
concentration of refractory detritus in suspension, sedimented, or
dissolved (g/m3).
Because microbial  colonization and decomposition involves microflora with  a wide range of
temperature tolerances, the effect of temperature is modeled in the traditional way (Thomann and
Mueller,  1987), taking the rate at an observed  temperature and correcting it for the ambient
temperature up to a user-defined, high maximum temperature, at which point it drops to 0:

                            DecTCorr = Thetaemp~TObs where
                             Theta = 1.047 if Temp > 19° else
                             Theta = 1.185- 0.00729 • Temp
                                                                                (156)
                          If Temp > TMax Then DecTCorr = 0
The resulting curve has a shoulder similar to the Stroganov curve, but the effect increases up to
the maximum rate (Figure 96).

              Figure 96. Colonization and decomposition as an effect of temperature
                               10
          H	1	1	1	1	1	1	1	h
          20    30    40    50    60    70
         TEMPERATURE (C)
                                      143

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


The nitrogen limitation construct, which is original with AQUATOX, is parameterized using an
analysis of data presented by Egglishaw (1972) for Scottish streams. It is computed by:
                              »n--  •         N-MinN                               „„.
                              NLimit =	                         (157)
                                      N - MinN + HalfSatN
                                 N = Ammonia + Nitrate                             (158)
where:
       N            =      total available nitrogen (g/m3);
       MinN        =      minimum level of nitrogen for colonization (= 0.1 g/m3);
       HalfSatN     =      half-saturation constant for nitrogen stimulation (= 0.15 g/m3);
       Ammonia     =      concentration of ammonia (g/m3); and
       Nitrate       =      concentration of nitrite and nitrate (g/m3).

Although it can be changed by the user, a default maximum colonization rate of 0.007 (g/g-d)  is
provided,  based  on Mclntire and Colby  (1978,  after  Sedell  et  al.,  1975).  The rates  of
decomposition (or colonization) of refractory dissolved  organic matter  are comparable to those
for particulate matter.  Saunders (1980) reported values of 0.007 (g/g-d) for a eutrophic lake and
0.008 (g/g-d) for a tundra pond. Anaerobic rates were reported by Gunnison et al. (1985).
Decomposition

Decomposition is the process by which detritus is broken down by bacteria and fungi, yielding
constituent nutrients, including nitrogen, phosphorus, and inorganic carbon.  Therefore, it is a
critical  process in modeling nutrient recycling.   In  AQUATOX, following a  concept first
advanced  by  Park et al.  (1974),  the  process  is modeled as a  first-order equation with
multiplicative  limitations  for suboptimal environmental  conditions (see section 4.1  for a
discussion of similar construct for photosynthesis):

          Decomposition = DecayMax • DOCorrection • DecTCorr • pHCorr • Detritus     (159)
where:
       Decomposition   =   loss due to microbial decomposition (g/m3-d);
       DecayMax      =   maximum decomposition rate under aerobic conditions (g/g-d);
       DOCorrection   =   correction for anaerobic conditions (unitless), see (160);
       DecTCorr       =   the effect of temperature (unitless), see (156);
       pHCorr         =   correction for suboptimal pH (unitless), see (162); and
       Detritus         =   concentration of detritus, including dissolved but not buried (g/m3).

Note that biomass of bacteria is not explicitly modeled  in AQUATOX.  In some models (for
example, EXAMS, Burns et al., 1982) decomposition is represented by a second-order equation
using an empirical estimate of bacteria biomass.  However, using bacterial biomass  as  a site


                                       144

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 4


constant would constrain the model, potentially forcing the rate.  Decomposers were modeled
explicitly as a part of the CLEAN model (Clesceri et al.,  1977).  However, if conditions are
favorable, decomposers can double in 20 minutes;  this can result in stiff equations,  adding
significantly to the computational time.  Ordinarily, decomposers will grow rapidly as long as
conditions  are  favorable.   The  only time the biomass  of decomposers  might need  to  be
considered explicitly is when a new organic chemical is introduced and the microbial assemblage
requires time to become adapted to using it as a substrate.

The effect of temperature on biodegradation is represented by Equation (156), which also is used
for colonization. The function for dissolved oxygen, formulated for AQUATOX, is:

                    r^^^      •     -^       /,  -^     \ KAnaerobic                ,+ fK^
                    DOC orrection = Factor + (1- Factor) --                (160)
                                                        DecayMax
where  the predicted  DO concentrations are entered  into a Michaelis-Menten formulation to
determine the  extent to which degradation rates are affected by ambient  DO  concentrations
(Clesceri, 1980; Park et al., 1982):
                              Factor- -       -                          (161)
                                       HalfSatO + Oxygen
and:
       Factor         =     Michaelis-Menten factor (unitless);
       KAnaerobic     =     decomposition rate at 0 g/m3 oxygen (g/m3-d or ug/L-d);  Set to
                            0.3 g/m3-d for microbial degradation of sediments.  For chemicals,
                            (160) is also used and the "rate of anaerobic microbial degr." from
                            the chemical underlying data is used (KMDegr Anaerobic).
       Oxygen         =     dissolved oxygen concentration (g/m3); and
       HalfSatO       =     half-saturation constant for oxygen (g/m3) (0.5 g/m3 in the water
                            column or 8.0 g/m3 for sedimented detritus).

DOCorrection accounts for both decreased  and increased (Figure 97)  degradation rates under
anaerobic conditions, with KAnaerobic/DecayMax having values less than one and greater than
one, respectively. Detritus will always decompose more slowly under anaerobic conditions; but
some organic chemicals, such as some halogenated compounds (Hill and McCarty, 1967), will
degrade more rapidly.  Half-saturation constants of 0.1 to 1.4 g/m3 have  been reported (Bowie et
al., 1985); a value of 0.5 g/m3 is used in the water column and a calibrated value of 8.0 g/m3 is
used for the sediments to force anoxic conditions.
                                       145

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                          CHAPTER 4
                         Figure 97. Correction for dissolved oxygen
                       0.2
                                     34567
                                        Dissolved Oxygen
                                                                10
                             •KAnaerobic = 1.3 —KAnaerobic = 0.3 — KAnaerobic = 0
Another important environmental control on the rate of microbial degradation is pH. Most fungi
grow optimally between pH 5 and 6 (Lyman et al., 1990), and most bacteria grow between pH 6
to about 9 (Alexander, 1977).  Microbial oxidation is most rapid between pH 6 and 8 (Lyman et
al.,  1990).  Within the pH range of 5 and 8.5, therefore, pH is assumed to not affect the rate of
microbial degradation, and the suboptimal factor for pH is set to 1.0.  In the absence of good data
on the rates of biodegradation under extreme  pH conditions,  biodegradation is represented as
decreasing exponentially beyond the optimal range (Park et al., 1980a; Park et al.,  1982).  If the
pH is below the lower end of the optimal range, the following equation is used:
                                  pHCorr = e(
                                         ,, _  (pH - pHMin)
                                                                                 (162)
where:
      pH        =
      pHMin    =
                        ambient pH, and
                        minimum pH below which limitation on biodegradation rate occurs.
If the pH is above the upper end of the optimal range for microbial degradation, the following
equation is used:
where:
                                  pHCorr = e^Mma-^                             (163)

      pHMax =     maximum pH above which limitation on biodegradation rate occurs.

These responses are shown in Figure 98.
                                      146

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
                              Figure 98. Limitation due to pH
                                     EFFECT OF pH
                           3.0  4.1   5.2   6.3   7.4  8.5  9.6  10.7
                                             PH
Sediment oxygen demand (SOD in g O2/m d) is also calculated by taking the sum of detrital
decomposition and then multiplying by O2Biomass (the ratio of oxygen to organic matter).  This
can be  compared with  SOD values  derived from  the  optional sediment  diagenesis model
(Chapter 7).

Sedimentation

Depending upon which options the user chooses, sedimentation (i.e., the sinking of suspended
particles to the sediment bed) is calculated differently (Table 12). When the inorganic-sediment
model (sand-silt-clay) is included, the sedimentation and deposition  of  detritus is  assumed to
mimic the  sedimentation and resuspension of silt (see (235) and (233)).   If the  multi-layer
sediment  model  is  included (using  user-input erosion  and  deposition  time-series)  the
sedimentation of detritus is calculated using the deposition velocity for cohesives (assumed to be
a surrogate for organic matter) as follows:
                                            DepVel  „
                             Sedimentation = —	State
                                             Thick
                                                        (164)
When the inorganic-sediment model or the multiple-layer sediment model are not included in a
simulation (i.e. "classic"  AQUATOX formulations are used),  the  sedimentation of suspended
particulate detritus to bottom sediments can be modeled using simplifying assumptions (165).
The constructs are intended to provide general responses to environmental factors, but they could
be considerably improved upon by linkage to a hydrodynamic model (currently only  available
with the multi-layer sediment model).
                           o  /•      •     KSed  „   , „
                           Sedimentation =	Decel • State
                                          Thick
                                                        (165)
where:
       Sedimentation   =
transfer from  suspended to sedimented by  sinking (g/m3-d),
negative is effectively Resuspension (see below);
if
                                       147

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
      KSed
      DepVel

      Thick
      Decel
      State
sedimentation rate (m/d);
user input time-series of deposition velocities for cohesives (multi-
layer model only; m/d);
depth of water or thickness of layer if stratified (m);
deceleration factor (unitless), see (166); and
concentration of particulate detrital compartment (g/m3).
           Table 12: Summary of Detrital Deposition and Resuspension in AQUATOX

 Deposition of Suspended Detritus & Phytoplankton

"Classic" AQUATOX model
Sand-Silt-Clay submodel
Multi-layer Sediment Model
Sediment Diagenesis
Assumption
Sedimentation is a function of Mean Discharge
Follows "silt" as calculated by the Sand-Silt-Clay
submodel
Follows "cohesives" class, (which may be user input
or calculated using the sand-silt-clay model)
Choice of "Classic" AQUATOX or Sand-Silt-Clay
assumptions
Equation
(165)
(235)
(164);
(235)

 Resuspention of Sedimented Detritus

"Classic" AQUATOX model
Sand-Silt-Clay submodel
Multi-layer Sediment Model
Sediment Diagenesis
Assumption
Resuspension is a function of Mean Discharge
Follows "silt" in inorganic sediments model
Follows "cohesives" class, (which may be user input
or calculated using the sand-silt-clay model)
Resuspension is not enabled.
Equation
(165)
(233)
(167);
(233)

If the discharge exceeds the mean  discharge then sedimentation is slowed proportionately
(Figure 99):
                      If TotDischarge > MeanDischarge then
where:
       TotDischarge
      MeanDischarge
                             Decel =
                                      MeanDischarge
                                        TotDischarge
                                  else Decel = 1.0
                                                                                (166)
 total epilimnetic and hypolimnetic discharge (m3/d); and
 mean discharge, recalculated on an annual basis at the beginning of
 each year of the simulation (m3/d).
                                      148

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                   CHAPTER 4
           Figure 99.  Relationship ofdecel to discharge with a mean discharge of 5 m3/s.
                                   6  8  10  12  14  16  18  20  22  24
                                      Discharge (cu m/s)
If the depth of water is less than or equal to 1.0 m and wind speed is greater than or equal to 5.5
m/s then the sedimentation rate is negative, effectively becoming the  rate of resuspension. For
plants, if the depth of water is is less than or equal to 1.0  m and wind speed is greater than or
equal to 2.5 m/s then the sedimentation rate is assumed to be zero. If there is ice cover, then the
sedimentation rate is doubled to represent the  lack of turbulence.

If the multi-layer sediment model  is included (using user-input erosion and deposition  time-
series) the resuspension  of detritus is  calculated using  the  erosion velocity  for  cohesives
(assumed to be surrogate for organics) as follows
where:
       Resuspension
       ErodeVel

       Thick
                                           ErodeVel  0  70
                            Resuspension =	SedState
                                 L           rr-rl • /
                                             i hick
                                                          (167)
transfer from sediment to suspended by erosion (g/m -d);
user input time-series of cohesives erosion velocities (multi-layer
model only m/d);
depth of water or thickness of layer if stratified (m);
Daily Burial

When the quantity of sedimented refractory detritus exceeds its initial condition, it is transferred
to the deeply buried category (buried detritus).
                    BurialDetntus = ABS(ConcDetntus - InitialConditionDetntm
                                                         (167b)
where:
       Buriali)etritus    =   daily burial of detritus (g/m-d);
       ConcDetritus     =   sedimented detritus concentration (g/m3)
                                        149

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                          CHAPTER 4
       InitialCondition =   initial condition of detritus (g/m )
5.2 Nitrogen

In the water column, two nitrogen compartments, ammonia
and  nitrate, are  modeled.    Nitrite  occurs  in  very  low
concentrations and is rapidly transformed through nitrification
and  denitrification (Wetzel, 1975); therefore,  it is  modeled
with nitrate. Un-ionized ammonia (NHs) is not modeled as a
separate state  variable but is  estimated  as  a fraction  of
ammonia (177). In the sediment bed, if the optional sediment
diagenesis   model is included  (see  chapter 7), nitrogen is
explicitly  modeled;  otherwise  inorganic  nitrogen  in  the
sediment bed is ignored,  but  organic nitrogen is implicitly
modeled as a component of sedimented detritus.

In the water column, ammonia  is assimilated  by  algae  and macrophytes and is converted to
nitrate as a result of nitrification:
                                         Nitrogen: Simplifying Assumptions

                                          • Nitrite is not explicitly modeled
                                          • Both nitrogen fixation and
                                           denitrification are subject to
                                           environmental controls; therefore,
                                           the nitrogen cycle is represented
                                           with considerable uncertainty.
                                          • Lethal effects from un-ionized and
                                           ionized ammonia  are  assumed
                                           additive.
                                          • Ammonia makes up stoichiometric
                                           imbalances between trophic levels.
           dAmmonia
               ~dt

where:
    dAmmonia/dt
    Loading
    Remineralization =
    Nitrify
    Assimilation
    Washout
    Washin
    DiffusionSeg
    TurbDiff
   Flux
        Diagenesis
   Loading + Remineralization - Nitrify -  Assimilation Ammoma

   - Washout + Washin ± TurbDiff + Diffusion Seg + FluxDjagenesjs
(168)
=   change in concentration of ammonia with time (g/m3-d);
=   loading of nutrient from inflow (g/m3-d);
=   ammonia derived from detritus and biota (g/m3-d), see (169);
=   nitrification (g/m3-d), see (174);
=   assimilation of nutrient by plants (g/m3-d), see  (171);
=   loss of nutrient due to being carried downstream (g/m3-d), see (16)
=   loadings from linked upstream segments (g/m3-d), see (30);
=   gain or loss due to diffusive transport over the feedback link between
    two segments, (g/m3-d), see (32);
=   depth-averaged   turbulent   diffusion   between   epilimnion   and
    hypolimnion if stratified (g/m3-d), see (22) and (23);
=   potential flux from the sediment diagenesis model, (g/m3-d),  see (273)
Remineralization includes all processes by which ammonia is produced from animal, plants, and
detritus, including decomposition and excretion required to maintain variable stoichiometry (see
Table 14):
                                         150

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                            CHAPTER 4
                                                                                   (169)
Remineralization = PhotoResp + DarkResp + AnimalResp + AnimalExcr
                  + DetritalDecomp + AnimalPredation + NutrRelDefecation
                  + NutrRelPlantSink + NutrRe Mortality + NutrRelGameteLoss
                  + NutrRelColonization + NutrRelPeriScour
where:
   PhotoResp          =  algal excretion of ammonia due to photo respiration (g/m3-d);
   DarkResp           =  algal excretion of ammonia due to dark respiration (g/m3-d);
                          excretion of ammonia due to animal respiration (g/m3-d);
                          animal  excretion of excess  nutrients to  ammonia to  maintain
                          constant org. to N ratio as required (g/m3-d);
                          nitrogen release due to detrital decomposition (g/m3-d);
                          change  in nitrogen content necessitated when an animal consumes
                          prey with a different nutrient content (g/m3-d), see discussion in
                          "Mass Balance of Nutrients" in Section 5.4;
                          ammonia released from animal defecation (g/m3-d);
                          ammonia balance from sinking of plants and conversion to detritus
                          (g/m3-d);
                          ammonia balance from biota mortality and conversion to detritus
                          (g/m3-d);
                          ammonia  balance  from gamete loss and  conversion  to detritus
                          (g/m3-d);
   NutrRelColonization =  ammonia balance from colonization of refractory detritus into labile
                          detritus (g/m3-d);
   NutrRelPeriScour   =  ammonia balance when periphyton  is scoured and converted to
                          phytoplankton and suspended  detritus. (g/m3-d);

Nitrate  is  assimilated  by plants  and  is  converted  to free   nitrogen (and  lost)  through
denitrification:
   AnimalResp
   AnimalExcr

   DetritalDecomp
   AnimalPredation
   NutrRelDefecation   =
   NutrRelPlantSink

   NutrRelMortality    =

   NutrRelGameteLoss  =
          dNitrate
             dt
                   = Loading + Nitrify - Denitrify - Assimmtrate - Washout + Washin
where:
       dNitrate/dt
       Washin
       Diffusionseg

       Loading
       Denitrify
       r lUX
                                                                                   (170)
                    + TurbDiff ± Diffusion Sfn + Flux
                                         Seg
                                                 Diagenesis
                         change in concentration of nitrate with time (g/m3-d);
                         loadings from linked upstream segments (g/m3-d), see (30);
                         gain or loss due to diffusive transport over the feedback link between
                         two segments, (g/m3-d), see (32);
                         user entered loading of nitrate, including atmospheric deposition;
                         denitrification (g/m3-d), see (175);
                         potential flux from the sediment  diagenesis model, (g/m3-d),  see
                         (273)
                                       151

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
Free nitrogen can  be fixed by  cyanobacteria.  Both nitrogen fixation and denitrification are
subject to environmental  controls and are difficult to model with any accuracy; therefore, the
nitrogen cycle is represented with considerable uncertainty.

                     Figure 100. Components of nitrogen remineralization
                                                      Free  N  (not in
                                                      model domain)
         Decomposition
                         Excretion
                     Denitrification
            Ammonia
                                Nitrification
           Assimilation
                     Assimilation
                                                                       Fixation
AQUATOX will estimate and output total nitrogen (TN) in the water column.  Total nitrogen is
the sum of ammonia and  nitrate in the water column as well as nitrogen associated  with
dissolved and suspended particulate organic matter and phytoplankton (see section 5.4 for further
details).

Assimilation

Nitrogen compounds are assimilated by plants as a function of photosynthesis in the respective
groups (Ambrose et al., 1991):
            Assimilation Ammoma = Zpiant( Photosynthesis plant • Uptake    en • NH4Pref)
                                                       (171)
           Assimilationmtrate = 'Lpiant (Photosynthesis plant • Uptake    en • (1 - NH4PreJ))      (172)
where:
       Assimilation    =
       Photosynthesis  =
assimilation rate for given nutrient (g/m -d);
rate of photosynthesis (g/m3-d), see (35);
                                       152

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


       UptakeNitr0gen   =    fraction of photosynthate that is  nitrogen (unitless,  0.01975  if
                           nitrogen-fixing, otherwise 0.079);
       NH4Pref       =    ammonia preference factor (unitless).

Only 23 percent of nitrate is nitrogen, but 78 percent of ammonia is nitrogen. This results in an
apparent preference for  ammonia.  The preference factor is  calculated with an  equation
developed by Thomann and Fitzpatrick (1982) and cited and used in WASP (Ambrose et al.,
1991):

               ^TTT^  r         N2NH4 • Ammonia • N2NO3 • Nitrate
              NH4Pref =
                         (KN + N2NH4 • Ammonia) • (KN + N2NO3 • Nitrate)
                                                                                   (173)
                                         N2NH4 • Ammonia • KN
                       (N2NH4 • Ammonia + N2NO3 • Nitrate) • (KN + N2NO3 • Nitrate)
where:
       N2NH4      =      ratio of nitrogen to ammonia (0.78);
       N2NO3      =      ratio of nitrogen to nitrate (0.23);
       KN          =      half-saturation constant for nitrogen uptake (g N/m3);
       Ammonia     =      concentration of ammonia (g/m3); and
       Nitrate       =      concentration of nitrate (g/m3).

For algae other than cyanobacteria, Uptake is the Redfield (1958) ratio; although other ratios (cf.
Harris, 1986) may be used by editing the parameter screen. At this time nitrogen-fixation by
cyanobacteria is represented by using a smaller uptake ratio, thus "creating" nitrogen.  Nitrogen
fixation is not tracked explicitly as a separate rate in the plant's derivative.
Nitrification and Denitrification

Nitrification is the conversion of ammonia to nitrite and then to nitrate by nitrifying bacteria; it
occurs at the sediment-water interface (Effler et al.,  1996) and in the water column (Schnoor
1996).  The maximum rate of nitrification is reduced by limitation  factors for  suboptimal
dissolved oxygen and pH,  similar to the way that decomposition is modeled, but using the more
restrictive correction for suboptimal temperature used for plants and animals:

                 Nitrify = KNitri • DOCorrection • TCorr • pHCorr • Ammonia            (174)

where:

       Nitrify       =     nitrification rate (g/m3-d);
       KNitri       =     maximum rate of nitrification (m/d);
       DOCorrection =     correction for anaerobic conditions (unitless) see (160);
       TCorr       =     correction for suboptimal temperature (unitless); see (59);
       pHCorr      =     correction for suboptimal pH (unitless), see (162); and
       Ammonia    =     concentration of ammonia (g/m3).
                                       153

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 4
If the Sediment Diagenesis model is used, the KNitri value may need to be decreased to account
for sediment nitrification being represented  separately.  The nitrifying  bacteria have narrow
environmental optima; according to Bowie et al. (1985) they require aerobic conditions with a
pH between 7 and 9.8, an optimal temperature of 30°, and minimum and maximum temperatures
of 10"and 60"respectively (Figure 101, Figure 102).
 Figure 101.  Response to pH, nitrification
                   Figure 102. Response to temperature, nitrification
              EFFECT OF pH
       5     6.4      7.8     9.2    10.6
          5.7     7.1     8.5     9.9
                      pH
                              STROGANOV FUNCTION
                                  NITRIFICATION
                                                § 0.8 4
                              10
20   30    40   50
TEMPERATURE (C)
60
Denitrification is the conversion of nitrate and nitrite to free nitrogen and occurs as an anaerobic
process.  However, only a small part of the denitrification occurs at the sediment-water interface
and it can  also occur in the water column due to "anoxic microsites" such as the interior of
detrital particles (Di Toro 2001).  Therefore, AQUATOX follows the convention of other models
in representing denitrification as a bulk process (by combining sediment and  water-column
denitrification). This approach is a change from  earlier model versions, including AQUATOX
Release 3.0, where denitrification processes at the sediment-water interface and in the water
column were considered separately.   Low oxygen levels enhance the denitfification process
(Ambrose et al., 1991):
             Denitrify = KDenitri • (1 - DOCorrection) • TCorr • pHCorr • Nitrate
                                                        (175)
where:
       Denitrify
       KDenitri
       TCorr
       pHCorr
       Nitrate
denitrification rate (g/m -d);
user-input maximum rate of denitrification (1/d);
effect of suboptimal temperature (unitless), see (59);
effect of suboptimal pH (unitless), see (162); and
concentration of nitrate (g/m3).
KDenitri might need to be reduced when the sediment diagenesis model is included, because
denitrification in the sediment bed is explicitly tracked within that model (see (278))
                                       154

-------
 AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 4
Furthermore, denitrification is  accomplished by a large number  of reducing bacteria  under
anaerobic conditions and with broad environmental tolerances (Bowie et al., 1985; Figure 103,
Figure 104).
Figure 103. Response to pH, denitrification
      1

     0.8

   p 0.6
   O
   Q 0.4
   LU
   IT
     0.2
               EFFECT OF pH
3     4.8     6.6     8.4    10.2
   3.9    5.7     7.5     9.3
               PH
                   Figure 104. Response to temperature, denitrif
                              STROGANOV FUNCTION
                                 DECOMPOSITION
                                        TOpt
                                                         10
                                   20   30   40   50
                                   TEMPERATURE (C)
60
 lonization of Ammonia
 The un-ionized form of ammonia, NHs, is toxic to invertebrates and fish.  Therefore, it is often
 singled out as a water quality  criterion.   Un-ionized ammonia  is in equilibrium  with the
 ammonium ion, NH4+, and the proportion is determined by pH and temperature.  It is useful to
 report NHs as well as total ammonia (NHs + NH4+).
 The computation of the fraction of total ammonia that is un-ionized is relatively straightforward
 (Bowie etal. 1985):
                                                   1
where:
       FracNH3
       pkh
       NH3
       Ammonia
       TKelvin
                              FmcNH3     =  -
                                              l + lQp-p

                             NH3   =  FracNH3 • Ammonia

                                                 2729.92
                               pkh  =   0.09018
                                                 TKelvin
                                                       (176)
                                                       (177)

                                                       (178)
fraction of un-ionized ammonia (unitless);
hydrolysis constant;
un-ionized ammonia (mg/L);
total ammonia (mg/L) see (168);
temperature (°K).
                                       155

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 4
The relative contributions of temperature and pH can be seen by graphing the fraction of un-
ionized ammonia against each of those variables in simulations of Lake Onondaga (Figure 105
and Figure 106). As inspection of the construct would suggest, un-ionized ammonia has a linear
relationship to temperature and a logarithmic relationship to pH, which causes it to be  sensitive
to extremes in pH.

            Figure 105.  Fraction of un-ionized ammonia roughly following temperature.
            0.07
            0.
            0.
                                    Fraction NH3
              Sep-88   Apr-89   Oct-89  May-90   Nov-90   Jim
           Figure 106. Fraction of un-ionized ammonia affected by extreme values of pH.
                                    Fraction NH3
              Sep-88   Apr-89   Oct-89   May-90  Nov-90   Jun-91
The construct was verified with the same set of data from Lake Onondaga as was used for the pH
verification (Effler et al. 1996), see section 5.7. It fits the observed data well (Figure 107).
                                      156

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                              CHAPTER 4
     Figure 107. Comparison of predicted and observed fraction of NH3 for Lake Onondaga, NY.
                              Data from (Effler et al. 1996).
                                   Fraction NH3
                                                        -FracNHS
                                                        -Obsfrac NH3
                                                        -Poly. (Obs frac NH3)
              Feb-  Apr- May- Jul-89 Aug- Oct-  Dec-
               89    89   89         89   89    89
Ammonia Toxicity

Lethal effects of ammonia on animals have been implemented in AQUATOX based on Update
of Ambient Water Quality Criteria for Ammonia (U.S. Environmental Protection Agency, 1999).
Based on this document, it is preferable to base toxicity on total ammonia, taking into account
the contributions from the un-ionized and ionized ammonia (LCSOu and LCSOi):
                  LC50,, =
f

U + ]
LC50,<8
R 1
I
npHT-S 1_i_1Q8~-P
\

J
                                                     (179)
                  LC50, =
                            1+
                                R            1
               i+:
                                                     (180)
where:
      LC50U
      LC50l
      J^(^J U total ammonia
 LC50 for the unionized concentrations of ammonia
LC50 for the ionized concentrations of ammonia.
LC50U + LC50,
                                     157

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 4
       pHT

       R


       LC50
            t,8
=    transition pH at which LC50 is the average of the high- and low-
     pH intercepts (7.204);
=    shape parameter  defined  as the  ratio of the high- and low-pH
     intercepts (0.00704),  along with pHj,  defines the shape  of the
     curve;
=    user-input LC50total ammonia at 20 degrees centigrade and pH of 8.
LC50 parameters derived with the equations above are then applied  to the external  toxicity
formulation (see  section 9.3,  equations  (429)-(431)).   The  slope of the Weibull  curve is  a
constant 0.7 for both forms of ammonia.  This value produces the best general match of data
from Appendix 6 from the Ammonia Criteria update  (U.S. Environmental  Protection Agency,
1999). Lethal effects from un-ionized and ionized ammonia are assumed to be additive.
5.3 Phosphorus

The  phosphorus cycle is  simpler than the nitrogen cycle.
Decomposition,  excretion, and assimilation are important
processes that are similar to those described above.   As was
the case with ammonia and nitrate, if the optional sediment
diagenesis  model  is  included (see Chapter  7),  flux of
phosphate from the sediment bed may be added to the water
column,  especially under  anoxic conditions.  Additionally,
sorption to calcite may have a significant effect on phosphate predictions in high pH systems due
to precipitation of calcium carbonate.   This optional  formulation is important to adequately
simulate  marl lakes.
                                    Phosphorus: Simplifying
                                    Assumption

                                     • Total bioavailable soluble
                                       phosphorus is modeled
                                     • A constant sorption rate for calcite
                                       is used
                                     • Soluble phosphorus makes up
                                       stoichiometric imbalances between
                                       trophic levels.
          dPhosphate        ,    „       ,                          TTr  ,
                     - = Loadmg+Remmemhzation - Assimilation?^^ -Washout
               dt
                      + Washin ± TurbDiff ± DiffusionSe - SorptionP + FluxDia,
                                                                          agenesis
                                                             (181)
                  Assimilation = ZPlant (Photosynthesis plant • Uptake    oms)
                                                             (182)
where:
       dPhosphate/dt
       Loading

       Remineralization
       Assimilation
       TurbDiff
     change in concentration of phosphate with time (g/m -d);
     loading  of nutrient  from  inflow and  atmospheric  deposition
     (g/m3-d);
     phosphate derived from detritus and biota (g/m3-d), see (183);
     assimilation by plants (g/m3-d);
     depth-averaged  turbulent  diffusion   between  epilimnion  and
     hypolimnion if stratified (g/m3-d), see (22) and (23);
                                       158

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                           CHAPTER 4
       Washout
       Washin
       Diffusionseg

       SorptionP
       Photosynthesis
       Uptake
                           loss of nutrient due to being carried downstream (g/m -d), see (16)
                           loadings from linked upstream segments (g/m3-d), see (30);
                           gain  or  loss  due to diffusive transport over  the feedback link
                           between two segments, (g/m3-d), see (32);
                           rate of sorption of phosphorus to calcite (mgP/L-d), see (218);
                           potential  flux from the sediment diagenesis model, (g/m3-d), see
                           (273)
                           rate of photosynthesis (g/m3-d), see (35), and
                           fraction of photosynthate that is phosphate (unitless, 0.018).
As was the case with ammonia, Remineralization includes all processes by which phosphate is
produced  from animal, plants, and  detritus, including decomposition, excretion, and  other
processes required to maintain mass balance given variable stoichiometry (see Table 15):
                                                                                  (183)
                          algal excretion of phosphate due to photo-respiration (g/m -d);
                          algal excretion of phosphate due to dark respiration (g/m3-d);
                          excretion of phosphate due to animal respiration (g/m3-d);
Remineralization = PhotoResp + DarkResp + AnimalResp + AnimalExcr
                  + DetritalDecomp + AnimalPredation + NutrRelDefecation
                  + NutrRelPlantSink + NutrRelMortality + NutrRelGameteLoss
                  + NutrRelColonization + NutrRelPeriScour
where:
   PhotoResp
   DarkResp
   AnimalResp
   AnimalExcr         =  animal excretion of excess nutrients  to  phosphate to maintain
                          constant org. to P ratio as required (g/m3-d);
                          phosphate release due to detrital decomposition (g/m3-d);
                          change in phosphate content necessitated when an animal consumes
                          prey with a different nutrient  content  (g/m3-d), see discussion in
                          "Mass Balance of Nutrients" below;
                          phosphate released from animal defecation (g/m3-d);
                          phosphate balance from  sinking of plants and conversion to detritus
                          (g/m3-d);
   NutrRelMortality    =  phosphate balance from biota mortality and conversion to detritus
                          (g/m3-d);
   NutrRelGameteLoss  =  phosphate balance from gamete loss  and conversion to detritus
                          (g/m3-d);
   NutrRelColonization =  phosphate balance from colonization  of  refractory  detritus  into
                          labile detritus (g/m3-d);
   NutrRelPeriScour    =  phosphate balance when periphyton  is scoured and  converted to
                          phytoplankton and suspended detritus. (g/m3-d);

At this time AQUATOX models only phosphate  available for plants; a correction factor in the
loading screen allows the user to scale total phosphate loadings to available phosphate. A default
value is  provided for average atmospheric  deposition, but  this should be  adjusted for site
   DetritalDecomp
   AnimalPredation
   NutrRelDefecation
   NutrRelPlantSink
                                       159

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                     CHAPTER 4
conditions.  In particular,  entrainment of dust from tilled fields and new highway construction
can cause significant increases in phosphate loadings.   As with nitrogen, the default uptake
parameter is the Redfield (1958) ratio; it may be edited if a different ratio is desired (cf. Harris,
1986).

AQUATOX estimates and outputs total phosphate  (TP) in the water column.  TP is the sum of
dissolved phosphate  in the water column as well  as phosphate associated with dissolved and
suspended paniculate organic matter and phytoplankton(see section 5.4 for further details).
5.4 Nutrient Mass Balance

Variable Stoichiometry

The ratios of elements in organic matter are allowed to
vary  among  but not  within  compartments.    This is
accomplished by providing editable fields for N:organic
matter  and  P:organic matter for  each  compartment.
Furthermore,  the wet  to dry  ratio  is  editable  for  all
compartments; it has a default value of 5.

In  order to  maintain  the  specified  ratios  for  each
compartment, the model explicitly accounts for processes
that balance the ratios during transfers,  such as excretion
coupled  with consumption and nutrient uptake  coupled
with  detrital  colonization.    Nutritional  value  is  not
automatically related to Stoichiometry in the model, but it
is implicit in default egestion values  provided with various food sources.  Table 13  shows the
default stoichiometric values suggested for the model, although these can be edited.

                     Table 13: Default stochiometric values in AQUATOX
Nutrient Mass Balance: Simplifying
Assumptions

 • Stoichiometry within each model
   compartment is constant over time
 • Free nitrogen is not tracked within
   AQUATOX
 • Nutrients taken up by macrophyte
   roots come from sources  that are
   outside the modeled system
 • Mass  balance  may  fail  if total
   nutrients in the water column drop
   to  zero  (due  to  inter-organism
   interactions)
 • Ammonia loadings are assumed to
   be  12 to 15% when total nitrate
   loadings are input by the user.
 • Dissolved  nutrients  make  up
   stoichiometric imbalances  between
   trophic levels.
Compartment
Refrac. detritus
Labile detritus
Phytoplankton
Cyanobacteria
Periphyton
Macrophytes
Cladocerans
Cope pods
Zoobenthos
Minnows
Shiner
Perch
Smelt
Bluegill
Trout
Bass
Frac. N
(dry)
0.002
0.079
0.059
0.059
0.04
0.018
0.09
0.09
0.09
0.097
0.1
0.1
0.1
0.1
0.1
0.1
Frac. P
(dry)
0.0002
0.018
0.007
0.007
0.0044
0.002
0.014
0.006
0.014
0.0149
0.025
0.031
0.016
0.031
0.031
0.031
Reference
Sterner & Elser 2002
Redfield (1958) ratios
Sterner & Elser 2002
same as phytoplankton for now
Sterner & Elser 2002
Sterner & Elser 2002
Sterner & Elser 2002
Sterner & Elser 2002
same as cladocerans for now
Sterner & George 2000
Sterner & George 2000
Sterner & George 2000
Sterner & George 2000
same as perch for now
same as perch for now
same as perch for now
                                         160

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4
Nutrient Loading Variables

Often water  quality data are given as total nitrogen and phosphorus.   In  order to improve
agreement with monitoring data, AQUATOX can accept both loadings and initial conditions as
"Total N" and "Total P."  This approach is made possible by accounting for the nitrogen and
phosphorus  contributed  by suspended and dissolved detritus and phytoplankton and back-
calculating the amount that must be available as freely dissolved nutrients.  The precision of this
conversion is aided by the model's variable stoichiometry.  For nitrogen:

                        ^Dissolved ~  * Total   * SuspendedDetritus    * SuspendedPlants                  \*™^)

where:
        N Dissolved       =    bioavailable dissolved nitrogen (g/m3 d); see (170);
        N Total         =    loadings of total nitrogen as input by the user (g/m3 d);
        ATSuspendedDetritus =    nitrogen in suspended detritus loadings (g/m3 d);
        N SuspendedPlants  =    nitrogen in suspended plant loadings (g/m3 d).

When Total N inputs are used, ammonia is assumed to be a fixed percentage of bioavailable
dissolved nitrogen, based on the type of input:

   •   Inflow waters: Ammonia content of dissolved inorganic nitrogen = 12%
   •   Point sources: Ammonia content of dissolved inorganic nitrogen = 15%
   •   Non-point sources:  Ammonia content of dissolved inorganic nitrogen = 12%

These percentages are based on professional judgement, they are averages from several large
data sets.  However, if the user  wishes to  use  a different percentage, separate ammonia and
nitrate data sets can be derived from the Total N time-series and input individually.

In acknowledgment of the way it is used in  the model, the  phosphorus  state  variable is
designated "Total Soluble P." Phosphorus that is not bioavailable (i.e. immobilized phosphorus
and acid-soluble phosphorus) may be specified using the FracAvail parameter as shown here:

                 TSP = FracAvail(PTotal - PSmpendedDetntm - PSuspendedPlants)                 (185)

where:
        TSP           =    bioavailable phosphorus (g/m3 d); see (181);
        FracAvail     =    user-input bioavailable fraction of phosphorus;
        PTotal         =    loadings of total phosphorus (g/m3 d);
        PSuspendedDetntus =    phosphorus in suspended detritus loadings (g/m3 d);
        P SuspendedPlants  =    phosphorus in suspended plant loadings (g/m3 d).
Nutrient Output Variables

In order to compare model results with monitoring data, total phosphorus, and total nitrogen are
calculated as output variables. This approach is accomplished by the reverse of the calculations
                                        161

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


for the loadings: the contributions of the nutrient in the  freely dissolved  state and tied up in
phytoplankton and dissolved and paniculate organic matter are calculated and summed.

Carbonaceous biochemical  oxygen demand (CBODs) is estimated considering  the  sum  of
detrital  decomposition.    The contributions  from phytoplankton  and labile dissolved  and
particulate organic matter are included using an oxygen to biomass conversion factor entered in
the remineralization record.

Mass Balance of Nutrients

Variables for tracking mass balance and nutrient fate are included in the output as  detailed
below. Phosphorus and Nitrogen balance mass to machine accuracy.   To maintain mass  balance,
nutrients are tracked through many interactions.

The mass balance and nutrient fate tracking variables are:

       Nutrient Tot. Mass: Total mass of nutrient in the system in kg
       Nutrient Tot. Loss: Total loss of nutrient from system since simulation start, kg
       Nutrient Tot. Washout: Total washout since simulation start, kg
       Nutrient Wash, Dissolved: Washout in dissolved form since simulation start, kg
       Nutrient Wash, Animals:  Washout in animals  since start, kg
       Nutrient Wash, Detritus:  Washout in detritus since start, kg
       Nutrient Wash, Plants:  Washout in plants since start, kg
       Nutrient Loss Emergel: Loss of nutrients in emerging insects since start, kg
       Nutrient Loss Denitrif.: Denitrification since start, kg
       Nutrient Burial: Burial of nutrients since start, kg
       Nutrient Tot. Load: Total nutrient load since start, kg
       Nutrient Load, Dissolved: Dissolved nutrient load since  start, kg
       Nutrient Load as Detritus: Nutrient load in detritus since start, kg
       Nutrient Load as Biota: Nutrient load in biota since start, kg
       Nutrient Root Uptake:  Load of nutrients into sytem via macrophyte roots since start. (Macrophyte root
              uptake is currently assumed to occur from below the modeled sediment layer), kg
       Nitrogen Fixation: Load of nitrate into system since start via nitrogen fixation, kg
       Nutrient MB Test: Mass balance test, total Mass + Loss - Load: Should stay constant
       Nutrient Exposure: Exposure of buried nutrients
       Nutrient Net Layer Sink:  For stratified systems, sinking since start, kg
       Nutrient Net TurbDiff:  For stratified systems,  Turbdiff since start, kg
       Nutrient Net Layer Migr.: For stratified systems, migration since start, kg
       Nutrient Total Net Layer: Net nutrient movement to or from paired vertical layer, kg (This value is the
              sum of sinking, turbulent diffusion  and migration. This quantity also accounts for nutrient
              transport caused by water movement when the thermocline depth changes.)
       Nutrient Mass Dissolved:  Total mass of dissolved nutrient in system, kg
       Nutrient Mass Detritus:  Total mass of nutrient in detritus in system, kg
       Nutrient Mass Animals:  Total mass of nutrient in animals in system, kg
       Nutrient Mass Plants:  Total mass of nutrient in plants in system, kg

It is important to make careful note of the units presented in the list above.  Load and loss terms
are calculated in terms of "kg since the start of the simulation," total mass units are "kg at the
current moment."

Simplified diagrams  of the nitrogen and phosphorus cycles can  be found in Figure  108 and
                                           162

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 4


Figure 109.  A full accounting of the 18 nutrient linkages and all external loads and losses for
nitrogen and phosphorus is also provided in Table 14 and Table 15.

There are instances in which nutrients can be moved to and from compartments that are not in
the model domain.  For example, when NOs undergoes denitrification and becomes free nitrogen
the free nitrogen is no longer tracked within AQUATOX.  An example of nutrients entering the
model domain comes with the growth of macrophytes. Rooted macrophytes are not limited by a
lack  of nutrients in the water column as  nutrients are  derived from the sediment.  Therefore,
when photosynthesis  of macrophytes produces growth, the nutrient content within the leaves of
the macrophytes is assumed to originate from the pore waters of the sediments. However, this
implicit "nutrient pumping" is tracked in the mass balance output.  Nitrogen fixation is another
addition of nutrients  from outside of the  model domain that  is  tracked with the mass balance
output varaible called "N fixation."

Additionally, some simplifications are required as a result of  dietary imbalances. For example,
herbivores generally have higher nutrient concentrations than the plants that they are consuming.
When biomass is converted from a plant into an animal through consumption the imbalance has
to be satisfied to maintain mass balance. Sterner and Elser (2002) state: "There is no single way
that  consumers  maintain their  stoichiometry in the face of imbalanced resources."   As a
simplification, AQUATOX takes nutrients from the dissolved water-column compartments to
make up this difference (see AnimalPredation in (169)). However, these same herbivores ingest
plants  with  higher nutrient concentrations than the fecal matter that they defecate.  When
biomass is  converted from plants to  detrital  matter through defecation the  model simulates a
release of  nutrients  into  the  water  column (see  NutrRelDefecation  in  (169)). These two
simplifying algorithms, therefore, balance each other for the most part, and such interactions will
have only a minor effect on predicted water-column nutrient concentrations. Likewise, nutrient-
poor refractory detritus is converted to labile detritus through microbial colonization and growth;
this is stimulated by uptake of nutrients from the water column (Sterner and Elser 2002) and is
represented in the model.
                                      163

-------
Figure 108
                 Nitrogen Cycle in AQUATOX
                              mortality, defecation, gameteloss
                                                           animals
               detritus
                                             plants
                                                               excretion,
                                                               respiration
                            assimilation
                                      excretion, respiration
                                                            macrophyte
                                                            root uptake
                            dissolved in water
                                      nitrification
                                                                 N in pore waters
   free nitrogen
(not in model domain) |' demtrification
                                                              | (outside model domain) j

-------
 Table 14
   Nitrogen Mass Balance: Accounting
NO3 link
Load external load
Nitrif from NH4 a
De N itrif external loss
NOSAssim to plant b
Washout external loss
TurbDiff layer accountg
Detritus,
Participate Refr. link
Load external load
mortality from anim/plt k
Colonz to PartLabDetr g
Washout external loss
Predation to Animal h
Sedimentation to SedRefrDetr i
Scour from SedRefrDetr j
SinkToHyp layer accountg
SinkFromEpi layer accountg
TurbDiff layer accountg
NH4 link
Load external load
Nitrif to NO3 a
Assimil to plant b
Excretion from anim/plt c,o
Respiration from anim/plt m,n
DetritalDecomp from LabileDetr d
Washout external loss
TurbDiff layer accountg
Detritus, Participate
Labile link
Load external load
Decomp to NH4 d
mortality from anim/plt k
GamLoss from Animal q
Colonz from Diss.PartRefr g
Washout external loss
Predation to Animal h
Sedimentation to SedLabDetr i
Scour from SedLabDetr j
SinkToHypo layer accountg
SinkFromEpi layer accountg
TurbDiff layer accountg
Detritus, Sed.
Refractory link
Load external load
Defecation from animal e
Plant Sedmtn from plant f
Colonz to SedLabDetr g
Predation to Animal h
Sedimentation from PartRefrDetr i
Scour to PartRefrDetr j
Burial external loss
Exposure external load
Algae link
Load external load
Photosyn from NO3, NH4 b
Respiration to NH4 m
Photo Resp to diss detr, NH4 l,c
Mortality to Diss / Part Detr k
Predation to Animal h
Washout external loss
Sedimntn (Sink) to Sed Detr f
TurbDiff layer accountg
SinkToHypo layer accountg
SinkFromEpi layer accountg
Sloughing to detr., phytoplk r
ToxDislodge to detr., as mort k
Detritus, Sed.
Labile link
Load external load
Defecation from animal e
Plant Sedmtn from plant f
Colonz from SedRefrDetr g
Predation to Animal h
Decomp to NH4 d
Sedimentation from PartLabDetr i
Scour to PartLabDetr j
Burial external loss
Exposure external load
Macrophytes link
Load external load
Photosyn root uptake, external
Respiration to NH4 m
Photo Resp to diss detr, NH4 l,c
Mortality to Part Detr k
Predation to animal h
Breakage to detr., as mort k
Detritus,
Dissolved link
Load external load
Decomp (labile) to NH4 d
Mortality from anim/plt k
Colonz DissRefr->PartLab g
Excretion from anim/plt 1
Washout external loss
TurbDiff layer accountg
Animals link
Load external load
Consumption from anim/plt h
Defecation to sed detr e
Respiration to NH4 if req. n
Excretion to NH4 if req. o
TurbDiff layer accountg
Predation to animal h
Mortality to Part Detr k
Gamete Loss to PartLabDetr q
Drift external loss
Entrain external loss
Promotion to animal p
Recruit from animal p
Emergel external loss
Migration layer accountg
Linkage Notes
 a  Denitrification from NH4 to NO3.
 b  An appropriate quantity of NO3 and NH4 are taken into a plant as part of photosynthesis so that mass balance is maintained.
 c  When excretion & respiration takes place in plants and animals, all nitrogen lost goes directly to dissolved NH4.
 d  Labile detritus breaks down and the nutrient content is released  as NH4.
 e  Defecation is split into sedimented-labile and sed-refr detritus 50-50. Excess nitrogen is released as NH4.
 f  Plants sink and are split into sedimented-labile and sed-refr detritus (92-08). Excess nitrogen is released as NH4.
 g  Refractory detritus converts into labile detritus. Any nitrogen imbalance  is balanced using NH4 in water.
 h
 j
   Animals eat plants and detritus.  Animal homeostasis (const, org to N ratio) is managed through Respiration & Excretion.
   Suspended sediment sinks and joins bottom sediment. Any change in N between phases is made up using dissolved NH4.
   Bottom sediment is scoured up and joins suspended sediment.  Any change in N between phases is made up using dissolved NH4.
 k  Animals and plants die and are divided up among suspended and dissolved detritus. Excess nitrogen is released as NH4.
 I  Plants excrete organic matter to dissolved detritus.  Excess Nitrogen is released as NH4.
 m  Plant respiration, nutrients are released to NH4
 n  Animal respiration, nutrients are relased to NH4 to maintain animal constant org. to N ratio as required.
 o  Animal excretion of excess nutrients to NH4 to maintain constant org. to N ratio as required.
 p  If young and old age-classes have different ratios, a warning is raised. Prom/Recr takes place  outside derivatives so ratios must match.
 q  Through gameteloss,  biomass is converted to Part Lab Detr. Excess Nitrogen is released as NH4.
 r  1/3 of periphyton sloughing goes to phytoplankton, 2/3 to detritus as mortality.  Nutrients are balanced between compartments.

-------
      Figure 109
                   Phosphorus Cycle in AQUATOX
                                  mortality, defecation, gameteloss
Oi
Oi
                                               ingestion
                       decomp.
       Loadings
plants
                                         excretion, respiration
                                       phosphate
                                   dissolved in water
                 animals
                                                           ingestion
excretion,
respiration
              macrophyte
              root uptake
                                                                             Washout
                                                               I  P in pore waters    !
                                                               j (outside model domain) j

-------
Table 15
    Phosphorus Mass Balance: Accounting
Total Soluble P link
Load external load
Assimilation to plant b
Excretion from anim/plt c,o
Respiration from anim/plt m,n
DetritalDecomp from LabileDetr d
Washout external loss
TurbDiff layer accountg

Detritus,
Participate Refr. link
Load external load
mortality from anim/plt k
Colonz to PartLabDetr g
Washout external loss
Predation to Animal h
Sedimentation to SedRefrDetr i
Scour from SedRefrDetr j
SinkToHyp layer accountg
SinkFromEpi layer accountg
TurbDiff layer accountg
Detritus, Sed.
Refractory link
Load external load
Defecation from animal e
Plant Sedmtn from plant f
Colonz to SedLabDetr g
Predation to Animal h
Sedimentation from PartRefrDetr i
Scour to PartRefrDetr j
Burial external loss
Exposure external load

Detritus,
Participate link
Load external load
Decomp to TSP d
mortality from anim/plt k
GamLoss from Animal q
Colonz from Diss.PartRefr g
Washout external loss
Predation to Animal h
Sedimentation to SedLabDetr i
Scour from SedLabDetr j
SinkToHypo layer accountg
SinkFromEpi layer accountg
TurbDiff layer accountg
Detritus, Sed.
Labile link
Load external load
Defecation from animal e
Plant Sedmtn from plant f
Colonz from SedRefrDetr g
Predation to Animal h
Decomp to TSP d
Sedimentation from PartLabDetr i
Scour to PartLabDetr j
Burial external loss
Exposure external load

Algae link
Load external load
Photosyn from TSP b
Respiration to TSP b
Photo Resp todissdetr, TSP l,c
Mortality to Diss / Part Detr k
Predation to Animal h
Washout external loss
Sedimntn (Sink)to Sed Detr f
TurbDiff layer accountg
SinkToHypo layer accountg
SinkFromEpi layer accountg
Sloughing to detr., phytoplk r
ToxDislodge to detr., as mort k
Detritus,
Dissolved link
Load external load
Decomp (labile) to TSP d
Mortality from anim/plt k
Colonz DissRefr->PartLab g
Excretion from anim/plt 1
Washout external loss
TurbDiff layer accountg

Macrophytes link
Load external load
Photosyn root uptake, external
Respiration to TSP m
Photo Resp to diss detr, TSP l,c
Mortality to Part Detr k
Predation to animal h
Breakage to detr., as mort k

Animals link
Load external load
Consumption from anim/plt h
Defecation to sed detr e
Respiration toTSPifreq. n
Excretion to TSP if req. l,o
TurbDiff layer accountg
Predation to animal h
Mortality to Part Detr k
GameteLoss to PartLabDetr q
Drift external loss
Entrain external loss
Promotion to animal p
Recruit from animal p
Emergel external loss
Migration layer accountg
 Linkage Notes
  b An appropriate quantity of phosphorus is taken into a plant as part of photosynthesis so that mass balance is maintained.
  c When excretion & respiration takes place in plants and animals (organic matter becomes DOM) additional P lost goes directly to dissolved P.
  d Labile detritus breaks down and the nutrient content is released as dissolved P.
  e Defecation is split into sedimented-labile and sed-refr detritus 50-50. Excess phosphorus is released as dissolved P.
  f Plants sink and are split into sedimented-labile and sed-refr detritus (92-08).  Excess phosphorus is released as dissolved P.
  g Refractory detritus breaks down into labile detritus. Any P imbalance is balanced using dissolved P in water.
  h Animals eat plants and detritus. Animal homeostasis (const, org to P ratio) is managed through Respiration & Excretion.
  i  Suspended sediment sinks and joins bottom sediment. Any change in P between phases is made up using dissolved P.
  j  Bottom sediment is scoured up and joins suspended sediment. Any change in P between phases is made up using dissolved P.
  k Animals and plants die and are divided up among suspended and dissolved detritus. Excess phosphorus is released as dissolved P.
  I  Plants and animals excrete organic matter to dissolved detritus.  Excess phosphorus is released as dissolved P.
  m Plant respiration, nutrients are  released to dissolved phosphorus.
  n Animal respiration, nutrients are relased to dissolved P to  maintain animal constant org. to P ratio as required.
  o Animal excretion of excess nutrients to P to maintain constant org. to P ratio as required.
  p If young and old age-classes have different ratios, a warning is raised. Prom/Recr takes place outside derivatives so ratios must match.
  q Through gameteloss, biomass is converted to Part Lab Detr.  Excess phosphorus is released as dissolved P.
  r 1/3 of periphyton sloughing goes to phytoplankton, 2/3 to detritus as mortality. Nutrients are balanced between compartments.

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
          CHAPTER 5
In some cases, when concentrations of nutrients in the water column drop to zero, perfect mass
balance of nutrients will not be maintained. Nutrient to organic matter ratios within organisms
do not vary over time, therefore transformation of organic matter (e.g. consumption, mortality,
sloughing, and sedimentation) occasionally requires  that a nutrient difference be made up from
the water column.  If there are no available nutrients in the water column, a slight loss of mass
balance is possible.

The mass associated with each component can be plotted, as in Figure 110.

  	Figure 110 Distribution of predicted mass of nitrogen in Lake OnondagaNY.	
          ONONDAGA LAKE, NY (PERTURBED) Run on 03-23-09 3:29 PM
                          (Epilimnion Segment)
       1.0E+6
       1.0E+5
       1.0E+4
       1.0E+3
 NMass Dissolved (kg)
- N Mass Susp. Detritus (kg)
- N Mass Animals (kg)
• N Mass Plants (kg)
 N Mass Bottom Sed. (kg)
          1/12/1989   5/12/1989
                           9/9/1989
                                   1/7/1990
                                           5/7/1990
                                                   9/4/1990
                                                            1/2/1991
                                            168

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                              CHAPTER 5
5.5 Dissolved Oxygen

Oxygen is an important regulatory endpoint; very low levels
can result in mass mortality for fish and other organisms,
mobilization   of nutrients   and   metals,  and  decreased
degradation of toxic organic materials.  Dissolved oxygen is
usually simulated as a daily average and does not account for
diurnal fluctuations (however, see Diel Oxygen below).  It is
a  function   of  reaeration,  photosynthesis,  respiration,
decomposition, and nitrification:
                            Oxygen: Simplifying Assumptions

                             • Reaeration is set to zero if ice cover
                               is predicted
                             • Cyanobacteria  blooms  limit  the
                               depth of oxygen reaeration
                                                                   	,
                  = Loading + Reaeration + Photosynthesized - BOD - 7 Respiration
           dt                                                     ^               (186)
                    - NitroDemand - Washout + Washin ± TurbDiff ± DiffusionSeg

                   Photosynthe sized = O2Photo • ^Plant (Photosynthesis plant)              (187)

                     BOD = O2Biomass • (^Detntus (DecompositionDetntus) )                (188)

                               NitroDemand = O2N • Nitrify                          (189)
where:
       dOxygen/dt
       Loading
       Reaeration
       Photosynthesized
       O2Photo
       BOD
       NitroDemand
       Washout
       Washin
       Diffusionseg

       O2Biomass
       Photosynthesis
       Decomposition
       £ Respiration
       O2N
       Nitrify
change in concentration of dissolved oxygen (g/m -d);
loading from inflow (g/m3-d);
atmospheric exchange of oxygen (g/m3-d), see (190);
oxygen produced by photosynthesis (g/m3-d);
ratio of oxygen to photosynthesis (1.6, unitless);
instantaneous biochemical oxygen demand (g/m3-d);
oxygen taken up by nitrification (g/m3-d);
loss due to being carried downstream (g/m3-d), see (16);
loadings from linked upstream segments (g/m3-d), see (30);
gain or loss due to diffusive transport over the feedback link
between two segments, (g/m3-d), see (32);
ratio of oxygen to organic matter (unitless);
rate of photosynthesis (g/m3-d), see (35), (85);
rate of decomposition (g/m3-d), see (159);
sum of respiration for all organisms (g/m3-d), (63) and (100);
ratio of oxygen to nitrogen (unitless); and
rate of nitrification (gN/m3-d) see (174).
Reaeration is a function of the depth-averaged mass transfer coefficient KReaer, corrected for
ambient temperature, multiplied by the difference between the dissolved oxygen level and the
saturation level (cf. Bowie et al., 1985):
                                           169

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 5
                         Reaeration = KReaer • (O2Sat - Oxygen)                    (190)
where:
       Reaeration   =      mass transfer of oxygen (g/m3-d);
       KReaer       =      depth-averaged reaeration coefficient (1/d);
       O2Sat        =      saturation concentration of oxygen (g/m3), see (198); and
       Oxygen       =      concentration of oxygen (g/m3).

For reaeration in estuaries, see Chapter 10 and equation (445).

In conditions where ice cover is assumed, as well as in the hypolimnetic segment of a stratified
simulation, Reaeration is generally set to zero.  However, to prevent excessive oxygen buildup
under these conditions, oxygen is not allowed to  exceed  two times saturation (O2Sat).  Any
oxygen buildup beyond two times saturation is added to Reaeration as a loss term.

KReaer may be entered as a constant value within the site's "underlying data." Alternatively,
AQUATOX will calculate KReaer based on the site-type and other characteristics.  In standing
water KReaer is computed as a minimum transfer velocity plus the effect of wind on the transfer
velocity (Schwarzenbach  et al., 1993) divided by the thickness of the mixed layer to obtain a
depth-averaged coefficient (Figure 111):
                                             Thick
where:
       Wind        =      wind velocity 10m above the water (m/sec);
       864          =      conversion factor (cm/sec to m/d); and
       Thick        =      thickness of mixed layer (m).

Algal blooms  can generate dissolved oxygen levels that are as much as 400% of saturation
(Wetzel, 2001).  However, near-surface cyanobacteria blooms, which are modeled as being in
the top 0.1 m,  produce high levels of oxygen that do not extend significantly into deeper water.
An adjustment is made in the code so that if the cyanobacteria biomass exceeds 1 mg/L and is
greater than other phytoplankton biomass, the thickness subject to oxygen reaeration is set to 0.1
m. This does not affect the KReaer that is used in computing volatilization (see section 8.5).

In streams, reaeration is a function of current velocity and water depth (Figure 1 12) following the
approach of Covar (1978,  see Bowie et al., 1985) and used in WASP (Ambrose et al., 1991).
The decision rules for which equation to use are taken from the WASPS code (Ambrose et al.,
1991).

If Vel< 0.5 18 m/sec:

                                   TransitionDepth = 0                              (1 92)
else:
                                          170

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 5


                             TransitionDepth = 4.411- Vel2'9135                        (193)
where:
       Vel                 =     velocity of stream (converted to m/sec) see (14); and
       TransitionDepth      =     intermediate variable (m).

If Depth < 0.61 m (but > 0.06), the equation of Owens et al. (1964, cited in Ambrose et al., 1991)
is used:
                            KReaer = 5.349 • Vel   • Depth                           (194)
where:
       Depth        =      mean depth of stream (m).

Otherwise, if Depth is > TransitionDepth, the equation of O'Connor and Dobbins (1958, cited in
Ambrose et al., 1991) is used:

                             KReaer = 3.93- Vela5° • Depth1'50

Else, if Depth < TransitionDepth but not <0.60 m, the equation of Churchill et al. (1962, cited in
Ambrose et al., 1991) is used:

                            KReaer = 5.049 • Vel0'97 • Depth'1'67                        (195)
In extremely shallow streams, especially experimental  streams where depth is < 0.06 m,  an
equation developed by Krenkel and Orlob (1962, cited in Bowie et al. 1985) from flume data is
used:
                              „„      2 34 -(U- Slope)0'408
                             KReaer =	      /                               (196)
                                              H
where:
       U     =      velocity (converted to fps);
       Slope =      longitudinal channel slope (m/m); and
       H     =      water depth (converted to ft).

If reaeration due to wind exceeds that due to current velocity, the equation for standing water is
used.  Reaeration is set to 0 if ice cover is expected (i.e., when the depth-averaged temperature <
3deg.  C).
                                          171

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 5
Figure 111. Reaeration as a Function of Wind
    10
   LU
   < 4
   LU *•
   o:
              EFFECT OF WIND
             OXYGEN, DEPTH = 1 m
                      8      12      16
                  6      10      14
                   WIND (mis)
                         Figure 112. Reaeration in Streams
                                       VELOCITY (m/sec)
Reaeration is assumed to be representative of 20 deg. C, so it is adjusted for ambient water
temperature using (Thomann and Mueller 1987):
where:
       KReaer
       Theta        =
       Temperature  =
     T^T~*        T^T~*      ^n  i (Temperature-20)
     KReaerT = KReaer2o • Theta

     Reaeration coefficient at ambient temperature (1/d);
     Reaeration coefficient for 20deg. C (1/d);
     temperature coefficient (1.024);  and
     ambient water temperature (deg. C).
                                                                                   (197)
In Release 3, oxygen saturation is calculated using the formulation of Thomann and Mueller
(1987, p 277), see also APHA et al (1995).   Oxygen saturation is calculated as a  function of
temperature (Figure 113), salinity (Figure 114), and altitude (Figure 115):
O2Sat = AltEffect • exp
 where
                                 1.57570E + 5  6.6423 LE + 7   1.2438E + 10
              TKelvin
      8.62195E + 11
                                                 TKelvin1
TKelvin5
                               TKelvin4
                                         _S  0017674-^+
                                  TKelvin  TKelvin2
           100-(0.0035- 3.28083 -Altitude]
AltEffect =	-
                       100
                                                                                     (198)
and where:
       AltEffect

       TKelvin
       S
       Altitude
Fractional reduction in oxygen saturation due to the effects of altitude
(Thomann and Mueller 1987, from Zison et al. 1978);
Kelvin temperature;
salinity driving variable, set to zero if not included in model (ppt); and
site specific altitude (m).
                                          172

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                           CHAPTER 5
Figure 113. Saturation as a Function of Temp.



                = 0           = 0 m
                                              Figure 114. Saturation as a Function of Salinity
                                                 10.0
                                                           Oxygen Saturation
                                                        Temperature = 20C, Altitude = Om
~-J
JE
,2
re
3
5



14
12
10
8
6
4


0 %,
0
o 	 '\v 	
0 	 "*'*---•••*-_
""—--==•-..
""'•"^-.-.::j...
o 	
0 	 i 	 i 	 i 	 i
0 10 20 30 40
Temperatyre(C)
I "~~
! Qfl
! c
! 3
|3



9
9
8.
8
7
7


5
o K^ 	
5 ! "^--^
o -i ""*^--...
5 i ""*****...
0 4 	 1 	 :
0 10 20 30 40
Salinity (ppt)

                                    Oxygen Saturation
                                Temperature = 20 C, Salinity = 0 ppt
                           10

                        3  9

                        •=•  8
                        ,2

                        3
                        (^  £*
                        m  o

                            5
500    1000
  Altitude (m)
                                                   1500
                                                           2000
Diel Oxygen

Significant fluctuations in oxygen are possible over the course of each day, particularly under
eutrophic conditions.  This type of fluctuation may now be captured within AQUATOX when
the model is run with an hourly time-step.  If the model is run with a larger reporting time step
(but an hourly integration time-step) the minimum and maximum oxygen concentrations will be
output on the basis of the hourly results.
The instantaneous light climate (28) affects the photosynthesis within the system and this, in
turn, affects the amount  of oxygen released into  the  water  column (187).  To assist in this
simulation, hourly oxygen loadings may be input  into AQUATOX if such data are available.
Alternatively, the effects of oxygen loadings and washout may be turned off, assuming that
upstream processes governing oxygen are producing water concentrations identical to the current
stream segment being modeled; in this way, in-stream processes can be analyzed without being
dominated by upstream loadings.
                                          173

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 5


AQUATOX can also output minimum and maximum predicted oxygen levels if the "data storage
stepsize" is greater than the "simulation step size." In other words if a simulation is set up with
an hourly step size but results are integrated daily,  AQUATOX will plot the mininum  and
maximum hourly prediction during that day.


Lethal Effects due to Low Oxygen

AQUATOX represents both lethal  and non-lethal effects from low concentrations of dissolved
oxygen.  The  US EPA  saltwater criteria document suggests the following general model for
estimating time to mortality based on  data from two species  of saltwater juvenile fish,  one
species  of juvenile freshwater  fish, and three species of saltwater larval crustaceans (U.S.
Environmental Protection Agency, 2000, Equation 9):

                      LCTime = Slope _ • \n(LC24hours) + Intercept _                  (199)
                                   exptime                      exptime
where:
                                = Lethal Concentration for a given percentage of a population
                                over the given duration (mg/L);
                           Slope ^ = 0.191-ZC24fe_ +0.064                      (200)
                                exptime


and


                         Intercept^_ = 0.392 • LC24hours + 0.204                     (201)
                                  exptime
To produce a general model of low oxygen effects, concentrations at which different percentages
are killed (holding duration constant) also need to be related to one another.  That is to say, a
model that relates LC5 to LC50 to LC95  must be produced.  Examining available data (Figure
116 to Figure 118), a linear model seems appropriate
                 LCFracduration = Slope  conc  • LCKnowndumtion + Intercept  conc             (202)
                                   pctkilled                         pctkilled
where:
       LCFracduration    =   concentration at which given percentage of organisms are killed
                           estimated  from a known lethal  concentration (holding  duration

                                          174

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                     CHAPTER 5
                           constant).
       LCKnown^ration  =   known lethal concentration for a given percentage of organisms at
                           the given duration.

Further examination of available data indicates different slopes for different species (Figure
119).  Most important, however, is  that for all species, the range of slopes is  quite narrow,
ranging from -0.001 to -0.01.  This indicates that for all  species and all durations, the range at
which  mortality occurs  due to insufficient oxygen is  quite  narrow.   For this reason, the
intermediate value of -0.007 was chosen as it is likely to reproduce available data reasonably
well.  This is preferable to having a user input this slope as these data are unlikely to be available
to most users.  Given a known lethal  concentration  at a known duration and using this slope, the
Intercept can be calculated see (204).
                Figure 116. Menhaden percent killed vs. O2 exposure concentration
                              Menhaden Pet Killed vs. Cones
                             atvarious exposure times (hours)
                                 20
40      60
Percent Killed
80
100
                                           175

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                          CHAPTER 5
                 Figure 117.  Blue Crab percent killed vs. O2 exposure concentration
                         BlueCrabPct Killed vs. Cones, at various
                                 exposure times (hours)
                        1.2

                          1

                        0.8
                       CO
                       o
                       °0.6

                        0.4  -

                        0.2

                          0
      + 1.047
                      +•  6-12
                      *  14-24
                      	Linear(6-12)
                      	Linear(14-24)
                            0
50          100
  Percent Killed
150
                   Figure 118. Spot percent killed vs. O2 exposure concentration
                                Spot Pet. Killed vs. Cones at various
                                     exposure times (hours)
                                y=-0.0019x +0.7678
                            y=-0.0016x +0.5778
* 1
A 4
x 24
t- 72
I inonr C~*\

m 2
6
• 48
- 72


                                   20
                                           40       60
                                          Percent Killed
                                                                   100
                                              176

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 5


-0.002 -
5 -0.004 -
"2 -0.006 -
o
O
g. -0.008 -
55
-0.01 -
n n-io -
Figure 119. Slope
Slope Conc./PctKillei
• •

*\
* +
•
; vs. species type
d vs. Species Type
• u

* Menhaden, Blue Crab
• Spot
^ ^






                            0    20    40    60    80    100   120
                                          duration
Combining equations (199) to (202),  given a user input 24-hour lethal concentration (in the
"Animal underlying data" screen), the model can calculate the fraction killed at a given duration
and at a given concentration.
            PctKilled =
                         O2Conc - 0.204 + 0.064 • \n(ExpTime)
                             0.191- \n(ExpTime) + 0.392
                                 - Intercept
                                                                      pctkilled
                                              -0.007
                                                        (203)
where:
                   Intercept conc   = LCKnownduration + 0.007 • PctKilledKnown
                           pctkilled
                                                        (204)
and:
       PctKilled
       ExpTime
       LCKnOWnduration
       PctKilledKnown
estimated percent  killed at a  given  oxygen  concentration and
exposure time;
concentration of oxygen (mg/L);
exposure time (hours);
user input lethal concentration (24-hour) (mg/L);
user input percentage for lethal concentration (percentage);
The model presented in equation (203) requires a user to input 24-hour lethal concentration as

                                          177

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                              CHAPTER 5
this is the basis for the general model presented in the saltwater criteria document. If a user has a
lethal  concentration at  a different duration,  the user  must  estimate  the  24-hour lethal
concentration, bearing  in mind that  the relationship between  exposure time  and lethal
concentrations is usually  logarithmic in  nature (Figure 120).   There are insufficient data to
develop a general model that will estimate 24-hour lethal concentrations given different user
input durations.

AQUATOX tracks oxygen concentrations over the previous 96 hours from the current time-step.
The oxygen effects model is then applied with the durations  shown below:

   •   1 hour, 4  hours, 12 hours (when model is run with hourly time-step only)
   •   1 day, 2 days, 4 days (relevant to both hourly and daily time-steps)

AQUATOX finds the  minimum oxygen  concentration over  each of these time-periods  and
applies it to equation (203). The maximum percent killed over all of the durations tested is then
applied to the animal biomass by increasing mortality (equations (90) and (112)) .

Figure 121  shows an example of a three-dimensional response surface produced by this model.
This is a model of low oxygen lethality for Atlantic menhaden produced by entering a 24-hour
LC95 of 0.61  mg/L.  Figure 122 shows  model predictions using a 24 hour LC50 of 3 mg/L
overlaid on a figure  from the U.S. Environmental Protection Agency's 1986 Quality Criteria for
Water.  This plot shows that the default value of 3 mg/L works well  for many species, but for
white bass,  for example, the LC50 should be set to a lower concentration.

              Figure 120. LC50 to exposure time based on data from U.S. EPA 2000
                             LCSOto Exposure Time Relationships,
                                    Menhaden and Spot
                     1

                   0.9

                   0.8

                   0.7
                 o) 0.6  •

                 o 0.5  ^
                 o
                 -1 0.4
                   0.3  •

                   0.2

                   0.1

                     0
y= 0.0762ln(x) +0.6332
                  y= 0.0526ln x +0.4773
          Spot 88 mm

          Atlantic menhaden 132 mm

          Log. (Spot88mm)

          Log. (Atlantic menhaden 132mm)
                       0
    20         40          60
       Exposure Time, Hours
80
                                          178

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 5
           Figure 121. Example of low O2 lethality model- menhaden response surface
                120
              (Exposure
               Time in
               Hours)
                                       179

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                    CHAPTER 5
      Figure 122.  96-hour model predictions (in red) compared against continuous exposure data
            (Data from U.S. EPA 1986, model set up using a 24-hour LC50 of 3.0mg/L)
— . ~I20
"o
|5 IOO
GO
1 00
o
o
*0 6O
•
Q_
a 20
CO o

T 1 1 O 1 1 1 I
0 j_^Z_£,JJM.
1 °^ *
o/|^ >
'
— i
O Largemouth Bass
!• O Black Crappie
•^ White Sucker
1m ^ White Bass
1 • Northern Pike
• • Channel Catfish
A Walleye
* v SmallrnouTh Bas:
i • j /
2 3 4 56789 IO
Dissolved Oxygen (mg/L)
Non-Lethal Effects due to Low Oxygen

The same three dimensional model used for lethal effects is utilized to calculate non-lethal low
oxygen effects (functions  of exposure level and time.)  In this case, EC50 reproduction affects
the fraction of gametes that are lost and EC50 growth affects consumption rates.
          O2EffectFrac =
                         'O2Conc - 0.204 + 0.064 • \n(ExpTime)
                               0.191-\n(ExpTime) +0.392
                                                      - Intercept
                                                                       pctkilled
                                               -0.007
                                                                           (205)
and:
where:
                         Intercept conc  = EC50duration + 0.007 • 50
                                 pctkilled
O2EffectFrac    =
                                                                           (206)
                           calculated fraction of gametes lost or reduction in growth rate at a
                           given oxygen concentration and exposure time;
                           concentration of oxygen (mg/L);
                                          180

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                        CHAPTER 5
       ExpTime        =   exposure time (hours);
       EC50duration      =   user input 50% effect concentration (24-hour) (mg/L);

O2EffectFrac is then applied to ingestion (91) and gamete loss (126).

5.6 Inorganic Carbon

Many  models  ignore  carbon  dioxide  as  an  ecosystem
component (Bowie et al.,  1985).   However,  it can  be an
important limiting nutrient.  Similar to other nutrients,  it is
produced by  decomposition and is  assimilated by plants; it
also is respired by organisms:
                                        Carbon Dioxide: Simplifying
                                        Assumptions

                                         • Atmospheric exchange is treated
                                          similar to that for oxygen.
                                         • For saltwater systems, an alternative
                                          option is to import a time-series of
                                          equilibrium CO2 levels.
             dCO2
              dt
where:
= Loading + Respired ^Decompose - Assimilation - Washout

   + Washin± CO2AtmosExch ± TurbDiff ± Diffusion,,
                    Respired = CO2Biomass • ^Orgamsm( Respiration Orgamsm)

                    Assimilation = ^Plant( Photosynthesis plant • UptakeCO2)
(207)
                                                               (208)

                                                               (209)
                     Decompose = CO2Biomass • ^Detntus(DecompDetntus)
                                                               (210)
and where:
       dCO2/dt
       Loading
       Respired
       Decompose
       Assimilation
       Washout
       Washin
       Diffusion^

       CO 2A tmosExch
       CO2Biomass
       Respiration
       Decomposition
       Photosynthesis
       UptakeCO2
              change in concentration of carbon dioxide (g/m3-d);
              loading of carbon dioxide from inflow (g/m3-d);
              carbon dioxide produced by respiration (g/m3-d);
              carbon dioxide derived from decomposition (g/m3-d);
              assimilation of carbon dioxide by plants (g/m3-d);
              loss due to being carried downstream (g/m3-d), see (16);
              loadings from linked upstream segments (g/m3-d), see (30);
              gain or loss  due  to diffusive transport over the feedback
              link between two  segments, (g/m3-d), see (32);
              interchange of carbon dioxide with atmosphere (g/m3-d);
              ratio of carbon dioxide to organic matter (unitless);
              rate of respiration (g/m3-d), see (63) and (100);
              rate of decomposition  (g/m3-d), see (159);
              rate of photosynthesis (g/m3-d), see (35); and
              ratio of carbon dioxide to photosynthate (= 0.53).
                                          181

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 5
Carbon  dioxide also  is exchanged with the atmosphere;  this process is important, but is not
instantaneous:  significant undersaturation and over saturation are possible (Stumm and Morgan,
1996). The treatment of atmospheric exchange is similar to that for oxygen:
                       CO2AtmosExch = KLiqCO2 • (CO2Sat - CO2)
                                                       (211)
In fact, the mass transfer coefficient is based on the well-established reaeration coefficient for
oxygen, corrected  for  the  difference  in  diffusivity of carbon dioxide as recommended by
Schwarzenbach et al. (1993):
KLiqCO2 = KReaer
                                               MolWtO2
                                                         ,0.25
                                                                                  (212)
where:
       CO 2A tmosExch
       KLiqCO2
       C02
       CO2Equil
       KReaer

       MolWtO2
       MolWtCO2
interchange of carbon dioxide with atmosphere (g/m3-d);
depth-averaged liquid-phase mass transfer coefficient (1/d);
concentration of carbon dioxide (g/m3);
equilibrium concentration of carbon dioxide (g/m3), see (213);
depth-averaged reaeration coefficient for oxygen (1/d), see (191)-
(195);
molecular weight of oxygen (=32); and
molecular weight of carbon dioxide (= 44).
Keying the mass-transfer coefficient for carbon dioxide to the reaeration coefficient for oxygen
is very powerful in that the effects of wind (Figure 123) and the velocity and depth of streams
can be represented, using the oxygen equations (Equations (191)-(195)).

                       Figure 123.  Carbon dioxide mass transfer
                                    EFFECT OF WIND
                                CARBON DIOXIDE, DEPTH = 1 m
                                         6   8  10  12  14  16
                                         WIND (m/s)
Based on this approach, the predicted mass transfer under still conditions is 0.92, compared to
the observed value of 0.89 + 0.03 (Lyman et al., 1982).  This same approach is used, with minor
modifications, to predict the volatilization of other chemicals (see Section 8.5). Computation of
equilibrium of carbon dioxide is based on the method in Bowie et al. (1985; see also Chapra and
                                          182

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 5
Reckhow, 1983) using Henry's law constant, with its temperature dependency (Figure 124), and
the partial pressure of carbon dioxide:
                             CO2Equil = CO2Henry • pCO2
where:
                  CO2Henry = MCO2 • 1Q
                                         2385.73
                                               -14.0184 + 0.0152642 • TKelvin
                                                      (213)


                                                      (214)
and where:
       CO2Equil
       CO2Henry
      pC02
      MCO2
       TKelvin
       Temperature
                             TKelvin = 273.15+ Temperature
equilibrium concentration of carbon dioxide (g/m3);
Henry's law constant for carbon dioxide (g/m3-atm):
atmospheric partial pressure of carbon dioxide (= 0.00035);
mg carbon dioxide per mole (= 44000);
temperature in deg.K, and
ambient water temperature (deg. C).
                                                      (215)
                       Figure 124. Saturation of carbon dioxide
                              CARBON DIOXIDE SATURATION
                                 7.5
          12 16.5 21 25.5 30 34.5  39
            TEMPERATURE (C)
The equilibrium CO2 equations described above cannot be applied to a seawater system as the
chemistry in seawater is significantly different from freshwater.  Over the years, several models
and constants used to describe the dissociation of carbon dioxide in seawater have been proposed
by investigators.

For saline conditions, the equilibrium parameters of the CO2 system can be derived by using
CO2SYS (Yuan, 2006) or CO2calc (USGS, 2010) and the results used as inputs for CO2Equil in
the AQUATOX simulation.  Within these models, the user needs to provide two  of the five
measurable CO2  system  parameters:  Total alkalinity (TA),  Total carbon dioxide (TCO2), pH
and Partial pressure of carbon dioxide (pCO2) or fugacity of carbon dioxide (fCO2); along with
temperature  (T),  pressure (P) and  salinity (S). The user can then select appropriate constants
                                         183

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 5


from proposed literature values and the program will calculate the remaining carbonate system
parameters including a time-series of CO2 concentrations in water.

For maximum flexibility, AQUATOX 3.1 has an interface that will accept these time-series of
CO2Equil.  In this  manner, the user can select the most appropriate model for their site and
import these values into the AQUATOX interface.  A time-series of pH can also be estimated by
these ocean-water chemistry models.

5.7 Modeling Dynamic pH
Dynamic pH is important in simulations for several reasons:
                                                             Assumptions
      o   pH affects the ionization of ammonia and potential
                                                             Dynamic pH: Simplifying
                                                              • Simple semi-empirical formulation
                                                              • Computation is good for the pH
                                                               range of 3.75 to 8.25
          resulting toxicity;
      o   pH affects the hydrolysis and ionization of organic
          chemicals which potentially has effects on chemical
          fate and the degree of toxicity;
      o   pH also affects the decay of organic matter and denitrification of nitrate which could
          eventually feed back to the animals;
      o   if pH exceeds 7.5, calcite precipitation can take place which has  a significant effect
          on the food-web.

A user-input time-series of pH levels may be used to drive  the  model or AQUATOX  can
calculate pH levels.

Many models follow the example of Stumm and  Morgan (1996) and  solve  simultaneous
equations for pH,  alkalinity, and  the  complete carbonate-bicarbonate equilibrium  system.
However, this approach requires more data than are often available,  and the iterative solution of
the equations entails  an additional computational burden — all for a precision  that is unnecessary
for ecosystem models.  The alternative is to restrict the range of simulated pH to that of normal
aquatic systems and to make simplifying assumptions that allow a semi-empirical computation of
pH (Marmorek et al.  1996, Small and Sutton 1986). That is the approach taken for AQUATOX.

The computation is good for the pH range of 3.75 to 8.25, where the carbonate ion is negligible
and can thus be ignored.  (Any predictions above 8.25 are truncated to 8.25 and any  predictions
below 3.75  are  set to 3.75.)   The derivation is  given  by  Small  and Sutton (1986),  with a
correction for dissolved organic carbon (Marmorek et al.  1996).   It incorporates a quadratic
function of carbon  dioxide;  and it is  a  nonlinear function of mean  alkalinity and  the
concentration of refractory dissolved organic carbon (humic  and fulvic acids), by means of an
inverse hyperbolic sine function:
                                                Alkalinity -5.1  -DOC}
                                            „
                  pHCalc   =  A + B • ArcSmH
                                                         C         J              (216)
                                          184

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 5
where:
      pHCalc
      ArcSinH
      Alkalinity
      DOC

      5.1
pH;
inverse hyperbolic sine function;
mean Gran alkalinity (ueq CaCOs/L);
refractory dissolved organic carbon (mg/L); sum of (143), (144);

average ueq of organic ions per mg of DOC;
       B  =  l/ln(10)
       Alpha  =  H2CO3 * • CCO2 + pkw
       H2CO3*  =  10
where:
         H2CO3*   =    first acidity constant;
         CCO2     =    CO2 expressed as ueq/L; see (207) multiplied by conversion factor
                         of 22.73 (ueq/mg);
         pkw        =    ionization constant for water (le-14);
         T         =    temperature (°C); see (24);
         0.92       =    correction factor for dissolved CO2.

Calibration and verification of the construct used data from nine lakes and ponds in the National
Eutrophication Survey (U.S. Environmental Protection Agency, 1977), two observations on Lake
Onondaga, NY, from before and after closure of a chlor-alkali plant (Effler et al., 1996), and one
observation in a river (Figure 125). The correction factor for CO2 was  obtained by fitting the
data to the unity line, but ignoring the two highest points because the construct does not  predict
pH above 8.25.
                                         185

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                      CHAPTER 5
          Figure 125.  Comparison of predicted and observed pHs from selected lakes.
                          6.0
                                Observed vs. Predicted pH
7.0
    8.0
Predicted pH
9.0
10.0
The construct also was verified using time-series data from Lake Onondaga, NY (Figure 126).
The observed data were interpolated from the 2-m depth pH isopleths on a graph (Effler et al.
1996), introducing some uncertainty into the comparison.

     Figure 126. Comparison of predicted and observed pH values for Lake Onondaga, NY.
                              Data from (Effler et al. 1996).
                         Predicted pH, Lake Onondaga NY
                  Feb-  Apr-  May- Jul-89  Aug-  Oct-   Dec-
                   89   89    89        89   89   89
                                                         •AQUATOX
                                                         • Observed
                                                         •Poly. (Observed)
                                         186

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                            CHAPTER 5
                                                            Calcite Precipitation:  Simplifying
                                                            Assumptions
                                                             • Biogenic origin
                                                             • pH  of 7.5  is considered as  a
                                                              threshold for precipitation
                                                             • Dissolved  phosphate  sorbs  to
                                                              calcium carbonate but desorption is
                                                              not modeled
5.8 Modeling Calcium Carbonate Precipitation and Effects

Precipitation  of  calcium  carbonate  (mostly  calcite in
freshwater), with the potential for sorption and removal of
phosphorus,  is  modeled  as  an  extension  of the  pH
approach.  The prediction of pH in AQUATOX does not
extend past 8.25 because  the carbonate-bicarbonate system
becomes dominant.  We use a predicted  pH of 7.5 as a
threshold for  precipitation  of   calcium  carbonate  in
freshwater  ecosystems.   Almost  all  calcite  is  formed
biogenically,  primarily by plants using bicarbonate  as a source of carbon (McConnaughey et al.
1994). Even "whitings"  (sudden precipitation of fine-grained calcite) have been shown to be a
consequence  of cyanobacteria photosynthesis (Thompson et al. 1997).  Calcareous plants are
characterized by pH  polarization  with  acidic and alkaline  poles; calcification occurs  at the
alkaline pole (McConnaughey et al.  1994).  Proton generation  leads to formation of twice as
much CO2 than  is used in  the  process, providing  CO2 that is immediately taken  up for
photosynthesis.  As  a result, calcification and  photosynthesis use equivalent moles of C, as
shown by both theory and experiments (McConnaughey et al. 1994).  Three chemical reactions
represent this process:
         2+
       Ca+ + CO2 + H2O -» CaCO3 + 2H+
2H+
Ca
2+
             2HCO3'
             2HCO3-
                        2CO2 + 2H2O
                        CaCO3 + CO2 + H2O
Not all  plants can use bicarbonate.  However, it is difficult to generalize; mosses do not and
many chrysophytes (golden algae) do not.   Evidence suggests  that other  groups,  including
greens,  cyanobacteria, diatoms, and macrophytes, have species that do use bicarbonate and that
these will dominate in alkaline systems.

The algorithm simulates precipitation of calcite as being the molar equivalent to photosynthesis
of most plants and as occurring when the threshold pH of 7.5 is reached:
          If pH   >=  7.5  then   CalcitePcpt = C2Calcite
                                                          Photosynthesis
                                                                       PlantSubset
                                                                 C2OM
                                                                                   (217)
where:
      pH
      CalcitePcpt
      C2Calcite
      Photosynthesis
      PlantSubset
      C2OM
                          pH calculated by Eq. 204 or observed time series;
                          calcite precipitated (mg calcite/L • d);
                          stoichiometric constant for C and calcite (8.33, g calcite /g C);
                          rate of photosynthesis for a subset of plants (g/m3 • d);
                          all plants except Bryophytes and Other Algae;
                          stoichiometric constant for C and organic matter (1.9, g C/g OM).
Precipitated calcite is protected, in part, by sorbed organic material.  Therefore, it is assumed to
be insoluble—an assumption  also made in the sediment diagenesis model  (Di Toro 2001).
Because the settling rate is fast, it is also assumed that the calcite goes directly to the sediment.
                                           187

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 5


Phosphorus is adsorbed to the surface and coprecipitates with calcium carbonate (Wetzel 2001).
The rate of coprecipitation seems to be dependent on the rate of calcite precipitation (Otsuki and
Wetzel 1972).   However, the sorption is weak and can be reversed easily (Murphy et al. 1983).
Therefore, the default partition coefficient (300 L/kg) is based on equilibration experiments with
sediments from a marl lake (Van Rees et al. 1991).

                SorptionP  =  KDPCalcite • Phosphate • CalcitePcpt • \e - 6            (218)

where:
       SorptionP    =      rate of sorption of phosphorus to calcite (mgP/L •  d);
       KDPCalcite   =      partition coefficient for phosphorus to calcite (L/kg);
       Phosphate    =      concentration of phosphorus in water (mgP/L) (see (181));
       1 e-6         =      conversion factor (kg/mg).

Ironically, precipitation is impeded by phosphorus levels that are  too high. The threshold for
inhibition is about 30 mg-P/L (Neal 2001).  Furthermore, dissolved organic matter also can
inhibit precipitation, with  120  mg C/L  being  the threshold  (Neal 2001).  However, these
concentrations are so high that they are ignored in the model.
                                          188

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                 CHAPTER 6
                             6. INORGANIC SEDIMENTS

Inorganic sediments can have significant effects on light climate and inorganic sediment effects
on biota can also be explicitly modeled (see the section on suspended sediment effects starting
on page 107). Release 3 of AQUATOX contains four levels of inorganic sediment submodels:

   •   a very simple model based on a regression relationship between sediment deposition and
       total suspended sediments,  see  (122).   This approach should be used when the only
       inorganic sediment data available are TSS.   Add  the "TSS" state variable to use this
       option.

   •   a simple inorganic sediments submodel described in Section 6.1.  This model can be used
       to estimate the scour and deposition of inorganic sediments at a site as a function of water
       flows; therefore it is only applicable to streams and  rivers. This model requires additional
       data about the types of inorganic sediments (i.e., sand, silt, or clay) and their average rate
       of scour and deposition under different water-flow regimes.  This model may be selected
       under the sediment menu by choosing "Add Sand Silt Clay Model."

   •   a complex multiple-layer sediment submodel described in  Section 6.2.  This model can
       be used to estimate the  sequestration of organic toxicants within the deeper layers of the
       sediments and  the  potential for scour of such toxicants from these deep layers.  This
       submodel should be linked to  a hydrodynamic  model  to calculate the scour  and
       deposition of sediments in the modeled segment. This model may be selected under the
       sediment menu by choosing "Add Multi-Layer Sediment." Additional layers may also be
       added or removed using the options listed under the  sediment menu.

   •   a sediment  diagenesis  model described in  Section 7.  This model  provides a more
       sophisticated accounting of the  decay  of organic matter and remineralization in an
       anaerobic sediment bed and the effects on sediment oxygen demand.  The diagenesis
       model assumes a depositional environment; scour of sediments is not incorporated. This
       model  may be  selected under  the sediment menu by choosing  "Add  Sediment
       Diagenesis."
Within an AQUATOX simulation it is  also possible to ignore the effects of inorganic sediments
on ecosystem characteristics altogether, by including none of the models listed above. However,
the model will always track the remineralization of organic material within the  sediment bed and
the water column.
6.1 Sand Silt Clay Model

The original version was contributed by Rodolfo Camacho of
Abt Associates Inc. AQUATOX simulates scour, deposition
and transport of sediments and calculates the concentration of
sediments in the water column and sediment  bed  within a
river reach.  For running waters, the sediment is divided into
three categories according to the particle size:
   •   sand,  with  particle sizes  between 0.062  to  2.0
       millimeters (mm),
Sand, Silt, Clay: Simplifying
Assumptions

 • River reach is short and well-mixed
 • Channel is rectangular
 • Daily average flow regime
   determines scour, and deposition
 • Model for streams / rivers only
                                          189

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


   •   silt (0.004 to 0.062 mm), and
   •   clay (0.00024 to 0.004 mm).

Wash load (primarily clay and silt) is deposited or eroded within the channel reach depending on
the daily flow regime.   Sand transport is also computed within the channel reach.  The river
reach is assumed to be short and well mixed so that concentration  does not vary longitudinally.
Flow routing is not performed within the river reach.  The daily average flow regime determines
the amount  of scour, deposition  and transport of sediment.   Scour, deposition and transport
quantities are also limited by the amount of solids available in the  bed sediments and the water
column.

Within the bed, the mass of sediment in each of the three sediment size classes is a function of
the mass in the previous time step, and the mass of sediment in the overlying water column lost
through deposition, and gained through scour:

         MassBedSed = MassBedSed,t=-i + (DepositSed - ScourSed) • VolumeWater • TimeStep    (219)

where:
      MassBedsed        =     mass of sediment in channel bed (kg);
      MassBedsed, t = -i    =     mass of sediment in channel bed on previous day (kg);
      Depositsed         =     amount of suspended sediment deposited (kg/m d); see (230);
      Scour sed           =     amount of silt or clay resuspended (kg/m3 d); see (227);
       Volume water        =     volume of stream reach (m );  see (2); and
       TimeStep          =     derivative time-step (d).

The volumes of the respective sediment size classes are calculated as:
                                T, ,        MassBedsed                            0,m
                                Volume Sed	                            (220)
                                              RhOsed
where:
       Volumesed    =      volume of given sediment size class (m3);
       MassBedsed   =      mass of the given sediment size class (kg);
       Rhosed       =      density of given sediment size class (kg/m3);
       Rhosand              =     2600 (kg/m3); and
       Rftosat. ciay    =      2400 (kg/m3).

The porosity of the bed is calculated as the volume weighted average of the porosity of its
components:

                           BedPorosity = JLFracSed • Porosity Sed                      (221)
where:
       BedPorosity  =      porosity of the bed (fraction);
       Fracsed       =      fraction of the bed that is composed of given sediment class; and
       Porositysed    =      porosity of given sediment class (fraction).
                                          190

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6
The total volume of the bed is calculated as:
                     D ^T/ i      Volume sand + Volume sat+ Volumeaay                ,„.,
                     BedVolume =	                (222)
                                          1- BedPorosity
where:
      BedVolume    =     Volume of the bed (m3).

The depth of the bed is calculated as

                        r, ,T^   ,           BedVolume                           ,--~
                       BedDepth =	                   (223)
                                  ChannelLength • ChannelWidth

where:
      BedDepth           =     depth of the sediment bed (m);
      ChannelLength      =     length of the channel (m); and
      ChannelWidth       =     width of the channel (m).

The concentrations of silt and clay suspended in the water column are computed similarly to the
mass  of those sediments in the bed, with the addition  of loadings from upstream  and losses
downstream:

                      KgLoad- ,
              ConcSed = _  0£4nn + Cone Sed,t=-i + ScourSed - DepositSed - WashSed         (224)
                       (J • oo4UU
where:
      Concsed       =    concentration of silt or clay in water column (kg/m3);
      Concsed, t = -i    =    concentration of silt or clay on previous day (kg/m3);
      KgLoadsed     =    loading of clay or silt (kg/d);
      Q             =    flow rate (m3/s);
      86400         =    conversion from m3/s to m3/d;
      Scour Sed       =    amount of silt or clay resuspended (kg/m3); see (227);
      Deposited     =    amount of suspended sediment deposited (kg/m3); see (230); and
      Washsed       =    amount  of sediment  lost through downstream transport  (kg/m3);
                          see (231).

The concentration of sand is computed using a totally different approach, which is described in
the section on Sand below.
Deposition and Scour of Silt and Clay

Relationships for scour and deposition of  cohesive  sediments  (silts  and clays)  used in
AQUATOX are the  same as the ones used by the Hydrologic Simulation Program in Fortran
(HSPF, U.S. Environmental Protection Agency, 1991).  Deposition and scour of silts  and clay
are modeled using the relationships for deposition (Krone, 1962) and scour (Partheniades, 1965)
as summarized by Partheniades (1971).

                                         191

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6



Shear stress is computed as (Bicknell et al., 1992):

                           Tau = H2ODensity • Slope • HRadius                      (225)
where:
       Tau          =      shear stress (kg/m2);
       H2ODensity  =      density of water (1000 kg/m3);
       Slope        =      slope of channel (m/m);

and hydraulic radius (HRadius) is (Colby and Mclntire, 1978):

                                 ,,„ ,.      Y-Width                              .„,.
                                 HRadius =	                            (226)
                                          2 • Y + Width

where:
       HRadius      =      hydraulic radius (m);
       Y            =      average depth over reach (m); and
       Width        =      channel width (m).
Resuspension or scour of bed sediments is predicted to occur when the computed shear stress is
greater than the critical shear stress for scour:

                                if Tau > TauScoursed then
                                  Erodibility, .  (    Tau     ^                   (227)
                        Scour Sed = - ±m •  -- 1
                                       Y
where:
       Scour sed      =      resuspension of silt or clay (kg/m3 d);
       Erodibilitysed =      erodibility coefficient (0.244 kg/m2 d); and
       TauScoursed  =      critical shear stress for scour of silt or clay (kg/m ).

The amount of sediment that is resuspended is constrained by the mass of sediments stored in the
bed.  An intermediate variable representing the maximum potential mass that can be scoured is
calculated; if the mass available is less than the potential, then scour is set to the lower amount:
                             Check sed ~ Scour sed ' Volumewater                        (228)

                              if Mass sed  ^  Check sed then
                                             Masssed                             (229)
                                 Scourged
where:
       Checksed      =      maximum potential mass (kg); and
                                          192

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


       Masssed      =      mass of silt or clay in bed (kg).

Deposition occurs when the computed shear stress is less than the critical depositional shear
stress:
                     Deposit5eA = ConcSi
if Tau < TauDepSed then

               Sed'^*-"* ^f\\
                                                  (230)
                                         f    -VT&j-SecPerDay
                                                          TauDePsed
where:
       Depositsed    =      amount of sediment deposited (kg/m3 day);
       TauDepsed    =      critical depositional shear stress (kg/m2);
       Concsed      =      concentration of suspended silt or clay (kg/m3);
                           terminal fall velocity of given sediment type (m/s); and
       SecPerDay   =      86400 (seconds / day).

The terminal fall velocity is specified in the site's underlying data.

Downstream transport is an important mechanism for loss of suspended sediment from a given
stream reach:

                                         Disch- ConcSed
                                      d -- - — — -
                                           Segvolume
where:
       Washsed      =      amount  of given sediment lost to downstream transport  (kg/m3
       day);
       Disch        =      discharge of water from the segment (m /day);
       Concsed      =      concentration of suspended sediment (kg/m );
       SegVolume   =      volume of segment (m3).

When  the  inorganic sediment model is included  in an AQUATOX  stream simulation,  the
deposition  and erosion of detritus mimics the deposition and erosion of  silt.  The fraction of
detritus that is being scoured or deposited is assumed to equal  the fraction of silt that is being
scoured or  deposited. The following equations are used to calculate the scour and deposition of
detritus:


                   Frac Scour Detntus = Frac Scour Sllt = Scour Sllt • Volumesat              (232)
                                                            Mass snt

                    Scour Detntus = Frac Scour Detritus • ConcAiisedDetntus • 1000               (233)
where:
       FracScour      =     fraction of scour per day (fraction/day);
       Scour sat        =     amount of silt scoured (kg/m3 day) see (227);
       Volumesnt      =     volume of silt initially in the bed (m3);

                                          193

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


       Masssut        =     mass of silt initially in the bed (kg);
       ConcAiisedDetntus  =     all sedimented detritus (labile and refractory) in the stream bed
                            (kg/m3);
       ScourDetritus     =     amount of detritus scoured (g/m3 day); and
       1000           =     conversion of kg to g.

The equations for deposition of detritus are similar:
               7-7^     •,•         7-7^    •,•       Depositions, -1000
               Frac Deposition Detritus = Frac Deposition snt = —	—	          (234)
                                                             Cone sat

                    DepositionDetntus = Frac DepositionDetntus • ConcSuspDetntuS               (235)
where:

       Depositionsnt        =      amount of silt deposited (kg/m3 day) see (230);
       Cone sat             =      amount of silt initially in the water (g/m3);
       FracDeposition      =      fraction of deposition per day (firac / day); and
       Con
       and
Cone suspDetritus       =      amount of suspended detritus initially in the water (g/m3);
                                                                3
       Deposition Detritus     =      amount of detritus deposited (g/m  day).
Scour, Deposition and Transport of Sand

Scour, deposition and transport of sand are simulated using the Engelund and Hansen (1967)
sediment transport relationships as presented by Brownlie (1981).  This relationship was selected
because of its simplicity and accuracy.  Brownlie (1981) shows that this relationship gives good
results when compared to 13 others using a field and laboratory data set of about 7,000 records.

                      . -,     Rho             Velocity • Slope         r- — - —    ,~.-.^
         PotConcSand = 0.05 -- - --- .          ,                 ' ^TauStar    (236)
                           Rho Sand -Rho   RhoSand-Rho
                                                       ' 8 ' Dsand fl UUU
                                                ,
                                              Rho
where:
       PotConcsand  =      potential concentration of suspended sand (kg/m3);
       Rho          =      density of water (1000 kg/m3)
       Rhosand             =      density of sand (2650 kg/m3);
       Velocity      =      flow velocity (converted to m/s);
       Slope        =      slope of stream (m/m);
       Dsand        =      mean diameter of sand particle (0.30 mm converted to m); and
       TauStar      =      dimensionless shear stress.

The dimensionless shear stress is calculated by:
                       rr  o                   rrn  7-                                „,.,„<..
                       TauStar = -- HRadius -- - -                  (237)
                                                       Dsand/ 1000
                                           194

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 6


where:
       HRadius     =      hydraulic radius (m).

Once the potential  concentration has been determined for the given flow rate and  channel
characteristics, it is  compared with the present concentration.   If the potential concentration is
greater, the difference is considered to be  made available through scour, up to the limit of the
bed. If the potential concentration is less than what is in suspension, the difference is considered
to be deposited:

                           Check Sand = PotConcsand • Volumewater                       (238)

                           MassSuspSand = ConcSand • VolumeWater                       (239)

                        TotalMassSand = MassSuspSand + MassBedSand                    (240)

                             if CheckSmd ^ MassSuspSand then


                           DeP°SitSand = MaSSSuSPSand ' Checked                       (241)
                                  ConCSand = PotConCSand

                             if Checksand ^ TotalMasssand then


                                     MassBedsand = 0                                (242)
                                 „       _ TotalMasssand
                                 COnCSand --
                    if Checksand > MassSuspSand and < TotalMasssand then

                            Scoursand = Checksand - MassSuspSand                       (243)
                                    _ MassSuspSand + Scoursand
                            COnCSand --
                                          195

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6
Suspended Inorganic Sediments in Standing Water

At present, AQUATOX does not compute settling of inorganic sediments in standing water or
scour as a function of wave action. However, suspended sediments are important in creating
turbidity and limiting light, especially in reservoirs and shallow lakes. Therefore, the user can
provide loadings of total suspended solids (TSS), and the model will back-calculate suspended
inorganic sediment concentrations by subtracting the simulated phytoplankton and suspended
detritus concentrations:

                          InorgSed = TSS-Z Phyto - Z PartDetr                      (244)
where:
       InorgSed     =      concentration of suspended inorganic sediments (g/m3);
       TSS          =      observed concentration of total suspended solids (g/m3);
       Phyto        =      predicted phytoplankton concentrations (g/m3); and
       PartDetr     =      predicted suspended detritus concentrations (g/m3).

A radio button on the TSS loadings screen is used to specify whether user-input TSS loadings are
"total suspended (inorganic) sediments" or "total suspended solids."  If "inorganic sediments"
are  specified then equation (244) is not required as the TSS loading is not assumed to include
phytoplankton or organic matter.
                                          196

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                               CHAPTER 6
                                                          Multi-Layer   Sediment
                                                          Simplifying Assumptions
                                  Model:
6.2 Multi-Layer Sediment Model

As an  alternative to  the simple  sand-silt-clay  model
described above (section 6.1), AQUATOX also includes a
complex multiple layer  sediment transport model.   This
model can simulate up to  ten bottom layers of sediment.
Within  each  sediment layer, the  state  variables  consist of
inorganic  solids, pore waters,  labile   and  refractory
dissolved organic matter in pore waters, and sedimented
detritus.  Nutrient  concentrations  are not modeled in the
pore  waters  of the  sediment layers,  although  dissolved
organic matter is. Each  of these  state variables can also have up to twenty organic toxicant
concentrations associated with it.  The AQUATOX sediment transport component is summarized
in Figure 127.
         Figure 127: Components of the AQUATOX sediment transport model and units.
                                                             Top  layer is "active layer"  that
                                                             interacts with the water column
                                                             Individual sediment layers are well-
                                                             mixed
                                                             Density of each sediment layer
                                                             remains constant
                                                             Hardpan  barrier  assumed at  the
                                                             bottom of the svstem
     Susp. Inorg.
    Solids mg/L
DOM in Water
 Col. mg/Lwc
       Inorganic
       Solids  g/m
 DOM in Pore
 Water mg/L
                          Pore Water
                             m3/m2
                                          Sed Detr mg/L
       Inorganic  -5=
                          Pore Water
 DOM in Pore
 Water mg/Lpw
                                                                    Buried Detr g/m
       Inorganic
      Solids  g/m
                      Buried Detr g/m2
                                              DOM in Pore
                                              Water mg/L
Pore Water
   m3/m2
The AQUATOX sediment submodel was designed to be nearly identical in concept to IPX (In-
Place Pollutant eXport)  version 2.7.4 (Velleux et. al 2000).  Erosion  and deposition  cause
changes in the mass of sediments in the top or "active" layer.  When the active layer becomes too
large or too small, a conveyor-belt action takes place moving all of the layers up or down intact
("pez dispenser" action).  Because all layers are assumed well-mixed, moving partial layers up
and down and then recalculating concentrations within sediment layers would result in too much
mixing throughout the sediment layers (and advection of pollutants from the bottom layer to the
top).  During  development, the AQUATOX sediment submodel  was closely tested against the
IPX model and precisely reproduced results from that model.
                                           197

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6
Within AQUATOX, inorganic sediments in layered sediments are represented as three distinct
state variables:  cohesives  (clay),  non-cohesives  (silt),  and  non-cohesives2 (sand).   These
correspond to the variables described in Section 6.1.

For each inorganic compartment, the sediment transport model accepts daily input parameters for
interactions between the top sediment layer and the water column.  These interactions are input
as daily scour and daily deposition for each inorganic sediment type in units of grams per day.
The  model also requires deposition and erosion velocities for cohesive inorganic  sediments.
These inputs  are then used to calculate the deposition and erosion of organic matter within the
system.

AQUATOX assumes that the density of each sediment layer will  remain constant throughout a
simulation. Because of this, the volume and thickness of the top bed layer will vary in response
to deposition  and erosion. Additionally, the surface area of the multi-layer sediment bed is set to
remain constant.  Even if the sediment  surface at a site grows or shrinks due to  water volume
changes, this model tracks sediments under the initial-condition surface area.

When the top layer has reached a maximum thickness, it is broken  into two layers.  Other layers
in the system are moved down one layer without disturbing their concentrations or thicknesses.
This allows the model to maintain a toxicant concentration gradient within the sediment layers
during depositional regimes.  Similarly, when the top layer has eroded to a minimum size, the
layer beneath it is joined with the active layer to form a new top layer.  In this case, lower layers
are moved up one level, without changing their concentrations, densities, or thicknesses.  More
details about these processes can be found in section on sediment layer interactions below.

At the bottom  of the system, a hardpan barrier is assumed.  The model, therefore, has no
interaction beneath its lowest layer.  If enough erosion takes place so that this hardpan barrier is
exposed, no further erosion will  be possible. Deposition  can, however, rebuild the sediment
layer system.  This hardpan bottom prevents the  artificial inclusion  of "clean"  sediment and
organic  matter into the model's simulation during erosional events.  Because it is a barrier and
not a boundary, it prevents loss of toxicant to the system under depositional regimes.

AQUATOX writes output data for a fixed number of sediment layers.  When, due to deposition,
a layer  is buried below the fixed  number of sediment layers, AQUATOX  keeps track of that
layer, but does not write daily output. That deep layer is stored in memory and state variables in
that layer have  the potential to move back into the system  later due to erosion.  When,  due to
erosion, there are fewer than the fixed number of sediment  layers, AQUATOX writes zeros for
all layers below the hardpan barrier.

Pore water moves up and down  through the sediment system when layers move upward and
downward in  the system.  Substances dissolved in pore water also move through the system as a
result of diffusion.
                                          198

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6
Suspended Inorganic Sediments

As mentioned above, inorganic sediments are broken into three sets of state variables based on
particle size. Each of these three inorganic sediment types are found in the water column as well
as in each modeled sediment layer.

For inorganic sediments suspended in the water column, the derivative looks as follows:

            dSuspSediment   T    ,.    0       ^     . .    Tir  ,      Tir  , .         ,~*-^
            	= Loading + Scour - Deposition - Washout + Washin        (245)
                  dt

where:

       dSuspSediment/dt     =     change in concentration of suspended sediment (g/m3-d);
       Loading             =     inflow loadings (excluding upstream segments) (g/m3-d);
       Scour               =     scour from the active sediment layer (g/m3-d);
       Deposition           =     deposition to the active sediment layer (g/m3 • d);
       Washout             =     loss due to being carried downstream (g/m3-d), see (16);
       Washin              =     loadings from upstream segments (g/m -d), see (30);
There are two options for specifying deposition to and scour from the active layer when using the
multi-layer sediment option.  Deposition and scour can be simulated by a hydrodynamic model
and imported into AQUATOX.  In this case, for each of the three categories  of suspended
sediment, deposition to and scour from the active layer are input to AQUATOX as a daily time
series in units of g/d.  These inputs are converted into units of g/m3-d by dividing by the volume
of the segment.

Alternatively, based on user specification, the model can calculate deposition and scour using the
sand-silt-clay model specifications, see (230),  (227).  In the "Edit Sediment Layer Data" dialog,
where cohesives or non-cohesives are being input there is a checkbox that states "use sand-silt-
clay model" to toggle between these two options.

Unlike the  simple sediment model, suspended sediments can  sorb organic  toxicants when the
multi-layer sediment model is run.  More specifications about sorption of organic chemicals to
inorganic sediments can be found in Section 8.10 of this document.

Inorganics in the Sediment Bed

Inorganic sediments are found in each sediment layer that is modeled. The derivative, however
is relevant only for the active (top) layer.

                dBottomSediment   ^    .  .    0       n  „   ,  n  „              ,~.^
                	= Deposition - Scour + Bedload - Bedloss           (246)
                       dt
                                         199

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 6


where:
     dBottomSediment/dt  =   change in concentration of sediment in this bed layer (g/m2-d);
     Scour             =   movement to the water column (g/m2-d);
     Deposition         =   deposition from the water column (g/m -d);
     Bedload           =   bedload from all upstream  segments (g/m -d).  Only relevant for
                           the active layer of sediment, see (247);
     Bedloss            =   loss due to bedload to all downstream segments (g/m2-d).  Only
                           relevant for the active layer of sediment, see (248).

Deposition and scour are input into the model in units of g/d.  These inputs are divided by the
area of the system to get units of g/m2-d.

Bed load is input as a loading in g/d for each link between two segments, if multiple segments
are being modeled.  This process is only relevant for the top layer of sediment modeled.  The
total bed load for a particular segment can be calculated by summing the  loadings over all
incoming links.
                          D ^T   ^   ^           UPstreamlink
                          BedLoad = > - - — - -                     (247)
                                              AvgArea

where:

      BedLoad           =      total bedload from all upstream segments (g/m2- d);
      BedLoadupstreamhnk   =      bedload over one of the upstream links (g/d);
      AvgArea            =      average area of the segment (m );

Similarly, total bed loss is the sum of the loadings over all outgoing links:
                           n  IT       \^     -Upstreamlmk
                          BedLoss = >  - - - - -                      (248)
                                             AvgArea

                                                                         r\
      BedLoss            =      total bedloss to all downstream segments (g/m -d);
      BedLoss Downstreamiink  =      bedload over one of the downstream links (g/d);
      AvgArea            =      average area of the segment (m2);
As mentioned above, the derivative presented is relevant only for the active layer.  Inorganic
sediments below the active layer  do move up and down through the  system  as  a result of
exposure or deposition.  However, these sediments move as a part of their entire intact layer
when the active layer has reached its maximum or minimum level.

When the top layer reaches a minimum thickness, the layer below the active layer is added to the
active layer to form one new layer.  The inorganic sediments within these two layers  do undergo
mixing during this process.


                                         200

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6
Detritus in the Sediment Bed

State variables tracking sedimented labile and refractory detritus are also included in each layer
of sediment that is simulated.  The equations for sedimented detritus in the active layer are the
same as those for "classic" AQUATOX.

Like inorganic sediments, buried detritus below the active layer only moves up and down in the
system when its layer moves up and down intact.  Therefore,  detritus found below the active
layer has a very simple derivative:

                             dBuriedDetritus    ^                               ,-. in^
                             	= -Decomp                          (249)

where:
                                                                             r\
       dBuriedDetritus/dt   =     change in concentration of sediment on bottom (g/m -d);
       Decomp             =     microbial decomposition in (g/m2-d) see (159).
Pore Waters in the Sediment Bed

Pore water quantities are also tracked in the sediment bed.  The derivative for pore waters is
quite straightforward:

                             dPoreWater      .
                             	—	= Gamup - LossUp                         (250)

where:

      dPoreWater/dt  =     change in volume of pore  water in the sediment bed normalized
                           per unit area  (m3/m2 • d);
      GainUp        =     gain of pore water from the water column above (m3/m2 • d);
      LossUp         =     loss of pore water to the water column above (m3/m2 -d);
In the active layer, pore waters are assumed to move into the water column when scour occurs.
To keep the bed density constant, the loss of pore waters can be solved as follows:

                             (Erode n , Density,, , ) - (Erode ~, / BedDensity)
          LossTr =                         -                            -
               UP
                                                       -  \Q~6

where:
      Lossup              =     loss of pore water to the water column above (cm3/- d);
      Erode sed            =     scour of this sediment to the water column above, (g/d);
      Density sed           =     density of this sediment (g/m3);
      BedDensity          =     density of the active layer (g/m3);

                                        201

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


       le-6                 =     one over the density of water (m3/g);

Pore waters  are taken from  the water column when deposition occurs.   Keeping the density
constant, the gain of pore waters can be solved as follows:
            .    =         (Deposit SedDensitySed ) - (Deposit Sed I BedDensity}
           0m  ~
                                                        ~   \Q-6
where:
                                                                                3
                           =     gain of pore water from the water column above (cm/- d);
       Deposit sed           =     deposit of this sediment from the water column, (g/d);
       Density sed           =     density of this sediment (g/m3);
       BedDensity          =     density of the active layer (g/m3);
       le-6                 =     one over the density of water (m3/g);
When the active layer becomes too large it becomes split into two layers.  During this split, the
new second layer is assumed  compressed to the density of the old  second layer.   This
compression results in squeezing of pore water out into the water column. Details of this process
can be found in the section on sediment layer interactions, below.
Dissolved Organic Matter within Pore Waters

Another state variable tracked within the sediment bed is dissolved organic matter within pore
waters.  Dissolved labile and refractory detritus within pore waters are tracked as separate state
variables.   Like other dissolved  detritus, these variables use units  of mg/L.  However, it is
important to note that these are liters of pore water and not liters in the water column.


          dD°M"er = GainDOMUp - LossDOMUp ± DiffDown ± DiffUp - Decomp     (253)


where:
                   t =   change in concentration of DOM in pore water in the sediment bed
                        normalized per unit area (nig/L^ -d);
    GainDOMup    =    active layer only: gain of DOM due to pore water gain from the water
                        column (mg/L^-d);
    LossDOMup    =    active layer only: loss of DOM due to pore water loss to the water
                        column (mg/Lpn,-d);
    Diffup, DiffDown  =    diffusion over upper or lower boundary (mg/Lpw- d), see (256);
    Decomp        =    microbial decomposition in (mg/Lpw-d), see (159).
The increase of DOM due to pore water gain from the water column is simply the volume of


                                          202

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 6
water that is moving from the water column above multiplied by the DOM concentration in the
above sediment layer.  However, the concentration then needs to be normalized for the volume
of pore water in the current segment:
                 GainDOMu = ConcDOM . • GainPWu \
                           Up             "~l          uPore Water Vol
                                                            (254)
where:
      GainDOMup
(mg/LpW-d);
      ConcDOM „_!
      GainPW,
              up
      AvgArea
      Ie3
      PoreWaterVol
=  gain  of DOM  due to  pore  water gain  from the  layer  above

=  concentration of DOM in above layer (mg/LUpper water);
=  gain of pore water from above (m3UpperWater/in -d);
=  average area of the segment (m );
=  units conversion (L/m3);
=  pore water volume (L);
The loss of DOM in pore water to the water column is a simpler equation due to the fact that
there are no units conversions necessary:
                          (   LossPWu
  LossDOM  = ConcDOM \	^	
                          I PoreWaterConc
                                                                                 (255)
where:
      LossDOMup
      ConcDOM „
      LossPW,
             up
      PoreWaterConc  =
    loss of DOM in pore water to the layer above (mg/Lpw-d);
    concentration of DOM in this layer (mg/LpW);
    loss of pore water to above layer (m3pw/m -d);
    pore water concentration (m3pw/m2);
Because diffusion and decomposition of DOM in pore water occur throughout the system, not
just the active layer, the above derivative is relevant for the whole system.  DOM in pore water
also  moves  up  and down through a system when its layer  moves intact due  to erosion or
deposition.
Diffusion within Pore Waters

AQUATOX calculates the diffusion of dissolved organic matter within pore waters in the
sediment layers. This calculation requires that porosity be included in the diffusion equation:
                                         203

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


             T^-JX  •       DiffCoeff • Area • AvgPor \   	up      ^,**,,n   ,        ,~£^
             Diffusion^ = ——	—	 	-1	dmm_         (256)
                           CharLength • AvgPor  \ Porosityup   Porosity

where:
      DiffusionUp   =    gain of DOM due to diffusive transport over the upper boundary of the
                        sediment layer, (g/d);
      DiffCoeff    =    dispersion coefficient, (m  /d);
      Area         =    interfacial area of the upper boundary of the sediment layer (m );
      AvgPor      =    average porosity of the two layers. If the boundary is a sediment/water
                        boundary, AvgPor is the porosity of the sediment, (fraction);
      CharLength  =    characteristic mixing length, see text below, (m);
      Cone Layer     =    concentration of the relevant segment, (g/m3);
      Porosity Layer  =    porosity of the relevant layer (fraction).
For the characteristic mixing length, AQUATOX uses the distance between two benthic segment
midpoints.  For pore water exchange with a surface water segment, the characteristic mixing
length is taken to be the depth of the surficial benthic segment

Equation (256) is also used to calculate the diffusion of toxicants within pore waters.  In this
case,  the units of DiffusionUp  are mg/d  rather than g/d and the concentrations of toxicants
within the layers are in units of ug/L rather than mg/L.

Sediment Interactions

The mass of the top sediment layer increases and decreases as a result of deposition and scour.
Because the density of this layer remains constant, the volume and thickness of the top sediment
layer  also increases and decreases.  When the thickness of the top sediment layer reaches its
maximum, as defined by the user, the upper bed is split horizontally into two layers. The top of
these two layers maintains the same density it had before the layer was split up. It is assigned the
initial condition depth of the active layer.

The lower level is assumed to be compressed to the same density as the level below it.  This
compression results in pore water being  squeezed into the water column. The volume that is lost
as a result of this compression can be solved as follows:
              T7 ,    T  .            reompress           ower         7i    ess
              VolumeLost = - - - - -         (257)
                                          \e6- Density Lower
where:
       VolumeLost          =     volume of active layer lost due to compaction (m3);
                                 mass of the new second layer before compression (g);
                                 volume of the new second layer before compression (m3);
       Density lower          =     density of the layer below the active layer (g/m3);
       Ie6                 =     density of water (g/m3)
                                          204

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


The above equation also provides the quantity of pore water squeezed into the water column
because the compression of the active layer is entirely the result of pore water being squeezed
out.  Toxicants, dissolved organic matter, and toxicants associated with dissolved organic matter
in the pore water also move into the water column as a result of this compression.  If there is
only one layer in the system when the splitting of the active layer takes place, Densitylower is
assumed to be the initial condition density of the second layer in the system.

The volume of a sediment layer is defined as follows:

                                          V SedMass
                                 BedVoln = ^	                            (258)
                                          BedDensity
where:
       BedVoln            =     volume of bed at layer n (m3);
       SedMass            =     mass of sediment type (g);
       BedDensity         =     density of bed (g/m );
The porosity of a sediment layer is defined as:
                                         _      [  \ OTIC    \
                         FmcWatern = 1 - V             Sed                       (259)
where:
       FracWatern                =      porosity of the sediment layer (fraction);
       Cone sed             =     concentration of the sediment (g/m3);
       Sedtypes            =     all organic and inorganic sediments
       Density sed           =     density of the sediment (g/m3);
When the thickness of the top sediment layer reaches a minimum, as defined by the user, the two
top layers combine into one new active layer.  The density of this new active layer  is the
weighted average of the densities of the combined layers.

           ,r   „  ,„   .„   Volume'    -Density>     + Volume    -Density
           NewBedDensity =	—      (260)
                                        Volume Lcyer2 + Volume Layerl

where:
       NewBedDensity     =      density of new joined bed (g/m3);
       VolumeLayerN        =      volume of layer that was initially layer 1 or 2 (m3)
       DensityLayerN        =      density of layer that was initially layer 1 or 2 (g/m3);

The height of the new layer is the sum of the heights of the two layers being joined.
                                         205

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 6


The bottom  of the system is composed of a hardpan barrier.  When this bottom is exposed, no
further erosion can take place. When deposition occurs on this hardpan bottom, it is rebuilt with
the density of the layer that existed previously.  If enough deposition occurs so that two layers
are created, the new second layer is compressed to the density of the original second layer.

If a system starts with exposed hardpan as an initial condition,  the user must still specify the
density of the top layer so that AQUATOX knows what density to create the top layer with.  If
the user specifies a density for the second layer, this will be used when enough deposition occurs
so that two layers are created.
                                         206

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                  CHAPTER 7
                              7. SEDIMENT DIAGENESIS

AQUATOX has been modified to include a representation of
the sediment bed as presented in  Di Toro's Sediment Flux
Modeling (2001).  This optional sediment submodel tracks the
effects  of  organic  matter  decomposition  on  pore-water
nutrients,  and  predicts  the flux of nutrients from  the pore
waters  to  the  overlying  water   column  based   on  this
decomposition. It is a more realistic representation of nutrient
fluxes than the  "classic" AQUATOX model. It includes silica,
which will be  modeled  as  a  nutrient for diatoms in a later
version.

The model assumes a small aerobic layer (LI) above a larger
anaerobic layer (L2). For this reason, it is best to apply this
optional   submodel  in   eutrophic  sites  where  anaerobic
sediments are prevalent.

Because   AQUATOX    simulates   organic   matter   with
stoichiometric  ratios for nutrients  and  Di   Toro's model
simulates  separate  organic  nutrients,  the  organic-nutrient
relationships are redefined for the sediments. The additional 21
sediment diagenesis model is enabled (and one driving variable)
  Sediment Diagenesis Model:
  Simplifying Assumptions

   • Model assumes a depositional
    environment (no scour is modeled).
   • Two layers of sediment are
    modeled.
   • Aerobic (top) layer is quite thin
   • Model is best suited to represent
    predominantly anaerobic
    sediments.
   • Deposition of particulate organic
    matter moves directly into Layer 2.
    Particulate organic matter in Layer
    1 assumed to be negligible and is
    not modeled
   • The fraction of POP and PON
    within defecated or sedimented
    matter is assumed equal to the ratio
    of phosphate or nitrate to organic
    matter for given species.
   • All methane is oxidized or lost.
 state variables added when the
are as follows:
       POC (Particulate Organic Carbon) in sediment: three state variables to represent three
       reactivity classes (see below).  A component of the particulate organic matter (POM) that
       settles from the water column into the anaerobic layer (Layer 2) and decomposes.
       PON (Particulate Organic Nitrate) in sediment:  as with POC,  three state variables to
       represent three reactivity classes in the anaerobic layer.  Another component of POM.
       POP (Particulate Organic Phosphate) in sediment: as with POC, three state variables to
       represent three reactivity classes in the anaerobic layer. The third modeled component of
       POM.
       Ammonia: two  state variables to represent two layers.  Formed by the decomposition of
       PON, this process is  also called the diagenesis flux.  Ammonia in sediment undergoes
       nitrification and flux to or  from the water column.
       Nitrate:  two state variables (in Layers 1 and 2). Formed by nitrification of ammonia in
       the sediment bed. Undergoes denitrification and flux to or from the water column.
       Orthophosphate: two state variables (in Layers 1 and 2). Formed by the decomposition
       of POP in sediment (diagenesis flux).  Flux to or from the water column is predicted but
       may be limited by strong P sorption to oxidated ferrous iron in the aerobic layer.
       Methane: (Layer 2) Methane is formed due to the decomposition of POC in the sediment
       bed under low-salinity conditions.  Methane undergoes oxidation resulting in increased
       sediment oxygen demand.
       Sulfide:  two state variables (in Layers  1  and 2).  Hydrogen sulfide (H2S)  is formed,
       rather than methane under saline conditions.  Sulfide in sediment may undergo burial,
       flux to the water column, or oxidation (increasing SOD).
                                           207

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                           CHAPTER 7
       Biogenic Silica:   Silica in  sediment is modeled using three state  variables.  Silica
       deposited from the water column is bioavailable or "biogenic silica" and is modeled in
       Layer 2.  Biogenic silica can then either undergo deep burial or dissolution to dissolved
       silica.
       Dissolved Silica:  two state variables (in Layers 1 and 2). Produced when biogenic silica
       breaks down due to dissolution. Available  Silica in Layer 2 and Silica in Layers 1 & 2.
       Dissolved silica may undergo burial or flux to the water column.
       COD: Driving variable for chemical oxygen demand in the water column that affects the
       flux of sulfide to the water column.

         Figure 128: Simplified schematic of the AQUATOX sediment diagenesis model
                      (Diagram does not include Silica, Sulfide or COD)
  Water Column
      Organic Matter
        Flux to
        Water
        fn(Oxygen)
             Flux to
             Water
Flux to
Water
  Aerobic
                              Phosphate
                  Ammonia
                                                             Nitrification
                                                              SOD
                                                                              Denitri-
                                                                              fication
  Anaerobic
                                     Mineralization
                           Oxidation
                    CH4  kSOD
                                 Mineralization
                         POP
Mineralization
  	»
Phosphate
  Deeply Buried
                                                           r     .
Particulate organic matter in the sediment bed (POC,  PON,  and POP) is divided into three
reactivity classes as follows:

   •   GI - reactivity class 1, equivalent to labile organic matter
   •   G2 - reactivity class 2, equivalent to refractory organic matter
   •   G?, - reactivity class 3, nonreactive

Within the system of equations governing these state variables,  sediment oxygen demand (SOD)
is  a  function of specific  chemical  reactions following the decomposition of organic matter.
                                         208

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7


Specifically the oxidation of methane or sulfide and the nitrification of ammonia increases the
predicted SOD .  This in turn has effects on the amount of oxygen present in the water column.
The amount of oxygen  in the water column, however significantly affects the nitrification of
ammonia (275).

To optimize the solution of this feedback loop, an iterative solution is utilized to calculate SOD
in each time-step, (see Eq 263) An initial  value of SOD (SODinmai) is estimated.  (In the first
time-step, SODjnitiai is calculated by the model based on sediment initial conditions, in later time-
steps the SODinitiai is assumed to equal the SOD in the previous time-step.) Based on SODinitiai,
the concentrations of ammonia, nitrate, and sulfide or methane can be calculated by the model
Then, using those nutrient concentrations, a new estimate of  SOD may be obtained.   This
becomes the new "initial" estimate of SOD until the initial estimate and "new" estimate of SOD
converge (to within the relative error set in the AQUATOX setup  screen).

This iterative solution is likely not mandatory within AQUATOX as the water column model is
not decoupled  from  the sediment  diagenesis model (all  differential  equations  are  solved
simultaneously.)  However, by including this iterative solution, the solution  for SOD is not a
limiting factor when setting the variable differentiation time-step.

Most implementations of Di Toro's model solve state variables  in the thin aerobic upper layer
(Layer 1) using an assumption of steady-state. This  option was added to AQUATOX Release
3.1. A checkbox at the top of the diagenesis initial conditions screen can be selected for running
the model in this manner. Initial tests of the steady-state model produce results nearly identical
to non-steady-state model results and the model runs up to ten times faster.

However, precise balancing of the mass of nutrients is not generally possible when the steady-
state model is incorporated.  If there are two interacting state variables and one is solved with a
steady-state solution and the other  is solved using differential  equations, the conservation of
mass is not possible.  (For example, when solved under steady state, the nutrient mass in Layer 1
will change based on the conditions  prior to the time-step but that nutrient mass is not explicitly
added to or subtracted from another state variable.)

It would be advisable to simulate a site with steady state turned on during model calibration and
off for production runs if balancing the mass of nutrients is important.  When the steady-state
model  is not utilized, the  state variables in sediment  Layer 1 are solved  using  differential
equations.  The thickness of Layer  1 (a user input variable) might therefore have a significant
effect on model run time, with larger layer thicknesses resulting in shorter run-times.

7.1 Sediment Fluxes

State variables in the two model layers are subject to a number of fluxes to and from other
modeled and unmodeled compartments. Fluxes in the  model include:

   •   Diffusion of the dissolved component of state variables to  and from the water column;
   •   Diffusion of the dissolved component of the state variables between layers;
   •   Burial of the state variables below the lower layer and out  of the modeled system; and

                                          209

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                            CHAPTER 7
   •   Particulate mixing of the two layers and resultant exchange of state variable.

To calculate these fluxes, the diffusion velocity between layers must be solved as well  as a
particle mixing velocity between the two layers and a surface mass transfer coefficient.

Diffusion Velocity Between Layers

Diffusion between layers is  specified by a diffusion coefficient, provided by the user and
adjusted for the water temperature in the system. Enhanced diffusive mixing due to bioturbation
is  not currently  included in the AQUATOX  implementation,  though  direct mixing by
bioturbation is.
                                        j-\   r\ Temp -20
                                  KL=^-^	
                                             //,
                                                                                   (261)
      KL
      Temp
                          diffusion velocity between layers (m/d);
                          diffusion coefficient for pore water (m /d);
                          constant for temperature adjustment for Dd (unitless);
                          temperature of water (deg. C); and
                          depth of sediment layer 2 (m).
Particle Mixing Between Layers (Bioturbation)

In a departure from Di Toro's model, particle mixing between layers is a direct function of the
modeled benthic biomass in the system. Di Toro's formulation uses the assumption that benthic
biomass is proportional to the labile carbon in the sediment. As AQUATOX calculates benthic
biomass explicitly, this simplifying assumption is not required and a direct empirical relationship
based on benthic biomass is utilized.
where:
      Benthic Biomass
      H2
      le-4
                             -ir\(Log(Benthic_Biomass)-2.n&'l5'l  )   -i    A
                         l,2
                                                                                   (262)
                         particle mixing velocity between layers (m/d)
                         sum of benthic invertebrate biomass (g/m  dry);
                         depth of sediment layer 2 (m); and
                         pore water concentration (m2/cm2);
                                          210

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 7
               Figure 129: Relationship derived from Di Toro, 2001, Figure 13.1A
                "Diffusion coefficient for particle mixing versus benthic biomass"
                  OQ
                     0.001
                          0.1
               1         10       100
                  Biomass (g dry/m2)
1000
Additionally, the calculation of benthic biomass by AQUATOX includes benthic invertebrate
mortality due to low oxygen conditions and recovery when oxygen concentrations rise. Because
of this, Di Toro's benthic stress model incorporating accumulated stress and dissipation of stress
is not required nor included within AQUATOX.

Surface Mass Transfer Coefficient

Di Toro has  advanced the idea  that the  diffusive surface mass transfer coefficient  can  be
successfully related to the sediment oxygen demand (Di Toro et al. 1990).    The resulting
equation is as follows.
                                      SOD
                                   Oxygenm
                                           Water
                                                                                (263)
                               SOD = CSOD+NSOD
where:
      s
      SOD
      CSOD

      NSOD

      Oxygen Water
=  surface diffusive transfer (m/d)
=  sediment oxygen demand (g O2 / m2 d);
=  carbon based sediment oxygen demand (g O2 / m2 d) see (287) or
   (291);
=  sediment oxygen demand due to nitrification (g ©2 / m  d) see (275),
   converted into oxygen equivalent units (1.714 gO2/gN);
=  overlying water oxygen cone, (g ©2 / m3) (186).
                                         211

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                          CHAPTER 7
As shown above, SOD is the sum of the carbon based sediment oxygen demand and sediment
oxygen demand due to nitrification..
7.2 POC

Particulate Organic Carbon in sediment is assumed to be located exclusively in the second layer
of sediment.  Three state variables are utilized to represent three reactivity classes (Gi through
Gs).  POC is a component of the particulate organic matter that settles from the water column
into the anaerobic layer and decomposes; it is also subject to consumption by detritivores.  In this
case, the POC uptake from that predation must be calculated separately from the POP and PON.

             dt
                *=L = Deposition - Mineralization - Burial - (Predation I Detr2OC)    (264)
where:
      Deposition
      Mineralization
      Burial
      Predation
      Detr2OC
                         deposition from water column (g C/ m3 d) see (266);
                         decomposition (g C/ m3 d) see (267) ;
                         deep burial below modeled layer (g C/ m3 d) see (265);
                         predation by detritivores (g C/ m3 d) see (99); and
                         detrital organic matter is assumed to be 1.90 • organic  carbon as
                         derived from stoichiometry (Winberg 1971).
For all state variables burial is solved as a function of the user input burial rate W2-'
where:
      Burial
      POM
      W2
      Hn
                                  Burial = POM --
                         burial below modeled layer (g C/ m3 d); and
                         POP, POC, or PON (g C/ m3);
                         user input burial rate  (m/d); and
                         depth of sediment layer n (m).
                                                                                 (265)
Burial from the top layer is added to the second layer, whereas burial from the second layer is
considered deep burial out of the modeled system.

Deposition is solved as
       DepositionPOM Gi =\  £ Def • Def2POMGi
                          Animals
                                                         - Sed2POMGi      TOter    (266)
                                                Algae &Detritus
                                                                      ^^^
                                                                         sediment
where:
      Deposition POM Gi =  deposition of G; reactivity class of POP, POC, or PON from water
                         column (g OM/ m3 d);
      Def            =  defecation of animals, see (97) (g OM/m3water d);
      Def2POMGi     =  fraction of POP, POC, or PON reaction class G; in defecated matter;
                                         212

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7


      Sed             =   sedimentation of plants or detritus, see (165), (g OM/m3water d);
      Sed2POMa     =   fraction of POP, POC, or PON reaction class G; in sedimented algae
                          or detritus (unitless);
      Volwater         =   water volume (m3); and
      Volsediment       =   sediment volume (m3);

Assigning fractions of defecation to the relevant POM class (i.e., determining Def2POMoi)  is a
two-part process.  First, the fraction of POM, POC, or PON in the defecated material  must be
determined.  Second, each fraction must be again multiplied by a fraction to assign it to the three
reactivity classes (Gi to Gj).   In this manner, particulate organic matter is  separated into nine
different state variables in the sediment.

The  fractions of POP  and PON within defecated matter are  assumed to  equal the ratios of
phosphate or nitrate to  organic  matter for  sedimented  labile  detritus;   these  are  editable
parameters ("remineralization" screen).  The fraction of POC within defecated matter  is set to
52.6% (Winberg  1971). Defecated matter is split evenly between reactivity classes GI and G2,
with no defecation assigned to the non-reactive Gs class (Def2SedLabile=0.5).

Similarly, assigning  fractions of sedimentation to reactivity  classes  is a two-part process.  As
before, the fraction of POP and PON within sedimented matter is assumed equal to the ratio of
phosphate or nitrate to organic matter for the given species or  detritus (editable parameters). The
fraction of POC within  sedimented matter is again set to 52.6% (Winberg 1971). The amount of
refractory detritus that is converted to reactivity class Gs is a user entered parameter. The rest of
the refractory detritus is assigned to G2 and labile detritus becomes GI.  92% of sinking plants
are assumed to be  labile  (Gi) with no  sinking algae being converted to the  non-reactive
compartment (Gs).

The  decomposition of organic  matter is calculated as a first-order reaction with an exponential
temperature  sensitivity built in:
                   MineralizationPOM Gi = POMGi • KPOM_Gi • 0POM_G?emp~2°              (267)
where:
      Mineralization POM _a =   decomposition of G; reactivity class of POP, POC, or PON in
                              the sediment bed (g/m3 d);
      POMa              =   concentration of POM in reactivity class G; (g/m3);
      KpoM_Gi             =   decay rate of POM class (1/d);
      OPOM_GI              =   exponential temperature adjustment for decomposition of POM
                              class G; (unitless); and
      Temp               =   temperature (deg.C).
Feeding on GI is calculated based on preferences for labile detritus and feeding on G2 is based
on preferences for refractory detritus; these are set in the animal data screens.  As a simplifying
assumption, there is no feeding on nonreactive Gj.

                                           213

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 7
7.3 PON

Particulate Organic Nitrogen in sediment is also assumed to be in the second layer of sediment.
Three state variables are utilized to represent three reaction classes (Gi through Gj).
           dPON
                Sediment
               dt
                      = Deposition - Mineralization - Burial - Predation • N2Org
                                                           (268)
where:
      Deposition       =
      Mineralization   =
      Burial          =
      Predation       =
      N2Org
   deposition from water column (g N/ m3 d) see (266);
   decomposition to ammonia (g N/ m3 d) see (267) ;
   deep burial below modeled layer (g N/ m3 d) see (265);
   predation by detritivores (g N/ m3 d) see (99); and
   user input conversion factor between N  and refractory or labile
   detritus (g N / g OC).
7.4 POP
Particulate Organic Phosphate in sediment is solved in a very similar manner to POC and PON.
Mineralization rates may be different, however.
           dPOPi
                Sediment
               dt
where:
= Deposition - Mineralization - Burial - Predation • P2Org
                                                                                 (269)
      Deposition       =
      Mineralization   =
      Burial          =
      Predation       =
      P2Org
   deposition from water column (g P/ m3 d) see (266);
   decomposition to orthophosphate (g P/ m3 d) see (267) ;
   deep burial below modeled layer (g P/ m3 d) see (265);
   predation by detritivores (g P/ m3 d) see (99); and
   user input  conversion factor  between P  and refractory or labile
   detritus (g P / g OC).
7.5 Ammonia

Ammonia in the  sediment is solved using two  state  variables to  represent the  two layers.
Ammonia is formed  by  the decomposition of PON.   Ammonia in  sediment undergoes
nitrification, burial, and flux to or from the water column. The ammonia in each state variable is
the sum of dissolved and particulate ammonia. The fraction that is dissolved is solved below in
equation (274).  The ammonia differential equations are as follows:
                 dAmmonia
                           'L2Sed
                       dt
                                = Diag _ Flux - Burial + Flux2Anaerobic
                                                           (270)
                                         214

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7
           dAmmoniaLlSed
           	= -Nitrification - Burial - Flux2Water - Flux2 Anaerobic      (271)
where:
      Diag Flux      =  decomposition of PON, see (267) ;
      Burial          =  burial below relevant layer (g N/ m3 d) see (265);
      Flux2Anaerobic =  flux to layer 2 from layer 1 (g N/ m3 d, may be negative) see (272) ;
      Flux2Water     =  flux to water from layer 1 (g N/ m3 d, may be negative);  see (273);
      Nitrification     =  conversion to nitrate (g N/ m3 d) see (275);
           Flwc2Anaerobic = - o)   fConc  - fConc  + KL\fd2C
where:
      co 12            =  particle mixing velocity between layers (m/d), see (262);
      KL             =  diffusion velocity between layers (m/d), see (261);
      fpjayer           =  particulate fraction in layer 1 or 2 (unitless);  see (274)
                      =  dissolved fraction in layer 1 or 2 (unitless); see (274)
      Cone layer         =  total concentration of state variable in layer (g/m3); and
      //layer            =  depth of layer being evaluated (m);


                        Flux2Water = s(fdl • cone, - concmter col )/H,                    (273)

where:
      s                =  surface diffusive transfer (m/d); (263)
      fdi               =  dissolved fraction in layer 1;
      Cone iayer         =  total concentration of state variable in layer (g/m3); and
      HI               =  depth of layer 1 (m);

The fraction of ammonia that is dissolved in each layer is calculated as follows:


                               f          -
                               J d ammonia, layer   .
                                                                                     (274)
                               f          = ^- f
                               J p ammonia, layer     J d ammonia, layer

where:
                      =  dissolved fraction in layer;
                      =  user-input solids concentration in layer (kg/L);
                      =  editable  partition coefficient for ammonium (L/kg); and
      fp ammonia,iayer     =  parti culate fraction in layer.
                                           215

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                   CHAPTER 7
Ammonia in the top layer is converted to nitrate in the presence of oxygen, resulting in sediment
oxygen demand.  Since the nitrification reaction requires oxygen, no nitrification is assumed to
occur in the lower anaerobic layer. Nitrification in the aerobic layer is calculated as follows:
          Nitrification =
                              DOWC
                      KM
                                                   NH4
                         2-KM
                               02
                                             (NH4,
(275)
where:
      Nitrification
      DOWC.
      KMNH4

      KM02
      9

      Temp
conversion of ammonia to nitrate (g N/m d);
dissolved oxygen in the water column (g/m3);
user-input  nitrification  half-saturation coefficient for  ammonium
 (gN/m3);
user-input  nitrification  half-saturation   coefficient  for  oxygen
(g 02/m3);
reaction velocity for nitrification (m/d);  (user-input, differentiating
between fresh and salt water)
surface diffusive transfer (m/d); (263)
concentration of ammonia in layer 1 (g/m3); (168)
user-input depth of layer 1 (m);
user-input  exponential  temperature  adjustment  for  nitrification
(unitless); and
temperature (deg.C).
7.6 Nitrate

Nitrate is formed by the nitrification of ammonia in the top layer of the sediment bed.  Nitrate in
sediment undergoes denitrification, burial and flux to or from the water column.
                    dNitrate L2Si
                               'ed
                          dt
                                  = -Burial - Denitr + Flux2 Anaerobic
                                                          (276)
        dNitrate
                 LISed
             dt
                     = Nitrification - Denitr - Burial - Flux2Water - Flux2 Anaerobic    (277)
where:
      Burial
burial to layer below modeled layer or out of the system(g N/ m d)
see (265);
                                           216

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 7


      Flux2Anaerobic  =  flux to layer 2 from layer 1 (g N/ m3 d, may be negative) see (272) ;
      Flux2 Water      =  flux to water from layer 1  (g N/ m3 d, may be negative);  see (273);
      Nitrification     =  conversion of ammonia to nitrate (g N/ m3 d), see (275);
      Denitr          =  denitrification of nitrate to free nitrogen (g N/ m3 d), see (278);

Nitrate is assumed to be dissolved in the sediment bed so/,/ =1.0 andfp = 0.0.

Denitrification is solved as follows

                                         2 £.  Temp-20
                         Denitr = K^>^ '°>™ -      2=l                     (278)
where:
      K layer, No3        =  user-input reaction velocity for denitrification  given the layer and
                         salinity regime (m/d);
      6               =  user-input exponential temperature adjustment for denitrification
                         (unitless); and
      s               =  surface diffusive transfer (m/d);  (263)
      H iayer           =  depth of layer (m);
      NO 3 layer         =  concentration of nitrate in layer (g/m3); and
      Temp           =  temperature (deg.C).
7.7 Orthophosphate

Phosphate in the sediment is solved using two state variables to represent the two layers.  Like
ammonia, the phosphate in each state variable represents the sum of dissolved and particulate
phosphate.

                    dPO4 L2Sed
                    	= Diag _ Flux - Burial + Flux2 Anaerobic               (279)
                        dt
                   dPO4 n_ ,
                   	^^ = -Burial - Flux2Water - Flux2 Anaerobic              (280)
                       dt
where:
      Diag Flux      =  decomposition of POP, see (267) ;
      Burial           =  burial to layer below modeled layer or out of the system(g P/ m3 d)
                         see (265);
      Flux2Anaerobic  =  flux to layer 2 from layer 1 (g P/ m3 d, may be negative) see (272) ;
      FluxlWater      =  flux to water from layer 1  (g P/ m3 d, may be negative);  see (273);

When oxygen is present in the water column, the diffusion of phosphorus from sediment pore

                                          217

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7


waters is limited. This is due to strong P  sorption to oxidated ferrous iron in the aerobic layer
(iron oxyhydroxide  precipitate).  Under conditions of anoxia, phosphorus flux from sediments
increases significantly.

Di Toro incorporates the effect of oxygen on  phosphate flux into his  model by making the
dissolved fraction of phosphate a function  of oxygen in the water column. When the oxygen in
water decreases below a critical threshold the partition coefficient for phosphate is increased by a
user-entered factor.  As the oxygen goes to zero, the partition coefficient is smoothly reduced to
the anaerobic coefficient using an exponential function:
                 if  DOWC >DOCntP04 then Kdp04l = Kdp04i2AKdp04^

                                                                                   (281)
                                                                   DOwc
                                     else  Kdpn,, = Kdpn, ,AKdpn,,Doc
                                            *PO4 1 ~~ Jvt*PO4 2A-"vt*PO4 l"^ca,P04


Partitioning of phosphate between the dissolved and particulate forms will affect on the flux of
phosphate to the water column (273).



                            J d phosphate, layer   i  . ...     T^_/                              ^   '
where:
                      =  dissolved fraction in layer (unitless);
                      =  user-input solids concentration in layer (kg/L); and
      Kdpo4,2         =  partition coefficient for phosphate in layer 2 (L/kg);
      &Kdpo4,i        =  fresh or saltwater factor to increase  the aerobic (Li) partition
                         coefficient of PO4 relative to the anaerobic (L2) coeff (unitless);
      DOwc          =  dissolved oxygen in the water column (g/m3), see (186); and
      DOcrit,PO4.       =  critical  oxygen concentration for  adjustment of partition  coefficient
                         for inorganic P (g/m3);
7.8 Methane

Methane is formed due to the decomposition  of POC in the sediment bed under low-salinity
conditions. Methane undergoes oxidation resulting in increased sediment oxygen demand.

           dMethane L2Sd
           	—	= Drag _ FluxMethane - Flux2WaterMethane - OxidationMethane      (283)


where:
      Methane ^sed    =   methane in the anaerobic layer  expressed in  oxygen equivalence
                          units (g O2equiv / m3)
      Diag Flux      =   decomposition of POC in freshwater, adjusted for the organic carbon

                                          218

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 7


                         lost due to denitrification (g O2equiv / m3 d) see (284);
      Flux2 Water      =   methane flux to water (g O2equiv / m3 d),  see (288); and
      Oxidation        =   oxidation of methane (CSOD) (g O2equiv / m3 d) see (287);


In the manner of Di Toro, methane and sulfide are tracked in units of oxygen equivalents  (g
O2equiv / m3) to easily balance the model's computations.

In fresh water conditions, decomposing POC is converted to methane which is tracked in oxygen
equivalents.  In salt water, decomposing POC becomes sulfide. However, some POC  is lost due
to denitrification and does not decompose:

                                                  f32^
           Diag _ FluxMethangf Sulfide = Mineralizationpoc 1 — 1-2.86- Denitrification       (284)

where:
      Diag_FluXMethane,Sulfide
                      =   decomposition  of POC in water, adjusted for the organic carbon lost
                         due to denitrification (g O2equiv / m3 d);
      Mineralizationpoc =   decomposition  of POC in freshwater, (g POC / m3 d) see (267) ;
      Denitrification    =   denitrification of nitrate, (g N/ m3 d) see (278);
      32/12           =   conversion between POC and oxygen equivalents; and
      2.86             =   conversion between Nitrate and oxygen equivalents;
Oxidation of methane is solved as a function of the saturation concentration of methane in pore
water.
                           CH4  =100l +   sLi^2420-Temp                       (285)
                                       ^    10 )

          CSODMax = mm (j2KL • CH4sat • Diag _ FluxMethane , Diag _ FluxMethane }     (286)
                                                            Temp-20
                                  CSODM\l-sech\
                 OxidationMethane =	^	^             (287)
where:
      CH4sat         =  saturation concentration of methane in pore water (g O2equiv / m3);
      zmecm           =  mean depth of water column above the sediment bed (m);
      Temp          =  temperature (deg.C);
      CSODMax       =  maximum oxidation flux (g  O2equiv / m2 d);
      KL             =  diffusion velocity between layers (m/d);  (261)
      Diag_FluxMethane =  diagenesis  flux of methane  to water column, adjusted to be in units
                                         219

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 7
      OxidatioriMethane
      sech
      s
      KCH4
      H2
   of(g02equiv/m2d);
   oxidation of methane (g O2equiv / m3 d);
   hyperbolic secant function
   surface diffusive transfer (m/d); (263)
   reaction velocity for methane oxidation(m/d);
   exp. temperature adjustment for methane oxidation (unitless); and
   depth of layer 2 (m);  (methane mass arbitrarily tracked on the second layer)
All methane is assumed to be oxidized or to escape from the sediment to water.  Thus the
derivative for methane will remain at zero and the solution for the flux to water can be solved as
follows:
                   Flux2WaterMetham = Diag_FluxMethane - OxidationMethane
                                                             (288)
where:
      Diag Flux      =  decomposition of POC in freshwater, adjusted for the organic carbon
                         lost due to denitrification (g O2equiv / m3 d), see (284);
      Oxidation       =  oxidation of methane (g O2equiv / m3 d), see (287);
7.9 Sulfide

Sulfide  is formed, rather than methane, under saline conditions.   Sulfide  in  sediment may
undergo burial, flux to the water column, or oxidation, which increases SOD.
                 dSulfideL2Sea
                      dt
                             = Diag _ FluxSulflde - Burial + Flux2 Anaerobic
                                                             (289)
             dSulfide
                    LISed
                 dt
                         = -Oxidation - Burial - Flux2Water - Flux2 Anaerobic
                                                             (290)
where:
      Sulfide LnSed
      Diag Flux

      Burial
      Flux2Anaerobic
      Flux2Water
      Oxidation
=  suifide concentration in layer n of sediment, (g O2equiv / m3);
=  decomposition of POC in salt water, adjusted for the organic carbon
   lost due to denitrification (g O2equiv / m3 d), see (284);
=  burial to layer below modeled layer or out of the system (g O2equiv /
   m3 d); see (265);
=  flux to layer 2 from layer 1 (g O2equiv / m3 d, may be neg.) see (272) ;
=  flux to  water from LI (g O2equiv  / m3 d,  may be neg.) (Note the
   driving var. "COD" represents the water col. cone, of suifide.) see
   (273);
=  oxidation of suifide in the active layer;
                                          220

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7
                                  \KH2S,d  'J
   ,      2  f  \a   Temp-20\
dl~*~K~H2S,p  'Jpl}UH2S
        Oxidation
                 Sulfide        nzo,Li                         y-y-
                                                        s • Hl
where:
      Oxidation suifide   =  oxidation of sulfide (g O2equiv / m3 d);
      Concms.Li        =  concentration of sulfide in layer 1 (g O2equiv / m3);
      Kff2S,d            =  reaction velocity for dissolved sulfide oxidation (m/d);
      KH2S,p            =  reaction velocity for paniculate sulfide oxidation (m/d);
      DO we.           =  dissolved oxygen in the water column (g/m3);
      KMms.DP        =  sulfide oxidation normalization constant for oxygen (g O2/m3);
      0H2S             =  exp. temperature adjustment for sulfide oxidation (unitless);
      s                =  surface diffusive transfer (m/d); and
      HI               =  depth of layer 1 (m);
The fraction of sulfide that is dissolved in each layer is calculated as follows:


                              Jd sulfide, layer ~ ~.         ~7T~,
                                          [+miayer'KaH2S, Layer

where:
      fdsuifidejayer       =  dissolved fraction in layer;
      miayer            =  solids concentration in layer (kg/L); and
      KdNH4           =  partition coefficient for sulfide for layer (L/kg);

The particulate fraction  of sulfide in each layer is calculated as one minus the dissolved fraction.


7.10 Biogenic Silica

Silica in sediment is modeled using three state variables.  Silica associated with diatoms and
deposited from the water column is biogenic silica and is modeled in Layer 2.  Biogenic silica
can then either undergo  deep burial or dissolution to dissolved silica.
                  dBiogenic _ Silica L2 Sed
                                 - = Deposition - Dissolution - Burial            (293)
                           dt
where:
      Deposition       =  deposition from water column (g Si/ m3 d) see (294);
      Dissolution      =  dissolution of biogenic silica (g Si/m3d)
      Burial           =  deep burial below modeled layer (g Si/ m3 d) see (265); and
                                           221

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7


Deposition of silica is a function of the sinking of diatoms:
                        DepositionSi=\  ^Sed• FracSilica \	water                    (294)
                                      \Diatoms              J ^^"^ sediment

where:
      Depositions,     =  deposition of silica from water column (g Si/ m3 d);
      FracSilica       =  user-input fraction of silica in diatoms, (unitless);
      Sed             =  sedimentation of diatoms, see (165), (g OM/m3water d);
      Volwater          =  water volume (m3); and
      Volsediment        =  sediment volume (m3);
Biogenic silica can undergo dissolution to dissolved silica.  This reaction can  also operate in
reverse:
       T\-   1 +•        n  Temp-2Q        OnC Avail Si       /o.      /•         ^          \  ,-_„
      Dissolution = KSi9Sl       - - -       \(SiSat - fdlll  L2 • ConcSil    }  (295)
                                 Cone,.  ,,+KMf   '
                                     -Avail  Si '  ^r-1 PSi
where:
      Dissolution      =  dissolution of biogenic  silica (g Si/m3 d);
      KSI              =  user-input reaction velocity for dissolved silica dissolution (1/d);
      6si              =  user-input exponential temperature adjustment for silica dissolution
                          (unitless);
      Concmr,iayer      =  concentration of available silica or silica in layer 2 (g Si/ m3);
      KMpsi           =  user input silica  dissolution half-saturation  constant  for  biogenic
                          silica (g Si/m3);
      Si sat             =  user-input saturation concentration of silica in pore water (g Si/m3);
                       =  dissolved fraction of silica in layer.
7.11 Dissolved Silica

Dissolved  silica is produced when biogenic silica breaks down due to dissolution, and could
potentially be modeled as a  limiting nutrient for diatoms  in a later version  of AQUATOX.
Dissolved  silica (referred to hereafter as "silica") is modeled in two layers:

                    dSilica L2Sed
                    	= Dissolution - Burial + Flux2 Anaerobic               (296)
                         dt
                   dSilica n „,,
                   	^^ = -Burial - Flux2Water - Flux2Anaerobic               (297)

where:

                                            222

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 7


      Dissolution      =   dissolution of biogenic  silica (g Si / m3 d), see (295);
      Burial          =   burial to layer below modeled layer or out of the system (g Si / m3
                          d); see (265);
      Flux2Anaerobic =   flux to layer 2 from layer 1 (g  Si/ m3 d, may be negative) see (272) ;
      Flux2Water     =   flux to water from layer 1 (g Si/ m3 d,  may be negative); see (273);
Similar to  inorganic phosphate, dissolved oxygen causes a barrier to silica flux to the water
column.  This is modeled by increasing the partition coefficient by a factor when the dissolved
oxygen decreases below a critical threshold.
                   if DOwc>DOCnt^  then
                                                                                    (298)
                                                                 D0wr
                                        else  Kdsi l = Kdsi 2AKdSi j
                               Jd Si, layer
where:
      fd sihca,iayer       =   dissolved fraction in layer (unitless);
      miayer           =   solids concentration in layer (kg/L); and
      Kdst,2           =   partition coefficient for silica in layer 2 (L/kg);
      kKdsij          =   fresh  or  saltwater factor to  increase the  aerobic  (Li) partition
                          coefficient of silica relative to the anaerobic (L2) coeff (unitless);
      DOwc          =   dissolved oxygen in the water column (g/m3); and
                      =   critical oxygen concentration for adjustment of partition coefficient
                          for silica (g/m3);
                                           223

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                CHAPTER 8
                           8. TOXIC ORGANIC CHEMICALS
The  chemical  fate  module  of  AQUATOX  predicts  the
partitioning of  a compound between water, sediment,  and
biota (Figure 130), and estimates the rate of degradation of the
compound   (Figure    131).       Microbial    degradation,
biotransformation,  photolysis, hydrolysis, and  volatilization
are modeled  in AQUATOX.   Each  of these processes  is
described generally, and again in more detail below.

Nonequilibrium  concentrations,  as represented  by  kinetic
equations, depend on sorption, desorption, and elimination  as
functions of the chemical, and  exposure through water  and
food  as  a  function  of  bioenergetics  of  the  organism.
Equilibrium   partitioning  is   no   longer  represented   in
AQUATOX except as a constraint on sorption to detritus and
plants  and  as   a basis  for  computing  internal  toxicity.
Partitioning to inorganic sediments is not modeled unless the
multi-layer sediment model is included.
Toxic Organic Chemicals:
Simplifying Assumptions

 • Kinetic model of toxicant fate
 • Photolysis in sediments is not
   included
 • A generalized equation is used to
   calculate partitioning of polar
   compounds
 • Direct sorption onto the body of an
   animal is ignored
 • The exchange of toxicant through
   the gill membrane is assumed to be
   facilitated by the same mechanism
   as the uptake of oxygen
 • Estimation of the elimination rate
   constant k2 may be made based on
   logKow with two alternative
   formulations available
 • Biotransformation occurs at a
   constant rate throughout a
   simulation
Microbial degradation is modeled by entering a maximum biodegradation rate for a particular
organic toxicant, which is subsequently reduced to account for suboptimal temperature, pH, and
dissolved oxygen.  Biotransformation is  represented by user-supplied first-order rate  constants
with the option  of also modeling multiple daughter products.  Photolysis is modeled by using a
light screening factor (Schwarzenbach et al.,  1993) and the near-surface, direct photolysis first-
order rate constant for each pollutant.  The light screening factor is a function of both the diffuse
attenuation coefficient near the surface and the average diffuse attenuation coefficient for the
whole water column.  For those organic chemicals that undergo hydrolysis, neutral, acid-, and
base-catalyzed reaction  rates  are entered into  AQUATOX  as applicable.   Volatilization  is
modeled using a stagnant two-film model, with the air and water transfer velocities approximated
by empirical equations based on reaeration of oxygen (Schwarzenbach et al.,  1993).
                                            224

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 8
         Figure 130. In-situ uptake and release of chlorpyrifos in a pond, dominated by plants


430
-207
344


£

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


The  mass balance equations  follow.  The change in mass of toxicant in the  water includes
explicit representations of mobilization  of the toxicant from sediment to water as a result of
decomposition of the labile sediment detritus compartment, sorption to and desorption from the
detrital  sediment compartments, uptake  by algae and macrophytes, uptake across the gills of
animals, depuration by organisms, and turbulent diffusion between epilimnion and hypolimnion:
       dToxrcantWater = Loading +  ^.^ (Decomposition LMeDetr • PPBLMeDetr •  1 e - 6)

                     + T. Desorption DetrTox + T. Depuration Org- 1. Sorption SedTm
                     - J^GillUptake - Macro Uptake- J^AlgalUptakeAlga                (300)
                     - Hydrolysis - Photolysis - MicrobialDegrdn + Volatilization
                     -Discharge  + BiotransformMicrobIn ±  TurbDiff ±DiffusionSeg
                     + PorewaterAdvection + Diffusion Sediment- Washout + Washin
The equations for the toxicant associated with the two sediment detritus compartments are rather
involved, involving direct processes  such as sorption and indirect conversions such as defecation.
However, photolysis is not included based on the assumption that it  is not a significant process
for detrital sediments:
      dToxicant &
               SedLabileDetr
Sorption - Desorption + (Colonization •  PPB sedRefrDetr •  1 e - 6)
              dt
                    +  ZpredZprey(Def2SedLabile •  DefecationTox Pred Pre)

                    - (Resuspension + Scour + Decomposition) •  PPB sediabHeDetr  • le-6

                    - T.PredIngestionpre4SedLab,leDetr '  PPBSedLabUeDetr "16-6              (3Q1)
                    + Sedimentation  • PPBsuspLdbiUDetr • 1 e - 6
                    + E(Sed2Detr •  SmkPhyto- PPBPhyto-  le-6)
                     - Hydrolysis - MicrobialDegrdn - Burial + Expose
                                  ± BiotransformMjcrobml
                                  SedRefrDetr
                                     - - =
                                            r,    ..     ^      ,.
                                          = Sorption - Desorption
                                dt
                  + ^Pred^prey((l-Def2SedLabile) •  DefecationTox Pred Pre)

                (Resuspension + Scour + Colonization)  • PPB sedRefrDetr  ' le-6

                     -^PredIngeStiOnpre(iSedRefrDetr •  PPB SedRefrDetr  '16-6               (3Q2)

                     + (Sedimentation + Scour) • PPB suSpRefrDetr  • le-6

                       + Z(Sed2Detr • Sinkphyto •  PPBPhyto  • le-6)
                     - Hydrolysis - MicrobialDegrdn - Burial + Expose
                                  ± BiotransformMjcmbjal

                                          226

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8
Similarly for the toxicant associated with suspended and dissolved detritus, the equations are:
                     SuspLabileDetr 	
                              = Loading + Sorption - Desorption + WashinToxCarrie
                   dt
               + ^0rg((Mort2Detr-Mortality 0rg + GameteLoss0rg)  • PPB0rg • 1 e-6)
                - (Sedimentation + Deposition + Washout + Decomposition
                + EpredIngestionpre4SuspLaWeDetr)  • PPBsuspLauieDetr  • 1 e-6                (303)
                + Colonization • PPBSuSpRefrDetr  • 1 e-6 +  BiotransformMicrobial
               + (Resuspension + Scour) •  PPB sedLabiieDetr  ' 1 e -6 +  SedToHyp
               - Hydrolysis - Photolysis - MicrobialDegrdn ± TurbDiff ± DiffusionSeg
                             SuspRefrDetr
                               — -  =
                                        T    ,.     r,   ,.     ^      ,.
                                      = Loading + Sorption - Desorption
                           dt
                     + ^0fg(Mort2Ref • Mortality Org  •  PPB0rg  • 1 e -6)
                 - (Sedimentation + Deposition + Washout + Colonization
             ± BiotransformMwmbjal + ^Pred Ingestion SuspRefrDetr ) • PPBSuspRefrDetr • 1 e -6        (304)
                     + (Resuspension + Scour) •  PPBsedRefrDetr •  1 e -6
                ±  SedToHyp - Hydrolysis - Photolysis - MicrobialDegrdn
                +  TurbDiff + Diffusion Seg + WashinToxCarner

       dToxicantDlssLMeDetr = Loading + Sorption - Desorption +  SumExcrToxToDiss0rg
                    + T.0rg(Mort2Detr •  Mortality Org  • PPB0rg  • 1 e -6)
                    - (Washout + Decomposition) •  PPBDlssLabneDetr  •  1 e -6              (305)
                    + BiotransformMicrobial - Hydrolysis - Photolysis
                   - MicrobialDegrdn  ± TurbDiff ± DiffusionSeg + WashinToxCarrier
                   ± PorewaterAdvection ± Diffusion Sedment

        d Toxicant D,ssRefrDetr = Loading + Sorption . Desorption +  SumExcToxToDiss0rg

                     + ^0fg(Mort2Ref • Mortality Org  •  PPB0rg  • 1 e -6)
                     - (Washout + Colonization) • PPBDlsSRefrDetr  • 1 e -6               (306)
                     +  BiotransformMicmbial - Hydrolysis - Photolysis
                    - MicrobialDegrdn ± TurbDiff ± DiffusionSeg + WashinToxCarrier
                     ± PorewaterAdvection ± Diffusion Sediment
                                           227

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


When the simple sediment model is run, there are no equations for buried detritus, as they are
considered to be sequestered and outside of the influence of any processes which would change
the concentrations of their associated  toxicants.   When the multi-layer  sediment model is
included, equations for toxicants in pore waters and toxicants in buried sediments may be found
in sections 8.10 and 8.11.

Toxicants associated with algae are represented as:

    	— = Loading + AlgalUptake - Depuration  + TurbDiff + DiffusionSeg

             + WashinTmCarner  - (Excretion + Washout + T.Pred PredationPred, Aiga + Mortality  (307)
              + Sink ±  SinkToHypo ± Floating)  • PPBAiga ' 1 e -6 ± BiotransformAlga

Macrophytes are  represented similarly,  but reflecting the fact that they  are stationary unless
specified as free-floating:
                    Macrophyte _
                            = Loading + MacroUptake - Depuration - (Excretion
                  dt
                + Epred PredationPred, Macro + Mortality + WashoutFreeFloating + Breakage)     (308)
                • PPBMacro ^e-6 + Biotransform Macmphyte + WashinTmCarnerFreeFloat
The toxicant associated with animals is represented by an involved kinetic equation because of
the various routes of exposure and transfer:

             d ToxicantAmmal  = Loading + GillUptake + Z wwfe  + TurbDiff

                - (Depuration + *LPredPredationpred,Ammai + Mortality + Spawn          (399)
            + Promotion + Drift + Migration + Emergelnsect) • PPBAmmai  •  ^ e -6
                            + Biotransform Ammal + WashmToxCarner
where:
       Toxicantwater       =      toxicant in dissolved phase in unit volume of water (ug/L);
       ToxicantsedDetr      =      mass of toxicant associated with each of the two sediment
                                  detritus compartments in unit volume of water (ug/L);
       ToxicantsuspDetr      =      mass of toxicant associated with each of the two suspended
                                  detritus compartments in unit volume of water (ug/L);
       ToxicantDissDetr      =      mass of toxicant associated with each  of the two dissolved
                                  organic compartments in unit volume of water (ug/L);
       ToxicantAiga        =      mass of toxicant associated with given alga in unit volume
                                  of water (ug/L);
       Toxicant Macmphyte    =      mass  of  toxicant associated  with  macrophyte in  unit
                                  volume of water(ug/L);
       Toxicant Animal       =      mass  of toxicant associated with given animal in  unit
                                  volume of water (ug/L);
       PPBsedDetr          =      concentration of toxicant in sediment  detritus ((j,g/kg), see
                                  (310);

                                           228

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                           CHAPTER 8
       PPBsuspDetr
       PPBotssDetr
       PPBAlga
       i i BMacrophyte
       i it) Animal
       1  e-6
       Loading
       TurbDiff

       Washin
       WashinToxCarner

       Diffusionseg

       DiffusionSediment

       PorewaterAdvection

       Hydrolysis
       BiotransformMicrobmi


       Biotransform org

       Photolysis

       MicrobialDegrdn

       Volatilization
       Discharge

       Burial

       Expose

       Decomposition
       Depuration

       Sorption

       Desorption

       Colonization
concentration of toxicant in suspended detritus (ug/kg);
concentration of toxicant in dissolved organics (ug/kg);
concentration of toxicant in given alga (ug/kg);
concentration of toxicant in macrophyte (ug/kg);
concentration of toxicant in given animal ((j,g/kg);
units conversion (kg/mg);
loading of toxicant from external sources (ug/L-d);
depth-averaged turbulent diffusion between epilimnion and
hypolimnion (ug/L-d), see (22) and (23).
loadings from linked upstream segments (g/m3-d), see (30);
inflow load of toxicant sorbed to a carrier from an upstream
segment (ug/L-d), see (31);
gain or loss due to  diffusive transport over the feedback
link between two segments, (ug/L-d), see (32);
gain or loss due to diffusive transport to  porewaters in the
sediment (ug/L-d), see (256);
gain or loss of toxicant to  porewater  due to scour  or
deposition of sediment (ug/Lpw-d), see (394), (395);
rate of loss due to hydrolysis (ug/L-d), see (313);
biotransformation to or  from given  organic  chemical  in
given   detrital  compartment      due   to   microbial
decomposition (ug/L-d), see (375);
biotransformation to or from given organic chemical within
the given organism (ug/L-d);  (375)
rate of loss due to  direct photolysis (ug/L-d), see (320);
assumed not to  be significant for bottom sediments;
rate of loss due to microbial  degradation (ug/L-d),  see
(326);
rate of loss due to volatilization (ug/L-d), see (331);
rate of loss of toxicant  due to discharge  downstream
(ug/L-d), see Table 3;
rate of loss due  to deposition  and  resultant  deep burial
(ug/L-d) see (167b);
rate of exposure due to resuspension of overlying sediments
(ug/L-d),see(227);
rate of decomposition of given detritus (mg/L-d), see (159);
elimination rate for toxicant due to clearance (ug/L-d),  see
(362), (363), and (372);
rate of sorption to given organic or inorganic compartment
(ug/L-d), see (350);
rate of  desorption from  given organic  or  inorganic
compartment (ug/L-d),  see (351);
rate of conversion of refractory  to labile detritus (g/m3-d),
see (155);
                                           229

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                          CHAPTER 8
       DefecationToxpred, pre~-

       Def2SedLabile

       Resuspension

       Scour


       Sedimentation

       Deposition


       Sed2Detr

       Sink

       Breakage
       Mortalityorg

       Mort2Detr
       GameteLoss
       Mort2Ref
       Washout or Drift


       SedToHyp
                  , prey
                   prey
       ExcToxToDissorg

       Excretion
       SinkToHypo
       AlgalUptake
       MacroUptake
       GillUptake
rate of transfer of toxicant due to defecation of given prey
by given predator (ug/L-d), see (379);
fraction of defecation that goes to sediment labile detritus,
= 0.5;
rate of resuspension of given sediment detritus (mg/L-d)
without the inorganic sediment model attached, see (165);
rate of resuspension of given sediment detritus (mg/L-d); in
streams with the inorganic sediment model attached, see
(233);
rate of sedimentation of given suspended detritus (mg/L-d);
without the inorganic sediment model attached, see (165);
rate of sedimentation of given suspended detritus (mg/L-d)
in streams with the inorganic sediment model attached, see
(235);
fraction of sinking phytoplankton that goes to given detrital
compartment;
loss rate of phytoplankton to bottom sediments (mg/L-d),
see (69);
loss of macrophytes due to breakage (g/m2-d), see (88);
nonpredatory mortality of given organism (mg/L-d), see
(66), (87), and (112);
fraction of dead organism that is labile (unitless);
loss rate for gametes (g/m3-d), see (126);
fraction of dead organism that is refractory (unitless);
rate  of loss  of given  toxicant,  suspended detritus  or
organism  due  to being carried downstream  (mg/L-d), see
(16), (71), (72), (130), and (131);
rate  of settling  loss  to hypolimnion  from  epilimnion
(mg/L-d).  May  be  positive  or negative  depending  on
segment being simulated,  see (69);
rate of ingestion of given food or  prey by  given  predator
(mg/L-d),see(91);
predatory  mortality by  given  predator on given  prey
(mg/L-d), see (99);
toxicant  excretion  from plants  to  dissolved  organics
(mg/L-d);
excretion rate for given organism  (g/m3-d), see (64), (111);
rate of transfer of phytoplankton  to hypolimnion (mg/L-d).
May be positive or negative depending  on  segment being
modeled, see (69);
rate of sorption by algae (ug/L - d),  see (360);
rate of sorption by macrophytes (ug/L - d), see (356);
rate of absorption of toxicant by the gills (ug/L - d), see
(365);
                                          230

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8

       DietUptakeprey      =      rate of dietary absorption of toxicant associated with given
                                  prey (ug/L-d), see (369);
       Recruit             =      biomass gained  from  successful  spawning  (g/m3-d),  see
                                  (128);
       Promotion          =      promotion from  one age class  to the next (mg/L-d),  see
                                  (136);
       Migration          =      rate of migration (g/m3-d), see (133); and
       Emergelnsect       =      insect emergence (mg/L-d), see (137).
The concentration in each carrier is given by:


                                PPB<=   ToxState>  -7e6                           (310)
                                       Carrier State,
where:
       PPBi         =      concentration of chemical in carrier /' (ug/kg);
       ToxStatei     =      mass of chemical in carrier /' (ug/L);
       CarrierState  =      biomass of carrier (mg/L); and
       Ie6          =      conversion factor (mg/kg).

8.1 lonization

Dissociation of an organic acid or base in water can have a significant effect on its environmental
properties.  In particular, solubility, volatilization, photolysis, sorption, and bioconcentration of
an ionized compound can be affected.  Rather than modeling ionization products, the approach
taken in AQUATOX is to  represent  the modifications to the fate and transport of the neutral
species, based on the fraction that  is not dissociated.  The acid dissociation constant is a measure
of the strength of the acid or base,  and is expressed as the negative \og,pKa, and the fraction that
is not ionized is:

                                Nondissoc =	                            (311)
                                            1+ jn(pH~pKa)

where:
       Nondissoc    =      nondissociated fraction (unitless).

If the compound  is a base then the  fraction not ionized is:


                                Nondissoc =	                            (312)
                                            J+jQ(pKa-pH)                            ^   '

Note: IfpKa is set to zero then ionization is ignored (i.e. NonDissoc is set to 1.0).

When pKa = pH  half the compound is ionized and half is  not (Figure 132).  At ambient
environmental pH values, compounds with  a pKa in the  range of 4 to 9 will exhibit  significant
dissociation (Figure 133).

                                          231

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                       CHAPTER 8
Figure 132. Dissociation of pentachlorophenol
(pKa = 4.75) at higher ph values
1
TJ
•s n R
lo.e
73
c
0
z n A
c
o
^3 n o
ro u-^
ul
0
2










"~\
t
\



\
\
\
1











\
\
\
\









^__^































4 6 8 10
pH
                        Figure 133.  Dissociation as a function ofpKa at
                        an ambient pH of 1
•o
(U
ra 0.8
'o
c
o
~z. n 4
c
o
|0.2
0
0




























































^y





/

/
/


7
/
/






f




























































2 4 6 8 10 12 14
pka
8.2 Hydrolysis

Hydrolysis is the degradation of a compound through reaction with water.  During hydrolysis,
both a pollutant molecule and a water molecule are split, and the two water molecule fragments
(H+ and OH") join to the two pollutant fragments to form new chemicals.  Neutral and acid- and
base-catalyzed hydrolysis are modeled using the approach of Mabey and Mill (1978) in which an
overall pseudo-first-order rate constant is computed  for a given pH, adjusted for the  ambient
temperature of the water:
                             Hydrolysis = KHyd • Toxicant
                                                        Phase
                                                              (313)
where:
and where:
       KHyd
       KAcidExp

       KBaseExp

       KUncat
       Arrhen
KHyd = (KAcidExp + KBaseExp + KUncat) • Arrhen                (314)

 =     overall pseudo-first-order rate constant for a given pH and
       temperature (1/d);
 =     pseudo-first-order acid-catalyzed rate constant for a given pH
       (1/d);
 =     pseudo-first-order base-catalyzed rate constant for a given pH
       (1/d);
 =     the measured first-order reaction rate at pH 7 (1/d); and
 =     temperature adjustment (unitless), see (319).
In neutral  hydrolysis reactions, the pollutant reacts  with a water molecule  (H2O)  and the
concentration of water is usually included in KUncat. In acid-catalyzed hydrolysis, the hydrogen
ion reacts with the pollutant, and a first-order decay rate  for a given pH can be estimated as
follows:
                                          232

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                        CHAPTER 8
                               KAcidExp = KAcid • Hlon
where:
and where:
      KAcid
      Hlon
      PH
                                                               (315)
                                                                                 (316)
        acid-catalyzed rate constant (L/mol-d);
        concentration of hydrogen ions (mol/L); and
        pH of water column.
Likewise for base-catalyzed hydrolysis, the first-order rate constant for a reaction between the
hydroxide ion and the pollutant at a given pH (Figure 134) can be described as:
                               KBaseExp = KBase • OHIon
where:

and where:
      XBase
      OHIon
                                                               (317)
                                                               (318)
  =     base-catalyzed rate constant (L/mol • d); and
  =     concentration of hydroxide ions (mol/L).

Figure 134. Base-catalyzed hydrolysis of pentachlorophenol
6.05E-03
6.00E-03
Q 5.95E-03
j-U c Qf)p no
OH
2





































































_^S


|
I
/
/
/
7
/



4 6 8 10
PH
Hydrolysis reaction rates are adjusted for the temperature of the waterbody being modeled by
using the Arrhenius rate law (Hemond and Fechner 1994).  An activation energy value of 18,000
cal/mol (a mid-range value for organic chemicals) is used as a default:
                                            En
                                                   En
where:
      En
      R
                               Arrhen =
        Arrhenius activation energy (cal/mol);
        universal gas constant (cal/mol • Kelvin);
                                                               (319)
                                         233

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


       KelvinT      =      temperature for which rate constant is to be predicted (Kelvin); and
       TObs        =      temperature at which known rate constant was measured (Kelvin).


8.3 Photolysis

Direct photolysis  is the  process  by  which  a compound  absorbs  light and undergoes
transformation:
                             Photolysis = KPhot • Toxicant Phase                        (320)
where:
       Photolysis    =      rate of loss due to photodegradation (ug/L-d); and
       KPhot       =      direct photolysis first-order rate constant (1/day).

For consistency, photolysis is  computed for both  the epilimnion and hypolimnion in  stratified
systems.   However,  photolysis is not a significant  factor at hypolimnetic depths and is also
ignored in sediments.

lonization may result in a significant shift in  the  absorption of light  (Lyman et al.,  1982;
Schwarzenbach et al., 1993). However, there is  a  general  absence of information on the effects
of light on ionized species.  The user provides an observed half-life for  photolysis, and this is
usually determined either with distilled water or with water from a representative site, so that
ionization may be included in the calculated lumped parameter KPhot.

Based on the approach of Thomann and Mueller  (1987; see also Schwarzenbach et al. 1993), the
observed first-order rate constant for the compound is modified by a light attenuation factor for
ultraviolet light so that the process as represented  is depth-sensitive  (Figure 135); it also is
adjusted by a factor for time-varying light:

                     KPhot = PhotRate • ScreeningFactor • LightFactor                 (321)
where:
       PhotRate     =      direct, observed photolysis first-order rate constant (I/day);
       ScreeningFactor =    a light screening factor (unitless), see (322); and
       LightFactor   =      a time-varying light factor (unitless), see (323).
                                          234

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 8
                  Figure 135. Photolysis of pentachlorophenol as a function of
                  	light intensity and depth of water	
                      100    200    300    400    500    600
                                   LIGHT INTENSITY (ly/d)

                          	DEPTH(m)	
                                          700
                         — 0.5    1   —1.5 — 2
                                2.5
A light screening factor adjusts the observed laboratory photolytic transformation rate of a given
pollutant for field conditions with variable light attenuation and depth (Thomann and Mueller,
1987):
                                        n  77-v •  ,    7     (- Extinct • Thick)
                      „     .   „        RadDistr  7-exp
                      ScreemngFactor =	
                                       RadDistrO   Extinct • Thick
                                                       (322)
where:
       RadDistr


       RadDistrO


       Extinct

       Thick
radiance  distribution function, which is the  ratio of the average
pathlength to the depth (see Schwarzenbach et al., 1993) (taken to
be 1.6, unitless);
radiance  distribution function for the top of the segment (taken to
be 1.2 for the top of the epilimnion and 1.6 for the  top  of the
hypolimnion, unitless);
light extinction coefficient (1/m) not including periphyton,  see
(40);
thickness of the water body segment  if stratified or maximum
depth if unstratified (m).
The equation presented above implicitly makes the following assumptions:

       quantum yield is independent of wavelength; and,
   •   the value used for PhotRate is a representative near-surface, first-order rate constant for
       direct photolysis.
                                          235

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 8


The rate is modified further to represent seasonally varying light conditions and the effect of ice
cover:
                                 LightFactor=                                     (323)
                                              Ave Solar
where:
       Solar 0    =  time-varying average light intensity at the top of the segment (ly/day); and
       AveSolar  =  average light intensity for late  spring or early summer, corresponding to
                    time when photolytic half-life is often measured (default = 500 Ly/day).

If the system is unstratified or if the epilimnion is being modeled, the light intensity is the light
loading:

                                     SolarO = Solar                                (324)
otherwise we are interested in the intensity at the top of the hypolimnion and the attenuation of
light is given as a logarithmic decrease over the thickness of the epilimnion:

                              o 7  n  o 7       (-Alpha • MaxZMix)                        SI^E\
                              SolarO = Solar • exp                                   (325)
where:
       Solar        =     incident solar radiation loading (ly/d), see (25); and
       MaxZMix    =     depth of the mixing zone (m), see (17).

Because  the ultraviolet  light intensity exhibits greater  seasonal variation than the  visible
spectrum  (Lyman  et al., 1982), decreasing markedly when the angle  of  the sun  is low,  this
construct  could  predict higher rates  of photolysis in the winter than might actually  occur.
However, the model also accounts for significant attenuation of light due to ice cover (see
section 3.6) so that photolysis, as modeled, is  not an important process in northern waters in the
winter.

8.4 Microbial Degradation

Not only can microorganisms decompose  the detrital organic material in ecosystems, they also
can degrade xenobiotic organic compounds  such  as fuels,  solvents, and  pesticides  to  obtain
energy. In AQUATOX this process of biodegradation of pollutants, whether they are dissolved
in the water column or adsorbed to organic detritus in the water column or sediments, is modeled
using the same  equations as for  decomposition of detritus,  substituting the pollutant  and its
degradation parameters for detritus in Equation (159) and supporting equations:

             MicrobialDegrdn = KMDegrdnphase • DOCorrection • TCorr • pHCorr
                                                                                   (326)
                                      •  Toxicant phase
where:
       MicrobialDegrdn  =       loss due to microbial degradation (g/m3-d);
       KMDegrdn        =       maximum aerobic  microbial  degradation  rate,  either in
                                  water column or sediments  (1/d),   in sediments this is
                                  assumed to be four times the user-entered value for water;

                                          236

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


       DOCorrection     =        effect of anaerobic conditions (unitless), see (160);
       TCorr             =        effect of suboptimal temperature (unitless), see (59);
       pHCorr           =        effect of suboptimal pH (unitless), see (162); and
       Toxicant          =        concentration of organic toxicant (g/m3).

Microbial degradation  of toxicants  proceeds  more quickly if the material is associated with
surficial or paniculate sediments rather than dissolved in the water column (Godshalk and Barko,
1985);   thus,  in calculating the  loss due to microbial  degradation in the sorbed phase, the
maximum degradation rate is converted by the model to four times the user entered maximum
chemical degradation rate in  the water (Max. Rate of Aerobic Microbial Degradation).   The
model assumes that reported maximum microbial degradation rates are for the dissolved phase; if
the reported degradation value is from a study with additional organic matter, such as suspended
slurry  or wet soil samples, then the parameter value that is entered should be one-fourth that
reported.
8.5 Volatilization

Volatilization is modeled using the "stagnant boundary theory", or two-film model, in which a
pollutant molecule must diffuse across both a stagnant water layer and a stagnant air layer to
volatilize  out of a  waterbody (Whitman, 1923; Liss and  Slater,  1974).  Diffusion rates of
pollutants in these  stagnant boundary  layers  can be related to the known diffusion rates of
chemicals such  as oxygen and water vapor. The thickness of the stagnant boundary layers must
also  be taken into  account  to estimate the volatile  flux of a chemical  out  of  (or into)  the
waterbody.

The time required for a pollutant to diffuse through the stagnant water layer in a  waterbody is
based on the well-established equations for the reaeration of oxygen, corrected for the difference
in diffusivity as indicated by the respective molecular weights (Thomann and Mueller, 1987, p.
533).  The diffusivity through the water film is greatly enhanced  by the degree  of ionization
(Schwarzenbach et al., 1993, p. 243), and the depth-averaged reaeration coefficient is multiplied
by the thickness of the well-mixed zone:

                                       (           2   Y2i      i
                  KLiq = KReaer • Thick • MolWtO —-—	             (327)
                                       ^        MolWt)    Nondissoc
where:
       KLiq        =      water-side transfer velocity (m/d);
       KReaer      =      depth-averaged reaeration coefficient  for oxygen  (1/d), see (191)-
                           (195);
       Thick       =      thickness  of  the water body segment if stratified or maximum
                           depth if unstratified (m);
       MolWtO2    =      molecular weight of oxygen (g/mol, =32);
       MolWt      =      molecular weight of pollutant (g/mol); and
       Nondissoc   =      nondissociated fraction (unitless), see (311).
                                          237

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 8


Likewise, the thickness of the air-side stagnant boundary layer is also affected by wind.  Wind
usually is measured at 10m, and laboratory experiments are based on wind measured at 10 cm,
so a conversion is  necessary (Banks, 1975).   To estimate the air-side transfer velocity of a
pollutant, we used the following empirical equation based on the evaporation of water, corrected
for the difference in diffusivity of water vapor compared to the toxicant (Thomann and Mueller,
1987, p. 534):
                                  (            o
                      KGas = 168-\ MolWtH2 — - —    • Wind • 0. 5                  (328)
                                  I,         MolWt)
where:
       KGas        =      air-side transfer velocity (m/d);
       Wind        =      wind speed ten meters above the water surface (m/s);
       0.5           =      conversion factor (wind at 10 cm/wind at 10 m); and
       MolWtH2O   =      molecular weight of water (g/mol, =18).

The total resistance to the mass transfer of the pollutant through  both the stagnant boundary
layers can be expressed as the sum of the resistances- the reciprocals of the air- and water-phase
mass transfer coefficients  (Schwarzenbach et al., 1993), modified for the effects of ionization:
                                                                                  (329)
                      KOVol  KLiq  KGas • HenryLaw • Nondissoc

where:
       KOVol  =    total mass transfer coefficient through both stagnant boundary  layers
(m/d);

                                      Henry • HLCSaltFactor                      ,~~n^
                           HenryLaw =	                      (330)
                                            R-TKelvin
and where:
       HenryLaw    =      Henry's law constant (unitlessj;
       Henry        =      Henry's law constant (atm m3 mol"1);
       HLCSaltFactor =     Correction factor for effect of salinity (unitless), see (444).
       R            =      gas constant (=8.2C
       TKelvin      =      temperature in °K.
R            =     gas constant (=8.206E-5 atm m3 (mol K)"1); and
The Henry's law constant is applicable only to the fraction that is nondissociated because the
ionized species will not be present in the gas phase (Schwarzenbach et al., 1993, p. 179).

The atmospheric exchange of the pollutant can be expressed as the depth-averaged total mass
transfer coefficient times the difference between the concentration of the chemical and the
saturation concentration:


                      Volatilization = ———— • (ToxSat - Toxicantwater)                 (331)
                                      Thick

                                          238

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


where:
       Volatilization   =    interchange with atmosphere (ug/L-d);
       Thick          =    depth of water or thickness of surface layer (m);
       ToxSat         =    saturation concentration of pollutant in  equilibrium with the gas
                           phase (ug/L), see (332); and
       Toxicantwater    =    concentration of pollutant in water (ug/L).

The saturation concentration depends on the concentration of the pollutant in the air, ignoring
temperature effects (Thomann and Mueller, 1987, p. 532; see also Schnoor, 1996), but adjusting
for ionization and units:
                           ToxSat =	Toxicantair	J00()                      (332)
                                   HenryLaw • Nondissoc
where:
       Toxicantair    =     gas-phase concentration of the pollutant (g/m3); and
       Nondissoc     =     nondissociated fraction (unitless).

Theoretically,  toxicants can be transferred  in  either direction across the water-air interface.
Often the pollutant can be assumed to have a negligible concentration in  the air and ToxSat is
zero.  However, this  general construct can  represent the transferral of volatile pollutants into
water bodies. Volatilization might become negative if toxicant  concentrations are high in the air,
and concentrations in the  water column may increase as a result of this interchange.  Because
ionized species do not volatilize, the saturation level increases if ionization is occurring.

The nondimensional Henry's law constant, which relates the concentration  of a compound in the
air phase to its concentration  in the  water phase,  strongly  affects  the  air-phase resistance.
Depending on the value of the Henry's law constant, the water phase, the air phase or both may
control volatilization.   For example, with a depth of  1 m and a wind of 1  m/s, the gas phase is
100,000 times as important as the water phase for atrazine (Henry's law constant = 3.0E-9), but
the water phase  is 50 times as  important as the air phase for benzene (Henry's law constant =
5.5E-3).   Volatilization of atrazine  exhibits a linear  relationship with wind (Figure 136) in
contrast to the exponential relationship exhibited by benzene (Figure 137).
                                           239

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                               CHAPTER 8
 Figure 136.
 Wind
Atrazine KOVol as a function of
Figure 137.  Benzene KOVol as a function of
Wind
         VOLATILIZATION OF ATRAZINE

       4E-05
     3.5E-05
       3E-05
               3.5  7
       10.5 14 17.5 21 24.5 28
        WIND (m/s)
                                          VOLATILIZATION OF BENZENE
                                                          3  6
                                                  9  12 15  18  21  24 27 30
                                                     WIND (m/s)
                                                            —AQUATOX
                                                              Schwarzenbach et al., 1993
8.6 Partition Coefficients

Although AQUATOX is a kinetic model, steady-state partition coefficients for organic pollutants
are computed in order to place constraints on competitive uptake and loss processes in detritus
and plants,  speeding up computations.   Bioconcentration factors also are used in computing
internal toxicity in plants and animals. They are estimated from empirical regression equations
and the pollutant's octanol-water partition coefficient.

Detritus

Natural organic matter is  the primary  sorbent for neutral  organic pollutants.   Hydrophobic
chemicals partition primarily in nonpolar organic matter (Abbott et al. 1995).  Refractory detritus
is  relatively nonpolar; its partition coefficient (in the non-dissolved phase) is a function of the
octanol-water partition coefficient (N = 34, r2 = 0.93; Schwarzenbach et al. 1993):
                                                       0.82
                                                                                    (333)
where:
       KOW
              detritus-water partition coefficient (L/kg); and
              octanol-water partition coefficient (unitless).
Detritus in sediments is simulated separately from inorganic sediments, rather than as a fraction
of the sediments as in  other models.  When the multi-layer sediment model is not included,
refractory detritus is used as a surrogate for sediments in general; and the sediment partition
coefficient KPSed, which can be entered manually by the user, is the same as KOMRefrDetr.

Equation (334) and the equations that follow are extended to polar compounds, following the
approach of Smejtek and Wang (1993):
                                           240

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
                                efroetr = 1.38- KOW0'82 • Nondissoc
                           (1- Nondissoc) • lonCorr -1.38- KOW°'82
                                                        (334)
where:
       Nondissoc
       lonCorr
un-ionized fraction (unitless); and
correction factor for decreased sorption, 0.01 for chemicals that are
bases and 0.1 for acids, (unitless).
Using pentachorophenol as a test compound, and comparing it to octanol, the influence of pH-
mediated dissociation is seen in Figure 138.  This relationship is verified by comparison with the
results of Smejtek and  Wang (1993) using egg membrane. However, in the general model Eq.
(334) is used for refractory detrital sediments as well.

Figure 138.  Refractory detritus-water and octanol-water partition coefficients for pentachlorophenol as a
                                      function of pH
1FR
-I cc; s

-r- "1 F4
O
^ 1F^i
1 P9
•1 P1






























^
i~_^
"~i



\,
"~~~~~|



*\
^




^^






\
N





N






s

3456789
PH
-*- PCP KOM for refractory detritus
-^- Un-ionized PCP - octanol/water
There appears to be a dichotomy in partitioning; data in the literature suggest that labile detritus
does not take up hydrophobic compounds as rapidly as refractory detritus. Algal cell membranes
contain  polar  lipids,  and it  is likely that this polarity is retained  in the  early stages  of
decomposition.  KOC does not remain the same upon aging, death, and decomposition, probably
because of polarity changes. In an experiment using fresh and aged algal detritus, there was a
100% increase  in KOC with aging (Koelmans et al., 1995).   KOC increased as the C/N  ratio
increased, indicating that the material was  becoming more refractory. In another  study, KOC
doubled between day 2 and day 34, probably due to deeper penetration into the organic matrix
and lower polarity (Cornelissen et al., 1997).

Polar substrates increase  the  pKa of  the  compound (Smejtek and  Wang,  1993).  This  is
represented in the model by lowering the pH of polar particulate material by one pH unit, which
changes the dissociation accordingly.

The partition equation for labile detritus (non-dissolved) is based  on a study by Koelmans et  al.
(1995) using fresh algal detritus (N = 3, r2 = 1.0):
                                          241

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8
                               KOC LAP* = 23.44 • KOW0'61                          (335)
In the  model,  the equation is generalized to polar compounds and transformed to an organic
matter partition coefficient:

                         KOMiaboetr = (23. 44 • KOWaa • Nondissoc
                     + (1 - Nondissoc) • lonCorr • 23. 44 • KOW°'61 ) • 0.526

where:
                    =      partition coefficient for labile particulate organic carbon (L/kg);
                    =      partition coefficient for labile detritus (L/kg);
       lonCorr      =      correction factor for decreased sorption, 0.01 for chemicals that are
                           bases and 0.1 for acids, (unitless); and
       0.526        =      conversion from KOC to KOM (g OC/g OM).

O'Connor and Connolly (1980;  see also Ambrose et al., 1991) found that the sediment partition
coefficient is the inverse of the mass of suspended sediment, and Di Toro (1985) developed a
construct to represent the relationship. However, AQUATOX models partitioning directly to
organic detritus and ignores inorganic sediments, which are seldom involved directly in sorption
of neutral organic pollutants.  Therefore, the partition coefficient is not corrected for mass of
sediment.

Association  of hydrophobic compounds  with colloidal and dissolved  organic matter (DOM)
reduces bioavailability; such contaminants are unavailable for uptake by organisms (Stange and
Swackhamer 1994, Gilek et al.  1996).  Therefore, it is imperative that complexation of organic
chemicals with DOM be modeled correctly. In particular, contradictory research results can be
reconciled by considering that DOM is not homogeneous.  For instance, refractory humic acids,
derived from decomposition of terrestrial  and wetland organic material,  are quite different from
labile exudates from algae and other indigenous organisms.

Humic acids exhibit high polarity and do not readily complex  neutral compounds.  Natural
humic  acids from a Finnish lake with extensive marshes were spiked with  a PCB, but  a PCB-
humic  acid complex could not be demonstrated (Maaret et al.  1992). In another study, Freidig et
al. (1998) used artificially prepared Aldrich humic acid to determine a humic acid-DOC partition
coefficient (n = 5, r2, = 0.80), although they cautioned about extrapolation to the field. Landrum
et al. (1984) found that KOC values for natural dissolved organic matter were approximately one
order of magnitude less than for Aldrich humic acids (Gobas and Zhang 1994);  incorporating
that factor into the equation of Freidig et al. (1998) yields:
                                                        7                          (337)
where:
                           refractory dissolved organic carbon partition coefficient (L/kg).
Until a better relationship is found, we are using a generalization of this equation to include polar
compounds, transformed from organic carbon to organic matter, in AQUATOX:

                                          242

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8
                         KOMRefrDOM = (2.88 • KOWa67 • Nondissoc
                                                                                  (338)
                     + (1 - Nondissoc) • lonCorr • 2.58 • KOW°'6'') • 0.526
where:
       KOMRefrDoM  =      refractory dissolved organic matter partition coefficient (L/kg).

Algae

Nonpolar lipids in algae occur in the cell  contents, and it is likely that they constitute part of the
labile dissolved exudate, which may be both excreted and lysed material. Therefore, the stronger
relationship reported by Koelmans and Heugens (1998) for partitioning to algal exudate (n = 6, r2
= 0.926) is:

                                KOCLabDoc = 0.88- KOW                            (339)

which we also generalized for polar compounds and transformed:

                          KOMLabooM = (0.88 • KOW • Nondissoc
                       + (1- Nondissoc
where:
                                                                                  (340)
                          - Nondissoc) • lonCorr • 0.88 • KOW) • 0.526
       KOCiabDOC   =      partition coefficient for labile dissolved organic carbon (L/kg); and
                    =      partition coefficient for labile dissolved organic matter (L/kg).
Unfortunately, older data and  modeling efforts failed to  distinguish between hydrophobic
compounds that were truly dissolved and those that were complexed with DOM. For example,
the PCB water concentrations for Lake Ontario, reported by Oliver and Niimi (1988) and used by
many subsequent researchers, included both dissolved and DOC-complexed PCBs (a fact which
they recognized).  In their steady-state model of PCBs in the Great Lakes, Thomann and Mueller
(1983) defined "dissolved" as that which is not particulate (passing a 0.45 micron filter).  In their
Hudson River PCB model, Thomann et al. (1991) again used an operational  definition of
dissolved PCBs.  AQUATOX distinguishes between truly dissolved and complexed compounds;
therefore, the partition coefficients calculated by AQUATOX may be larger than those used in
older studies.

Bioaccumulation  of  PCBs in  algae depends  on solubility,  hydrophobicity  and  molecular
configuration of the compound,  and growth rate, surface area and type, and content and type of
lipid in the alga (Stange and Swackhamer 1994).  Phytoplankton may double or triple in one day
and periphyton turnover may be  so rapid that some PCBs will not reach equilibrium (cf Hill and
Napolitano 1997).

Hydrophobic compounds partition to lipids in algae, but the relationship is not a simple one.
Phytoplankton lipids can range from 3 to 30% by weight (Swackhamer and Skoglund 1991), and
not all lipids are the same. Polar phospholipids occur on the surface. Hydrophobic compounds
preferentially partition to internal neutral lipids, but those are usually a minor fraction of the total
lipids,  and they vary depending on growth conditions and species (Stange and Swackhamer

                                          243

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


1994). Algal lipids have a much stronger affinity for hydrophobic compounds than does octanol,
so that the algal BCF^ > K0w (Stange and Swackhamer 1994, Koelmans et al. 1995,  Sijm et
al. 1998).
                                                                             r\
For algae, the approximation to estimate the dry-weight bioaccumulation factor (r = 0.87),
computed from Swackhamer and Skoglund's (1993) study of numerous PCB congeners, is:


                          \og(BCFAlga) = 0.41 + 0.91 • LogKOW                      (341)
where:
       BCFAiga       =      partition coefficient between algae and water (L/kg).

Rearranging and extending to hydrophilic and ionized compounds:

                                 w = 2.57 • KOW°'93 • Nondissoc
                                                                                 \   )
                        + (1 - Nondissoc) • lonCorr -0.257- KOW°'93

Comparing the results of using these coefficients, we see that they are consistent with the relative
importance of the various  substrates  in  binding  organic chemicals (Figure 140).   Binding
capacity of detritus is greater than dissolved organic matter in Great Lakes waters (Stange and
Swackhamer 1994, Gilek et al. 1996).   In a study using Baltic Sea water, less than 7% PCBs
were associated with dissolved organic matter and  most were associated with algae (Bjork and
Gilek 1999).   In  contrast, in  a study using algal  exudate and a PCB, 98% of the  dissolved
concentration was as  a  dissolved organic  matter complex and only 2%  was bioavailable
(Koelmans andHeugens 1998).

The influence of substrate polarity is evident in Figure 139, which shows the effect of ionization
on binding of pentachlorophenol to various types of organic matter.  The polar substrates, such
as algal detritus, have an inflection point which is one pH unit higher than that of nonpolar
substrates, such as refractory detritus. The relative importance  of the substrates for binding is
also demonstrated quite clearly.

Macrophytes

For macrophytes, an empirical relationship reported by Gobas et al. (1991) for 9 chemicals with
LogKOWs of 4 to 8.3 (r2 = 0.97) is used:

                          log(BCFMacJ = 0.98- LogKOW - 2.24                     (343)
Again, rearranging and extending to hydrophilic and ionized compounds:

                      BCF Macro = 0.00575 • KOW0'98 • (Nondissoc + 0.2)                 (344)
                                         244

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                               CHAPTER 8
Invertebrates

For  the  invertebrate bioconcentration factor,  the following  empirical  equation is used  for
nondetritivores,  based  on  7  chemicals  with  LogKOWs  ranging from  3.3  to 6.2  and
bioconcentration factors for Daphniapulex (r = 0.85; Southworth et al., 1978; see also Lyman et
al., 1982), converted to dry weight:
                     ( BCF vertebra J = (0.7520 • LogKOW - 0.4362) • WetToDry
                                       (345)
where:
       BCF'invertebrate  =      partition coefficient between invertebrates and water (L/kg); and
       WetToDry    =      wet to dry conversion factor (unitless, default = 5).
Extending and generalizing to ionized compounds:
                     BCF vertebrate = 0.3663 • ROW*7520 • (Nondissoc + 0.01)

For invertebrates that are detritivores the following equation is used, based on Gobas 1993:
                                       (346)
where:
                                FracLipid                ,.T   ,.       n n,\
                 BCF Vertebrate =	~	KOM RefrDetr ' (NondlSSOC + 0. 01)
                               FmcOCoetntus
                                       (347)
       BCFinvertebrate  =   partition coefficient between invertebrates and water (L/kg);
       FracLipid    =   fraction of lipid within the organism;
       FracOCDetntus =   fraction of organic carbon in detritus (= 0.526);
       KOMxefrDetr   =   partition coefficient for refractory sediment detritus (L/kg), see (334).
Figure 139.  Partitioning  to Various Types of
Organic Matter as Function of Kow
         1E10-
          1E9;
          1E8;

          1E6"

          1E4"
          1E3'
          1E2 '
                     5   6    1
                       Log KOW
                                  89
       - humie acids  - algae        exudate
       algal detritus * refr. detritus  - sediments
Figure 140.  Partitioning  to Various Types of
Organic Matter as a Function of pH
        1E6


        1E5*  "  '— --, _


        1E4*  "" "" ""  „   - "'  .  """" """ * ""


        1E3
          3456789
                        PH


     - humie acids  - algae        octanol/water
      exudate     * algal detritus  - refr. detritus
Fish

Fish take longer to reach equilibrium with the surrounding water; therefore, a nonequilibrium
bioconcentration factor is used.  For each pollutant, a whole-fish bioconcentration factor is based
                                            245

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
on the lipid content of the fish extended to hydrophilic chemicals (McCarty et al., 1992), with
provision for ionization:
where:
       Lipid
       WetToDry
                        sh = Lipid • WetToDry • KOW • (Nondissoc + 0.01)
partition coefficient between whole fish and water (L/kg);
fraction offish that is lipid (g lipid/g fish); and
wet to dry conversion factor (unitless, default = 5).
                                                        (348)
The bioconcentration factor is adjusted for the time to reach equilibrium as  a function of the
clearance or elimination rate and the time of exposure (Hawker and Connell, 1985; Connell and
Hawker, 1988; Figure 141):
                          BCFFlsh = KBFlsh • (J - e^atlon.TElapse^                      (349)
where:
       BCFpish      =     quasi-equilibrium bioconcentration factor for fish (L/kg);
       TElapsed     =     time elapsed since fish was first exposed (d); and
       Depuration   =     clearance, which may include biotransformation, see (372)  (1/d).

                   Figure 141. Bioconcentration factor for fish as a function
                                   of time and log KOW
1E7 I 	 =

^
Q
i= 1E6=,,
"(
o:
i— /
u 1ESQ
0 IP
-^ [/
£
o
0 1E4J
O


1 F^ r



^--

^^*
/










loq
f^—*
^t*

log





Inn
log



K( )

— j
^

Ku





k"o




A/ =



W -





A/ —
V V —



6
— —


8





Q
\J




-— "














•* P"














P« •*














^^














^^^


























0 200 400 600 800 1000 1200
DAY
8.7 Nonequilibrium Kinetics

Often there is an absence of equilibrium due to growth or insufficient exposure time, metabolic
biotransformation,   dietary  exposure,  and nonlinear  relationships  for very  large and/or
superhydrophobic compounds (Bertelsen et al. 1998).  Although it is important to have a
knowledge of equilibrium partitioning because it is an indication of the condition toward which
systems tend (Bertelsen et al. 1998), it is often impossible to determine steady-state potential due
                                          246

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
to changes in bioavailability and physiology (Landrum 1998). For example, PCBs may not be at
steady state even in large systems such as Lake Ontario that have been polluted over a long
period of time.  In  fact, PCBs in Lake Ontario  exhibit a 25-fold disequilibrium (Cook  and
Burkhard 1998).  The challenge is to obtain sufficient data for a kinetic model (Gobas et al.
1995).

Sorption and Desorption to Detritus

Partitioning to detritus appears to involve rapid sorption to particle surfaces, followed by slow
movement into,  and out of, organic matter and porous aggregates  (Karickhoff and Morris,
1985). Therefore attainment of equilibrium may be slow.   Because  of the need  to represent
sorption  and desorption  separately  in  detritus, kinetic formulations  are used (Thomann  and
Mueller,  1987), with provision for ionization:
                     Sorption = k jDetr • Toxicantwater • (Nondissoc + 0.01)
                             • Org2C • Detr • UptakeLimit • 7e - 6
                                                        (350)
                             Desorption = k 2Detr • Toxicant Detr
                                                       (351)
where:
       Sorption
       Nondissoc
       Toxicantwater
       Org2C

       Detr
       le-6
       Desorption
       UptakeLimit

       Toxicant DetT
rate of sorption to given detritus compartment (ug/L-d);
sorption rate constant (user-editable, default value of 1.39 L/kg-d),
see (355);
fraction not ionized (unitless), see (311);
concentration of toxicant in water (ug/L);
conversion factor  for organic matter to  carbon (= 0.526 g  C/g
organic matter);
mass of each of the detritus compartments per unit volume (mg/L);
units conversion (kg/mg);
rate  of desorption from  given sediment  detritus  compartment
(ug/L-d);
desorption rate constant (1/d), see (354);
factor to limit uptake as equilibrium is reached (unitless) see (352);
and
mass of toxicant in each of the detritus compartments (ug/L).
In order to limit  sorption to detritus and  algae  as  equilibrium is reached,  UptakeLimit is
computed as:
                     UptakeLimit Carner - Toxicants kpCarner- PPBCamer
                                                        (352)
                                                          earner
where:
       UptakeLimit carrier =  factor to limit uptake as equilibrium is reached (unitless);
                                          247

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


       kpcamer          =  partition coefficient (KOM) or bioconcentration factor (BCF) for
                           each carrier (L/kg), see (333) to (342);
       PPBcamer        =  concentration of toxicant in each carrier (ug/kg), see (310).


Desorption of the detrital compartments is the reciprocal of the reaction time, which Karickhoff
and Morris (1985) found to be a linear function of the partition coefficient over three  orders of
magnitude (r2 = 0.87):

                                   — * 0.03-24-KOM                              (353)
                                   k2
So k2 is taken to be:


                                       k2 = -^-                                  (354)
                                           KOM
where:
       KOM        =      detritus-water partition coefficient (L/kg OM, see section 8.6); and
       24           =      conversion from hours to days.

Because the kinetic definition of the detrital partition coefficient^QMis:


                                       KOM = —                                  (355)
                                              k2

the sorption rate constant kl is set by the user (Kl Detritus). The default value is 1.39 L/kg-d.

Bioconcentration in Macrophytes and Algae

Macrophytes: As Gobas et al. (1991) have shown,  submerged aquatic macrophytes take up and
release organic chemicals over a measurable period of time at rates related to the octanol-water
partition coefficient.  Uptake and  elimination are modeled assuming that  the  chemical  is
transported through  both aqueous and  lipid phases in the plant, with  rate constants  using
empirical equations fit to observed data (Gobas et al., 1991), modified to account for ionization
effects (Figure 142, Figure 143):

                     Macro Uptake = kl • ToxicantWater' StVarPiant -le-6                 (356)

                                Depuration plant = k2 • Toxicant Piant                     (357)
                                                                                   (358)
                                  0. 0020 +
                                          KOW • Nondissoc
If the user selects to estimate the elimination rate constant based on KOW (see section 8.8), the
following equation is used:
                                          248

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 8
                          k2 = -
where:
       MacroUptake
       Depuration piant
       StVarpiant
       1 e-6
       Toxicantpiant
       kl
       k2
       KOW
       Nondissoc
                              1.58 + 0.000015 • KOW • Nondissoc
 =   uptake of toxicant by plant (ug/L-d);
 =   cl earance of toxi cant from pi ant (ug/L • d);
 =   biomass of given plant (mg/L);
 =   units conversion (kg/mg);
 =   mass of toxicant in plant (ug/L);
 =   sorption rate constant (L/kg-d);
 =   elimination rate constant (1/d).
 =   octanol-water partition coefficient (unitless); and
 =   fraction of un-ionized toxicant (unitless).

Figure 142. Uptake rate constant for macrophytes
           (after Gobas et al., 1991)
                                                            (359)
^nn
4nn
Qfin
:*:
9flfi
•i nn



















0 2



















	 i




i


/
7
^


/
/
•/
/
/


./
4

























_














4 6 8 10
Log KOW
— Predicted • Observed
                     Figure 143. Elimination rate constant for macrophytes
                                  (after Gobas et al., 1991)
n 7
n R
n ^
n A
2°-4
m
n 9
O-i


























































S'
\
\
\















\
,\
x
*x
\













B.
^•^J













i














02468
Log KOW
— Predicted • Observed















10
Algae: Aside from obvious structural differences, algae may have very high lipid content (20%
for Chlorella sp.  according to J0rgensen et al.,  1979) and  macrophytes have a very low lipid
                                          249

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


content (0.2% in Myriophyllum spicatum as observed by Gobas et al. (1991), which affect both
uptake and elimination of toxicants.  However, the approach used by Gobas et al. (1991) in
modeling bioaccumulation in macrophytes provides a useful guide to modeling kinetic uptake in
algae.

There is probably a two-step algal  bioaccumulation mechanism for hydrophobic compounds,
with rapid surface sorption of 40-90% within 24 hours and then a small, steady increase with
transfer to interior lipids for the duration  of the exposure (Swackhamer and Skoglund 1991).
Uptake increases with increase in the surface area of algae (Wang et al.  1997).  Therefore,  the
smaller the organism the larger the uptake rate constant (Sijm et al. 1998).  However, in small
phytoplankton,  such as  the nannoplankton that dominate the Great  lakes, a high surface to
volume ratio can increase sorption,  but  high growth  rates can  limit internal  contaminant
concentrations (Swackhamer and  Skoglund 1991). The combination of lipid content, surface
area, and growth rate results in species differences in bioaccumulation factors among algae
(Wood et al. 1997).  Uptake of toxicants is a function of the uptake  rate constant  and  the
concentration of toxicant truly dissolved in the water, and is constrained by competitive uptake
by other compartments; also, because it is fast, it is limited as it approaches equilibrium, similar
to sorption to detritus :

                 AlgalUptake = kl • UptakeLimit Alga • ToxState • Carrier • 1 e - 6            (360)
where:
       AlgalUptake    =    rate of sorption by algae (ug/L-d);
       kl              =    uptake rate constant (L/kg-d), see (362);
       UptakeLimitAiga =    factor  to limit uptake as equilibrium is reached  (unitless),  see
                           (352);
       ToxState       =    concentration of dissolved toxicant (ug/L);
       Carrier         =    biomass of algal compartment (mg/L); and
       le-6            =    conversion factor (kg/mg).

The kinetics of partitioning of toxicants to algae is based on studies on PCB congeners in The
Netherlands by  Koelmans, Sijm, and colleagues and at the University of Minnesota by Skoglund
and Swackhamer. Both groups found uptake to be very rapid.  Sijm et al. (1998) presented data
on several congeners that were  used in this  study to develop the following relationship  for
phytoplankton (Figure 144):


                        kl =	                    (361)
                            1.8 E- 6 + 1/(KOW • (Nondissoc + 0.01))

Because size-dependent passive transport is indicated (Sijm et al., 1998), uptake by periphyton is
set arbitrarily at ten percent of that for phytoplankton.

Depuration is modeled  as a linear function;  it  does not  include loss due to excretion  of
photosynthate with associated toxicant, which is modeled separately:

                                  Depuration = k2 • State                             (362)
where:
                                          250

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 8
       Depuration   =      elimination of toxicant (ug/L-d);
       State         =      concentration of toxicant associated with alga (ug/L); and
       k2           =      elimination rate constant (1/d).

As a simplifying assumption, the depuration rate for periphyton is assumed to be two orders of
magnitude less:
                               Depuration = k2 • State • 0.01
                                                             (363)
The  elimination rate in plants may be input in the toxicity record  by the user or it may be
estimated using the following  equation  based in part  on Skoglund  et  al.  (1996).   Unlike
Skoglund, this equation ignores surface sorption and recognizes that growth dilution is explicit in
AQUATOX (see Figure 145):
where:
       LFrac

       WetToDry
                            k2
                                           2.4E + 5
                              Algae
                                   (KOW • LFrac • WetToDry)
                                                             (364)
      desorption rate constant (1/d);
      fraction  lipid (wet weight), entered  in  the  "chemical toxicity'
      screen; and
      translation from wet to dry weight (user input).

Figure 144. Algal sorption rate constant as a function
       of octanol-water partition coefficient
FIT
600000
9-500000
I5 400000
~ 300000
-55 200000
< 100000
0
c
TO DATA OF SUM ET AL. 1998


























































/
V






/



•



/'
/
(








^-



































2 4 6 8 10
LOG KOW
• ObsK1 — PredKI
                                          251

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                     CHAPTER 8
                    Figure 145. Rate of elimination by algae as a function of
                              octanol-water partition coefficient
1.2
1
"O n Q
fN r, r-
^ 0.6
(IS
•2)0.4












.
\
\
\
\












V
\
\
X











~-« 	

























2 4 6 8 10
Log KOW
Bioaccumulation in Animals

Animals can absorb toxic organic chemicals directly from the water through their gills and from
contaminated food through their guts. Direct sorption onto the body is ignored as a simplifying
assumption in this version of the model.  Reduction of body  burdens of organic chemicals is
accomplished through excretion and biotransformation, which  are often  considered together as
empirically determined  elimination rates.   "Growth dilution" occurs when  growth of the
organism is faster than accumulation of the toxicant.  Gobas (1993) includes fecal egestion, but
in AQUATOX egestion is merely the amount ingested but not assimilated; it is accounted for
indirectly in DietUptake. However, fecal loss is important  as an input  to the detrital toxicant
pool, and it is considered later in that context.  Inclusion of mortality and promotion terms is
necessary  for  mass balance,  but emphasizes the fact that average concentrations are being
modeled for any particular compartment.

Gill Sorption: An important route of exposure is by active transport through the  gills (Macek et
al., 1977).  This is the route that has been measured so often  in bioconcentration experiments
with fish. As the organism respires, water is passed over the outer surface of the gill and blood is
moved past the inner surface.  The exchange of toxicant through the gill membrane is assumed to
be facilitated by  the  same mechanism as the uptake of oxygen, following the approach of
Fagerstrom and Asell (1973,  1975), Weininger (1978),  and Thomann and Mueller (1987; see
also Thomann, 1989).  Therefore, the uptake rate for each animal can be calculated as a function
of respiration (Leung,  1978; Park et al., 1980):
where:
       GillUptake
                     GillUptake = KUptake • ToxicantWater • FracWaterColumn
                       T.Tr   ,    WEffTox • Respiration • O2Biomass
                       KUptake = —	
                                         Oxygen-WEffO2
=   uptake of toxicant by gills (ug/L - d);

                   252
                                                            (365)


                                                            (366)

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


       KUptake         =   uptake rate (1/d);
       ToxicantWater     =   concentration of toxicant in water (ug/L);
       FracwaterCoiumn    =   fraction of organism in water column (unitless), differentiates from
                           pore-water uptake if the multi-layer sediment model is included;
       WEffTox         =   withdrawal efficiency for toxicant by gills (unitless), see (367);
       Respiration      =   respiration rate (mg biomass/L-d), see (100);
       O2Biomass      =   ratio of oxygen to organic matter (mg oxygen/mg biomass; 0.575);
       Oxygen          =   concentration of dissolved oxygen (mg oxygen/L), see (186); and
       WEffO2         =   withdrawal efficiency for oxygen (unitless, generally 0.62);

The oxygen uptake efficiency WEffO2 is assigned a constant value of 0.62 based on observations
of McKim  et al.  (1985).  The toxicant uptake efficiency, WEffTox,  can be expected to have a
sigmoidal relationship to the log  octanol-water partition coefficient based on aqueous and lipid
transport (Spacie  and Hamelink, 1982).   This is represented by  an inelegant  but reasonable,
piece-wise  fit (Figure  146) to the data of McKim et al. (1985) using  750-g  fish, corrected for
ionization:
                                 lfLogKOW 3.0 then
                     WEffTox = 0.1 + Nondissoc • (0.3 • LogKOW - 0.45)
                               If 3.0 < LogKOW < 6.0 then
                                                                                  (367)
                             WEffTox = 0.1 + Nondissoc • 0.45
                               If 6.0 < LogKOW < 8.0 then
                  WEffTox = 0.1 + Nondissoc • (0.45 - 0.23 • (LogKOW - 6.0))


                                 If LogKOW > 8.0 then
                                     WEffTox = 0.1
       where:
       LogKOW     =      log octanol-water partition coefficient (unitless); and
       Nondissoc    =      fraction of toxicant that is un-ionized (unitless), see (311).
                                          253

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 8
                   Figure 146. Piece-wise fit to observed toxicant uptake data;
                    	Modified from McKim et al., 1985	
                        80h
                                        345
                                            LOG KOW
lonization  decreases the uptake efficiency  (Figure 147).  This  same algorithm is used for
invertebrates.  Thomann (1989) has proposed a similar construct for these same data  and a
slightly different construct for small organisms, but the scatter in the  data  does not seem to
justify using two different constructs.

                       Figure 147. The Effect of Differing Fractions of Un-
                       ionized Chemical on Uptake Efficiency
0.6
0.5
fr
§0.4
'o
£ 0.3

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 8


       GillUptake      =   uptake of toxicant by gills (ug/LwaterCoi - d);
       Toxicant poreWater  =   concentration of toxicant in pore waters (ug/Lporewater);
       Volume 'poreWater   =   volume of pore water (Lporewater); and
       Volume watered    =   volume of water column (
Dietary Uptake: Hydrophobic chemicals usually bioaccumulate primarily through absorption
from   contaminated  food.     Persistent,   highly   hydrophobic   chemicals  demonstrate
biomagnification or increasing concentrations as they  are passed up the food chain from one
trophic level to another; therefore, dietary exposure can be quite important (Gobas et al.,  1993).
Uptake from contaminated prey can be  computed as (Thomann and Mueller, 1987;  Gobas,
1993):

                          DietUptakepmy = KDPrey • PPBPrey • 1 e - 6                     (369)
where:

                       KDprey = GutEffTox • GutEffRed • Ingestionprey                  (370)
and:
      DietUptake prey   =   uptake of toxicant from given prey (ug toxicant/L-d);
      KDprey          =   dietary uptake rate for given prey (mg prey/L-d);
      PPBprey         =   cone, of toxicant in given prey (ug toxicant/kg prey), see (310);
       1 e-6            =   units conversion (kg/mg);
       GutEfJTox       =   efficiency of sorption of toxicant from gut (unitless);
       GutEffRed       =   reduction in GutEffTox due to non-lethal effects, see (371) ; and
                       =   ingestion of given prey (mg prey/L-d), see (91).
Gobas (1993) presents an empirical equation  for estimating  GutEffTox as a function  of the
octanol-water partition coefficient. However, data published by Gobas et al. (1993) suggest that
there is no trend in efficiency between LogKOW 4.5 and 7.5 (Figure 148); this is to be expected
because the digestive system has evolved to assimilate  a wide  variety of organic molecules.
Therefore, the mean value of 0.62 is used in AQUATOX as a constant for small fish. Nichols et
al. (1998) demonstrated that uptake is more efficient in larger fish; therefore, a value of 0.92 is
used for large game fish because of their size. Invertebrates generally exhibit lower efficiencies;
Landrum and Robbins (1990) showed that values ranged from 0.42 to 0.24 for chemicals with
log KOWs from 4.4 to 6.7;  the  mean value of 0.35  is used for invertebrates in  AQUATOX.
These values cannot be edited at this time.  (Note, the PFA model uses a relationship to chain
length, see (403) and (404).)
                                          255

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 8
                  Figure 148. GutEffTox constant based on mean value for data
                                  from Gobas et al., 1993
>,
o 1
<0
'o
it n 7^
£j u./si
c
o
'•K r, c.
o

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


where:
       Depuration Animal      =     clearance rate (ug/L-d);
       k2                  =     elimination rate constant (1/d);
       Toxicant Animal        =     mass of toxicant in given animal (ug/L); and
       TCorr               =     correction for suboptimal temperature (unitless), see (59).

If the multi-layer sediment model is included,  the amount of depuration that goes to the water
column vs. the active layer of pore waters is determined by the user input  "Frac. in Water
Column" parameter.

Estimation of the elimination rate constant k2 is based on a slope related to  log KOW and an
intercept that is  a direct function of respiration, assuming an allometric relationship between
respiration and the weight of the animal (Thomann, 1989), and an inverse function of the lipid
content in a construct unique to AQUATOX:

If WetWt < 5 g then
                                                               WpiWiRB
               Log k2 = - 0.536 • Log Kow • Log NonDissoc + 0.065 •  n   n—          (373)
                                                              LipidFrac
else
                                                               WetWtRB
               Log k2 = - 0.536 • Log Kow • Log NonDissoc + 0.116 •  vv   vv—          (374)
                                                              LipidFrac
where
       KOW         =      octanol-water partition coefficient (unitless);
       NonDissoc    =      fraction of toxicant that is un-ionized (unitless), see (311);
       LipidFrac    =      fraction of lipid in organism (g lipid/g organism wet);
       WetWt       =      mean wet weight of organism (g);
       RB           =      allometric exponent for respiration (unitless).
                                         257

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
                 Figure 149. Depuration rate constants for invertebrates and fish
               based on AQUATOX "classic" formulation (equations 373 and 374)
                4

                3

                2


             O) 0
             o
             -1  -1-

                -2

                -3

                -4
                  1
                                K2 for Various Animals
            456
                Log KOW
                           Daphnia
                           10-gfish
                           Eel
                           Linear (Daphnia pred)
                           Diporeia
                        A Eel obs
                        ^—Linear (10-g fish pred)
                        ^—Linear (Diporeia pred)
In AQUATOX Release 3.1, an alternative k2 estimation procedure is available based on Barber
(2003):
                                  ,„   C-WetWf0'197
                                      LipidFrac • K
                                                                                 (374b)
                                                  ow
where
       C
       WetWt
       LipidFrac
       KOW
constant of 445 for fish and 890 for invertebrates;
mean wet weight of organism (g);
fraction of lipid in organism (g lipid/g organism wet);
octanol-water partition coefficient (unitless);
Barber's (2003) formulation is based on uptake rates divided by LipidFrac*Kow fas a surrogate
for BCF).  The uptake rate equation utilized is based on an allometric analysis of 517 data points,
though there is a high degree of uncertainty  in this relationship.   Figure  150 shows that the
AQUATOX and Barber formulations have different relationships between predicted elimination
rates and Kow.  Our testing suggests that some studies benefit from one uptake formulation and
some benefit from the  other;  however, at this point there is no general  guidance as to which
formulation to use in a given application.
                                          258

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                          CHAPTER 8
            Figure 150. k2 predictions by Log KQW for a lOOg fish with 5% lipid
              10.0000
               1.0000
              0.1000
            t; o.oioo
              0.0010
              0.0001
              0.0000
                                                                    •Barber K2

                                                                    •AQUATOX K2
Biotransformation:  Biotransformation can  cause  the  conversion of a toxicant to another
toxicant  or to  a harmless  daughter product  through  a  variety  of pathways.    Internal
biotransformation to given daughter products by plants and animals is modeled by means of
empirical rate constants provided by the user in the "Chemical Biotransformation" screen:
                 Biotransformation = Toxicant orgamsm • BioRateConstorgamsm,t,
                                                                irgamsm, tox
                                                 (375)
where
       Biotransformation   =
       BioRateConst       =
rate of conversion of chemical by given organism (ug/L d),
biotransformation  rate  constant  to  a  given  toxicant,
provided by user (I/day)
 with the model keeping track of both the loss and the gains to various daughter compartments.
 A  simplifying assumption of the model  is that biotransformation occurs at a constant rate
 throughout a simulation.

Biotransformation  also can take  place  as  a  consequence of  microbial decomposition.  The
percentage of microbial biotransformation  from and into  each of the organic chemicals in a
simulation can be specified, with different values for aerobic and anaerobic decomposition. The
amount of biotransformation into a given chemical can then be calculated as follows for aerobic
conditions:
         Biotransform
                     Microbln
 Microbial DegradnOTox • FracAerobic • Frac
                                                                           OrgTox
(376)
                                          259

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
and for anaerobic conditions:
       BiotramformMicrobIn = Z0rgToxMicrobialDegradn0rgTox • (1 - FracAerobic) • Frac0rgTOX  (377)
where:
       BiotransforniMicrob in

       MicrobialDegradn

       FracAerobic

       FracorgTox
   Biotransformation to  a given  organic  chemical in  a  given
   detrital compartment due to microbial decomposition (ug/L d);
   total microbial degradation of a different toxicant in this detrital
   compartment ((J,g/L d) see (326);
   fraction of the microbial degradation that is aerobic (unitless),
   see (378); and
   user input fraction of the organic toxicant that is transformed to
   the current organic  toxicant (inputs can differ depending  on
   whether the degradation is aerobic or anaerobic).
To calculate the fraction of microbial decomposition that is aerobic, the following equation is
used:
                              FracAerobic = -
                                                Factor
                                            DOCorrection
                                                        (378)
where:
       Factor       =
       DOCorrection =
Michaelis-Menten factor (unitless) see (161);
effect of oxygen on microbial decomposition (unitless) see (160).
Bioaccumulation Factor:   Customarily, bioaccumulation is expressed as a bioaccumulation
factor (BAF), which is the ratio of the concentration in the organism to that in the water. The
BAF can be expressed as  a wet-weight, dry-weight, or lipid-normalized basis (Gobas  and
Morrison 2000). In AQUATOX, the BAFs are output  as both wet-weight and wet-weight lipid-
normalized values.  The concentration in an organism is wet-weight, and the  lipid  fraction is
input by the user as a wet-weight value:
                        BAFLlpld=LogW
                                        PpB0rgamsm I Toxicantwater
                                               FracLipid
                        BAFWet=Logw(pPB0rgamsm/ToXicantWater)
                                                                                  (378b)
where:
       PPBorganism    =   concentration of toxicant in given animal (ug/kg wet);
       FracLipid     =   fraction of organism that is lipid (g lipid/g organism wet); and
       Toxicantwater   =   concentration of toxicant in water (ug/L);
                                          260

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 8


Linkages to Detrital Compartments

Toxicants are transferred from organismal to detrital  compartments  through defecation and
mortality. The amount transferred due to defecation is the unassimilated portion of the toxicant
that is ingested:

                     DefecationTox = Z(KEgestpre4Prey • PPBPrey • / e - 6)                 (379)


                         ed,Prey = (1 - GutEffTox • GutEffRed) • Ingestionpredprey             (380)
where::
      DefecationTox    =    rate of transfer of toxicant due to defecation (ug/L-d);
               ed^prey    =    fecal  egestion  rate for  given  prey  by  given  predator  (mg
                             prey/L-d);
                        =    concentration of toxicant in given prey (ug/kg), see (310);
       1 e-6             =    units conversion (kg/mg);
      GutEffTox        =    efficiency of sorption of toxicant from gut (unitless); and
      GutEffRed        =    reduction in GutEffTox due to non-lethal effects, see (371) ;
      Ingestionpred, prey  =    rate of ingestion of given prey by given predator (mg/L-d), see
                             (91)

The  amount  of toxicant transferred due to mortality  may be large; it  is a function of the
concentrations of toxicant in the dying organisms and the mortality rates:

                          MortTox = Zf 'Mortality Org • PPB0rg • Ie6)                     (381)
where:
      MortTox     =     rate of transfer of toxicant due to mortality (ug/L-d);
      Mortality Org  =     rate of mortality of given organism (mg/L-d), see  (66), (87)  and
                           (H2);
      PPBorg       =     concentration of toxicant in given organism (ug/kg), see (310); and
       1 e-6         =     units conversion (kg/mg).
8.8 Alternative Uptake Model: Entering BCFs, Kl, and K2

When performing bioaccumulation calculations, the default behavior of the AQUATOX model is
to allow the user to  enter elimination  rate constants  (K2) for  all plants and animals for  a
particular organic chemical.  K2 values may also be estimated based on the Log KOW of the
chemical. Uptake in plants is a function of Log KOW  while gill uptake in animals is a function of
respiration and chemical  uptake efficiency.  The AQUATOX default model works well for a
wide variety of bioaccumulative organic chemicals, but some chemicals that are subject to very
rapid uptake and depuration are not efficiently modeled using these relationships; the rapid rates
create stiff equations that require shorter time-steps for solution.  In addition,  because of the
rapid rates, the chemical does approach equilibrium quickly.
                                          261

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


For this reason, an alternative uptake model is provided to the user.  In the chemical toxicity
record, the user may enter two of the three factors defining uptake (BCF, Kl, K2) and the third
factor is calculated using the below relationship:
                                       R^Z7   Kl
                                       BCF = •
                                              K2
                                                                                  (382)
where:      BCF   =  bioconcentration factor (L/kg dry);
              Kl   =  uptake rate constant (L/kg dry day);
              K2   =  elimination rate constant (1/d).

Given these parameters, AQUATOX calculates uptake and depuration in plants and animals as
kinetic processes.

                          Uptake = Kl • ToxState • Biomass • 1 e - 6                     (383}

                               Depuration = K2 • ToxState                          (384}

where:     Uptake   =  uptake rate within organism (|J,g/L day);
              Kl   =  uptake rate constant (L/kg dry day);
         ToxState   =  concentration of toxicant in organism in water (ng/L)
         Biomass   =  concentration organism in water (mg/L)
             le-6   =  (kg/mg)
      Depuration   =  loss rate within organism (ng/L day);
              K2   =  elimination rate constant (1/d).

Dietary uptake of chemicals by animals is not affected by this alternative parameterization.


8.9 Half-Life Calculation, DT50 and DT95

AQUATOX estimates time to 50% (half-lives, DTSOs) and time to 95% chemical loss (DT95s)
independently  in bottom sediment  and in  the water column.  Estimates are produced at each
output time-step  depending on the average loss rate during that time-step in that medium.
          HydrolysisWater + Photolysis + MicrobialWater + Washout + Volatilization. + Sorption ^>fic\
LossWater =	—	(385)
                                          MaSSWater
LOSSSed ~
                        MicrobialSed + Hydrolysis Sed + Desorption + Scour Sed
                        - — -
                                            MassSed
                                          262

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                      CHAPTER 8
where:
         LdSSMedia
   Hydrolysis Media
       Photolysis
    MicrobialMedia
         Washout
     Volatilization
         Sorption
        MaSSMedia
       Desorption
            Scour
  loss rate within media (1/d);
  hydrolysis rate in given media (ng/L d), see (313);
  photolysis rate in the water column (ng/L d), see (320);
  rate of microbial metabolism in given media (ng/L d), see (326);
  rate of toxicant washout from the water column (ng/L d); see (16);
  rate of chemical volatilization in the water column (ng/L d), see (331);
  sorption of toxicant to detritus, plants, and animals (ng/L d), see (350);
  mass of chemical in the media (|j,g/L);
  desorption of toxicant from bottom sediment, (|J,g/L d) see (351);
  resuspension of toxicants in bottom sediments, (ng/L d) see (233).
Loss rates are converted into time to 50% and 95% loss using the following formulae for first-
order reactions:
                                         = 0.693 1 Loss
                               DT95Media=2.996ILossMedia
                                                             (387)
                                                             (388)
where:  DTSOMedta   =   time in which 50% of chemical will be lost at current loss rate (d);
        DT95Media   =   time in which 95% of chemical will be lost at current loss rate (d);
         LossMedia   =   loss rate within media (1/d);

8.10 Chemical Sorption to Sediments

When the complex multi-layer sediment model is included, chemicals can sorb to and desorb
from suspended inorganic sediments based on user input rates that are applied to the model's
equations for sorption (249), and desorption (250).  To activate this model, required rates are:

       Kl      uptake rate constant        L/kg dry day
       K2      depuration rate constant    I/day
       Kp     partition coefficient         L/kg dry

The derivative  for toxicants sorbed to inorganic sediments is similar to that  for suspended
organics:
       dToxicant
                SuspSed
             dt
                      = Load - Microbial + Sorption -Desorption
- (Deposition + Washout)- PPBSuspSed
+ (Washin- PPB SuspSedupstream •  le-6)
                                                          le-6
(389)
  (Scour •  PPB Bot
                                      ttomed
                                             le-6)
where:
       Toxicant SuspSed  =   toxicant in relevant suspended sediment size-class (ug/L);
                                          263

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
       Load
       Microbial
       Sorption
       Desorption
       Deposition

       Washout

       Washin

       Scour
loading of toxicant from external sources (ug/L-d);
rate of loss due to microbial degradation (ug/L-d), see (326);
rate of sorption to given compartment (ug/L-d), see (350);
rate of desorption from given compartment (ug/L-d), see (351);
rate  of sedimentation  of given suspended  detritus  (mg/L-d) in
streams with the inorganic sediment model attached, see (230);
rate of loss of from sediment being carried downstream (mg/L-d),
see (16)
rate of gain from sediment  carried in from any upstream linked
segments (mg/L-d), see (30);
rate of resuspension of given  sediment (mg/L-d), see (227);
Chemicals also are tracked within inorganic sediments in the multi-layer sediment bed:

             dToxicantK
                      BottomSed _
                    dt
   = Sorption -Desorption - Microbial

 + (Deposition• PPBSuspSed  • le-6) + BedLoadTo
 -(Scour • PPBBottomSed  • 1 e -6) -BedLossTox
                                                                                  (390)
where:
       Toxicant BottomSed     =      toxicant in bottom  sediment (relevant sediment size-class
                          ug/m2);
                                                                    r\
       Microbial      =   rate of loss due to microbial  degradation (ug/m -d), see (326);
       Sorption        =   rate  of  sorption to  given  compartment  (ug/m2-d  after units
                          conversion), see (350);
                                                                          r\
       Desorption     =   rate of desorption from given compartment (ug/m -d after units
                          conversion), see (351);
       Deposition     =   rate of sedimentation of given  suspended detritus (ug/m2-d  after
                          units conversion) in streams  with the inorganic sediment model
                          attached,  see (230);
       Scour          =   rate  of  resuspension  of given  sediment  (ug/m2-d  after units
                          conversion), see (227);
                                                             r\
       BedLoadrox     =   rate of bed load of given toxicant (ug/m -d), see (391);
       BedLossTox     =   rate of bed loss of given toxicant (ug/m2-d), see (392).
                                                                  r\
In several cases above, units need to  be converted from ug/L-d to ug/m -d when moving from
sediment suspended in the water column to bed sediment.  This is done by multiplying by water
volume and then  dividing by the sediment bed surface area.  Toxicant mass balance has been
verified to be conservative through this process.

Toxicant movement due to bedload and bedloss are straightforward calculations:
                                          264

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                          CHAPTER 8
             D  .n   J
             BedLoad   =
Tox
where:
      BedLoadxox
      BedLoadupstreamhnk
      AvgArea
       le-
                                    AvgArea
                                                    nnr>          ->    o
                                                    PPBUpstreamBed • 1 e - 3
                                                                                 (391)
                                                                              r\
                                 toxicant bedload from all upstream segments ((J-g/m -d);
                                 bedload over one of the upstream links (g/d);
                                 average area of the segment (m2);
                                 toxicant concentration in the relevant upstream link (ug/kg)
                                 units conversion (kg/g)
Similarly, total bed loss is the sum of the loadings over all outgoing links:
      BedLossxox
      BedLoss
      AvgArea
      PPBBed
       le-3
                        Tox

                                {      AvgArea

                                 toxicant bedloss from current segment (ug/m2-d);
                                 bedloss over one of the downstream links (g/d);
                                 average  area of the segment (m ) ;
                                 toxicant concentration in the current segment (ug/k
                                 units conversion (kg/g)
                                                                                 (392)
8.11 Chemicals in Pore Waters

When the complex multi-layer sediment model is included, pore waters may contain toxic
organic chemicals.  Chemicals in pore waters are separated into those that are freely dissolved
and those that are complexed to dissolved organic carbon within the pore waters.
        dToxicant,;
                             = GainToXup - LossTaxUp + DiffDom + DiffUp + Decomp
                    - GillUptake - Microbial - Sorption + Desorption + Depuration

where:
    Toxicant FreefyDissoivedp.w.

    GainToxup

    LossToxup

    Diffup, DiffDown
                               change in  concentration  of pore water in the  sediment bed
                               normalized per unit area ((ig/L^ -d);
                               active layer only: gain of toxicant due to pore water gain from
                               the water column ((jg/L^-d), see (394);
                               active layer only: loss of toxicant due to pore water loss to the
                               water column (fig/L^, • d), see (395);
                               diffusion over upper or lower boundary ((j,g/Lpw- d), see (256);
                                         265

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 8
    Decomp                =

    GillUptake             =

    Depuration             =

    Microbial              =

    Sorption, Desorption    =
        freely dissolved toxicant gain due to microbial decomposition
        of organic matter ((j,g/Lpw-d), see (159);
        active layer  only:  uptake of  toxicant into organisms that
        reside at least partially in the sediment ((j,g/Lpw-d) (365);
        active layer  only:  excretion  of toxicant by organisms that
        reside at least partially in the sediment ((j,g/Lpw-d), (362);
        loss of toxicant in  pore waters due to microbial degradation
        (Hg/Lpw-d)see(326);
        sorption to and desorption from organic matter and inorganic
        matter in the current layer ((j,g/Lpw-d). (350), (351)
               GainToxUp =
                             GamUp • AreaSedLayer • ConcToxWateiCol • Ie3
                                           Volume
                                                  PoreWater
                                                          (394)
where:
               LossToxUp =
                             LossUp • AreaSedLayer • ConcToxporeWater • Ie3
       GainToxup
       LossToxup
           i Up, Loss up

      AreaSedLayer
       Volume poreWater
       Ie3
                                          Volume
                                                  PoreWater
                                                          (395)
    gain of toxicant in pore water from the water col. ((j,g/Lpw-d);
    loss of  toxicant in pore  water to the water  column  above
    (Hg/Lpw-d);
    gain or loss of pore water from the water column above (m3/m2-d);
    see (252), (251);
    sediment layer area (m2);
    concentration of toxicant in relevant media (|j,g/L);
    pore water volume (Lpw);
    units conversion (L/m ).
Chemicals also sorb to dissolved organic matter within pore waters:
        dToxicant
                 DOMPoreWate,
                dt
^- = GainDOMToxUp - LossDOMToxUp ± DiffDown ± DiffUp

- (Decomp • PPB • le - 6) - Microbial - Sorption + Desorption
where:
       GainDOMToxjjp =

       LossDOMToxup =
    gain  of toxicant  sorbed  to  DOM  from  the  water  column
    (Hg/Lpw-d)see(394);
    loss of toxicant sorbed to DOM in pore water to the water column
                                         266

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8


                           above ((j,g/Lpw-d) see (395);
       Diffup, DiffDown   =   diffusion over upper or lower boundary ((j,g/Lpw- d), see (256);
       Decomp         =   Decomposition of DOM ((j,g/Lpw-d), see (159);
       Microbial       =   loss of toxicant sorbed to DOM due to microbial degradation
                           (Hg/Lpw-d)see(326);
       Sorption         =   sorption to DOM (|j,g/LpW-d). (350)
       Desorption      =   desorption from DOM ((j,g/Lpw-d). (351)


8.12 Mass Balance Capabilities and Testing

A chemical mass balance testing capability was added to the code during the development of the
estuarine version  of AQUATOX.  This capability ensured  that all linkages between stratified
layers were properly developed with no loss of mass balance.  New PFA (perfluorinated acid)
formulations were also tested for mass balance with this capability.  Current testing indicates that
AQUATOX balances chemical mass to machine accuracy.

The chemical mass balance testing comprehensively tracks the mass of all chemical loadings and
losses  to the system.   Chemical mass balance is explicitly tested with this capability;  mass
balance of state variables containing chemicals is implicitly tested. The Chemical MBTest output
variable keeps track of all chemical by the following equation:

MBTest  =  Chemical Mass + Chemical Loss - Chemical Load - Net Layer Exchange    (397)

In this  manner, the MBTest will stay constant (within machine accuracy) throughout a simulation
if mass balance is being maintained.  However, the chemical mass balance function does not
work if the "Keep Freely Dissolved Contaminant Constant" option is selected within the  setup
screen.

The  chemical mass balance capability also provides a chemical tracking capability that allows
the user to see exactly what is happening to the chemical within the system. Chemical fate may
be tracked using the following output categories (all units are in kilograms):

   Chem. MBTest:        Mass balance test as described above,  see (397).

   Chem. Mass:           Total chemical mass in  the system including  chemicals within
                           biota.

   Chem. Loss + Mass:     Chemical loss plus chemical mass in the system.
                                         267

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                              CHAPTER 8
   Chem. Tot Wash:

       Chem. WashH2O:
       Chem. WashAnim:
       Chem. WashDetr:
       Chem. WashPlnt:

   Chem. Tot Loss:

       Chem. Hydrol:
       Chem. Photol:
       Chem. Volatil:
       Chem. MicrobMet:
       Chem. BioTrans:
       Chem. Emergel:
       Chem. Fishing Loss:

   Chem. Tot Load:

       Chem. H2O Load:
       Chem. Detr Load:
       Chem. Biota Load:

   Net LayerExch:

       Chem. Net Sink:
       Chem. Net Entrain:
Washout of chemical from the system since the simulation start.
The sum of the below four categories:
Washout of chemical dissolved in water
Washout of chemical in drifting animals.
Washout of chemical in suspended & dissolved detritus.
Washout of chemical in plants

Total loss of chemical from the  system since the simulation start.
The sum of the following eight categories plus washout:
Chemical loss due to hydrolysis.
Chemical loss due to photolysis.
Chemical loss due to volatilization.
Chemical loss due to microbial metabolism.
Chemical loss due to biotransformation.
Chemical loss due to the emergence of insects.
Chemical loss due to fishing.

Total loading of chemical into the system since  the simulation
start. The sum of the following three categories:
Load of chemical directly into water.
Load of chemical within detritus loadings.
Load of chemical within plant and animal loadings.

Net of layer exchange between the other layer in the system.  The
sum of the below five categories:
Net sinking from upper to lower layer.
Net entrainment of chemical.
       Chem. Net TurbDiff: Net turbulent diffusion of chemical.
       Chem. Net Migrate:  Net migration of chemical in animals.
       Chem. Delta Thick:  Chemical movement due to changes in the thickness of the two
                          layers.
                                        268

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8
8.13 Perfluoroalkylated Surfactants Submodel

As  mentioned in the  introduction  (section  1.5),  the perfluorinated compounds  of interest as
bioaccumulators are the perfluorinated acids (PFAs). Perfluoroctane sulfonate (PFOS) belongs
to the sulfonate group and perfluorooctanoic acid (PFOA) belongs to the carboxylate group. Due
to their use in industrial manufacturing, these persistent chemicals  are found in  humans, fish,
birds, and marine and terrestrial mammals throughout the world. PFOS has an especially high
bioconcentration factor in fish.

Sorption

Perfluorinated surfactants are  quite different from hydrocarbon surfactants.  The  nonpolar
perfluorocarbon tail repels both water and oil, and the perfluorinated surfactants are much more
active than their hydrocarbon counterparts (Moody and Field 2000).  A field is provided for the
user to  input a value  for the organic matter partition coefficient ("Kom  for Sediments"); this
empirical approach was taken in lieu of sufficient theory to support a  mechanistic formulation.
Sorption to algae and macrophytes are also modeled empirically ("BCF for Algae" and "BCF for
Macrophytes" parameters).

Biotransformation and Other Fate Processes

PFOS and other related chemicals are anionic surfactants and, as such, they are not subject to
volatilization.  However, the worldwide detection of PFOS suggests that there are one or more
precursors that are volatile.  Therefore,  a fate model for these compounds would not be complete
if it were not able to  represent the movement and transformation  of significant precursors to
PFOS  and  other bioaccumulative fluorinated  organics.   In  particular,  some fluorinated
compounds are subject to biodegradation of the nonfluorinated portion  (Key et al. 1998, Moody
and  Field 2000, Giesy and Kannan  2001);  these can  yield  both  volatile and nonvolatile
biotransformation products (Key et al. 1998).  For example, 7V-EtFOSE alcohol  is subject to
microbial  degradation, yielding 92% PFOS  and 8% PFOA (Lange  2000 cited in Cahill et al.
2003).  AQUATOX has the capability of representing biotransformation from one congener or
homolog to one  or more others when  there  are sufficient data to parameterize that part  of the
model.

Bioaccumulation

PFOS and PFOA and  similar compounds bioaccumulate differently  than PCBs and chlorinated
pesticides (Kannan et  al. 2001).  The perfluorinated compounds  of interest as bioaccumulators
are the acids.  At least for PFOS the salts dissociate instantaneously at neutral pH (OECD 2002).
Perfluorinated acids (PFAs) are  oil repelling and  are taken up by protein rather than lipids
(Kannan quoted in Scientific American,  March,  2001).  Therefore,  their kinetics  cannot  be
modeled as functions of the octanol-water partition coefficient.  Instead, relationships based on
perfluoroalkyl chain length (Martin et al. 2003a) are used.
                                          269

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 8
Gill Uptake

Data  on PFAs were insufficient at the time this submodel was  first  developed (2005) to
determine withdrawal efficiencies and  explicitly include respiration such as is done for other
organic compounds simulated by AQUATOX. Based on the data of (Martin et al. 2003a), the
uptake rate for all but the longest chain-length carboxylates can be represented as:
                       k\  =  SizeCorr •W-5-12l3+0-1164-cha"lLength                         (398)

where
       kl           =     uptake transfer rate (L/kg d);
       ChainLength =     length of perfluoroalkyl chain (integer).

If chain length exceeds 11, the value for 11 is used.  These data were based on 5-g trout, and
uptake is implicitly a function  of respiration, which is sensitive to size.  A size correction is
based on a standard allometric relationship and the reciprocal of that value for a 5-g fish:
                          SizeCorr  =  MeanWeight™	                     (399)
                                                       Sizeref

where
       SizeCorr     =     allometric correction for size (unitless);
       MeanWeight  =     mean wet weight of organism (g);
       RB          =     allometric exponent for respiration (unitless);
       SizeRef      =     reference value (0.7248).

The respiration rate decreases with larger sizes. The allometric exponent RB is assigned values
based on the Wisconsin Bioenergetics Model (Hewett and Johnson 1992).  If RB = -0.2 then the
correction for a 10-g fish is 0.63, that is, uptake is 63% that of the fish for which the kl values
were determined; the correction for a 100-g fish is 55% of the reference; and the correction for a
1000-g fish is 35%. For invertebrates RB is assigned a value of-0.25 (Moloney and Field 1989).

Although there are only two data points for sulfonates (Martin  et al. 2003a), the trend defined by
those points provides an approximation:

                           k\ =  SizeCorr •lO-6M+0-966-chc""Le"gth                      (400)

However, the kl values were determined from the first few observed uptake values and not the
observations just before the depuration phase of the experiment.  Adjusting the intercept actually
provides a better fit to the overall experiment:

                          kl   =   SizeCorr •W-5-S5+0-966'chamLength                        (401)
                                          270

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 8
Uptake rates (kl values) must be estimated  and entered in the "chemical toxicity" screen and
estimates can be modified with user-supplied  values if appropriate.

With the greater intercept, the sulfonates are taken up more rapidly than the carboxylates, as
shown in Figure 151. Gill uptake is calculated as:
                     GillUptake  = ToxicantWater • kl • StVarAmmal -le-6
       (402)
where:
       GillUptake   =      uptake of toxicant by gills (ng/L d);
       WetToDry    =      conversion factor for wet to dry weights (5);
       SizeCorr     =      allometric correction for size (unitless), see (400)(400);
       Toxicant water  =      concentration of toxicant in water (ug/L);
       kl           =      uptake transfer rate (L/kg d);
       StVarAnimal    =      biomass of given animal (mg/L);
       1 e-6         =      units conversion (kg/mg).

       Figure 151: Predicted and observed uptake transfer rates for carboxylates and sulfonates.
-5 C -
•5

9 K -
^
ra ">
m
^ 15
,- 1-5
J£
O 1
J> '
.5 -
n
0 R -
t



_/_
/
mf
/
f
/
/
/ /
/
5 8 10 12
Perfluoroalkyl Chain Length



i


• Obs carboxylate
.

Obs sulfonate



Dietary Assimilation

Martin et al. (2003b) found that assimilation of PFAs was quite efficient, exceeding that for the
normal hydrophobic chemicals. However, many of the calculated values reported (Martin et al.
2003b) exceeded  1.0, so the observed assimilation efficiencies were normalized to a maximum
of 1.0, and equations were derived for uptake from the gut (GutEfJTox).  If a carboxylate:
                log GutEff  =  - 0.91 + 0.085 • ChainLength    r2 = 0.897
       (403)
If a sulfonate:
                                          271

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 8
              GutEff  =  - 0.68 + 0.21 • ChainLength       r2 =1.0 (2 points)

In the absence of information on other organisms, these equations are used for all animals.

                Figure 152: Gut assimilation efficiency as a function of chain length.
                                                        (404)
1 Tl
o 1 nn -
o>
.— n &fi -
1C
ill
S n £fi -
IB
£
•S A Af\ _
'w
« n 9n
n nn -
A
/ z
/ z
* /^





— « — Pred/Obs su If on ate
Normalized
carboxylate

5 7 9 11 13
Perfluoroalkyl Chain Length
Depuration

Based on regression of published data from experiments with juvenile trout (Martin et al. 2003a,
Martin et al. 2003b), carboxylate depuration can be estimated as:
                  kl   =  SizeCorr-W
                                      -0.0873 - 0.1207 • ChainLength
                                   r2 =0.98
(405)
where:
              k2
       SizeCorr
depuration rate (1/d).
allometric correction for size (unitless), see (400);
Only four data points are available for two sulfonate compounds (Martin et al. 2003a, Martin et
al. 2003b); but they indicate that depuration is much slower than for carboxylates.  The model
extrapolates from those two pairs of points, but this estimation procedure should be used with
caution (Figure 153):
                   kl   =  SizeCorr -10
                                       -0.733 - 0.07 • ChainLength
                                  r2 =0.84
(406)
where:
              k2
       SizeCorr
depuration rate (1/d).
allometric correction for size (unitless), see (400);
Because uptake is so efficient in the gut, depuration may be largely across the gills. If this is true
                                           272

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                        CHAPTER 8
then depuration rate can be related to respiration rate, providing a correction for size.

In the absence of any data, this approach to modeling depuration is extended to invertebrates.
When data become available on depuration of PFAs in invertebrates, this series of constructs
may be modified.  Depuration rates  (k2s) must be estimated and entered in  the  "chemical
toxicity" screen and estimates can be modified with user-supplied values if appropriate.

             Figure 153. Depuration rate as a function of perfluoroalkyl chain length.
01 A -,
01 9
O-i
Ons
CM
^
n OR
Or\A
OCI9 -
n -
•
\
\
V
^•^
^""""^ 4

i i i

• Obs Caboxylate
	 Precl Cai boxylate
Obs Sulfonate
Pi eel Sulfonate
i
5 7 9 11 13
Perfluoroalkyl Chain Length
Available data indicate that concentrations of PFOS in wildlife are less than those known to
cause toxic effects in laboratory animals (Giesy and Kannan 2001).  AQUATOX provides a
means of factoring in toxicity data as they become available for aquatic species.
Bioconcentration Factors

The  steady-state  bioconcentration  factor (BCF) for  carboxylates,  used to  compute  time-
dependent toxicity, can be estimated by (Martin et al.  2003a):
              \ogBCF  =  -5.724 + 0.9146 • ChainLength
                                            r2 =0.995
(407)
where
       BCF   =
bioconcentration factor (L/kg).
Similar to uptake, the slope for the BCFs of sulfonates closely parallels that of carboxylates but
with a different intercept (Figure 154):
          logBCFsulfonate   =  -5.195 + 1.03- ChainLength       r2 = 1.0 (2 points)     (408)
                                          273

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
           CHAPTER 8
For compounds with perfluoroalkyl chain lengths in excess of 11, it is assumed that the BCF is
the same as that for chain length 11, as suggested by the outlier (Figure 154).
          Figure 154. Bioconcentration factors as functions of perfluoroalkyl chain length.
             •5
Pred Garb

Pred Sulf

Obs Carboxylate

Obs Sulfonate
                                                    13
                           Perfluoroalkyl Chain Length
                                           274

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
               CHAPTER 9
                                 9. ECOTOXICOLOGY

Unlike  most  ecological  models,  AQUATOX  contains  an
ecotoxicology submodel that computes both lethal and sublethal
acute toxic effects from  the concentration of a  toxicant in a
given  organism.   Furthermore,  because  AQUATOX  is  an
ecosystem model, it can simulate indirect effects such as loss of
forage  base,   reduction  in  predation,  and  anoxia  due  to
decomposition following a fish kill.

User-supplied values  for LC50, the  concentration of a toxicant
in water  that  causes  50% mortality,  form the  basis  for  a
sequence  of computations that lead to estimates of the biomass
of a given organism lost through lethal toxicity each day. The
sequence, which is documented in this chapter, is to compute:

    •  the internal concentration causing  50% mortality for a
       given period of exposure;
    •  the internal concentration causing 50% mortality after an
       infinite period of time based on an asymptotic concentration-response relationship;
    •  the time-varying lethal internal concentration of a chemical;
    •  the cumulative mortality for a given internal concentration;
    •  the biomass lost per day as an increment to the cumulative mortality.

The user-supplied ECSOs, the concentrations in water eliciting  sublethal toxicity responses in
50% of the population, are used to  obtain  factors relating the sublethal  toxicities to the lethal
toxicity.    Because AQUATOX can  simulate as  many as twenty  toxic organic chemicals
simultaneously, the simplifying assumption is made that the toxic effects are additive.
Ecotoxicology: Simplifying
Assumptions

 • Toxic effects of multiple chemicals
   are additive
 • Sublethal effects levels of
   chemicals may be estimated as a
   fraction of lethal effects levels
 • Regressions from one species to
   another are available regardless of
   the mode of action
 • The external toxicity model
   assumes immediate toxic effect to a
   level of external exposure
 • Cumulative toxicity considers
   differing tolerances in a population,
   but ignores inherited tolerance
 • Resistance to lower doses is
   conferred for the lifetime of an
   animal and for one year for a plant.
9.1 Lethal Toxicity of Compounds

Interspecies Correlation Estimates (ICE)

Often LC50 data will only be available for one or two of the many species that a user wishes to
include in a simulation.   To  alleviate  this problem,  a substantial  database  of regressions
(Interspecies Correlation Estimation,  ICE) is available as developed by the US. EPA Office of
Research  and Development, the University of Missouri-Columbia,  and  the US Geological
Survey (Asfaw and Mayer, 2003). At this time the Web-ICE database has over 2000 regressions
with over 100 aquatic species as "surrogates" (Raimondo et al. 2007). Regressions may be made
on  the basis of species, families,  or genera.   The database also includes goodness of  fit
information for regressions  so their suitability for a given application may be ascertained. Only
statistically significant regressions are included in the  database.

Using  the ICE database and the following regression equation, the model can be parameterized
to represent a complete food web.
                                            275

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 9
                   Log LC50Estimated = Intercept + Slope • Log LC50observed                (409)
where:
       LC50 Estimated  =      estimated LC50 (ug/L);
       Intercept     =      intercept for regression (ug/L);
       Slope        =      slope of the regression equation;
                    =      observed LC50 (ug/L).
The ICE database is integrated into the AQUATOX user interface. A link is provided to the
Web-based (Web-ICE) site so that the user can alternatively use the web tool. The steps that a
user can take to use ICE within AQUATOX to estimate unavailable LC50 data are as follows:

   •   Invoke the ICE interface from the AQUATOX "Chemical Toxicity Parameter" screen;
   •   Choose from the six available ICE databases (species,  genus,  and family  by either
       scientific names or common names);
   •   Either choose a "surrogate species" that matches a species for which there is observed
       LC50 data, or start with a "predicted species"  that matches a species that you  wish to
       model;
   •   The list box that you did not select from in the previous step will narrow to reflect the
       available surrogate or predicted species that match with your selection.  Select a choice
       from this list box as well. If you wish to start over again, you may select the "show all"
       button next to this list box.
   •   Examine the goodness  of fit for your model and evaluate whether it is appropriate for
       your purposes. Where  there are multiple surrogates for the desired predicted  species,
       compare the statistics and choose best surrogate/predicted pair;
   •   Apply the model by assigning  the surrogate and predicted species to species within the
       chemical's toxicity record.

Experimentally derived toxicity data  for  individual  species should be  used when  available.
However, ICE may then be used to estimate toxicity for species that have not yet been studied
given a particular chemical. There are uncertainties in  this estimation procedure, but the model
helps to track these uncertainties. When the ICE model  is invoked, data about the goodness of fit
and confidence interval  are copied  back into the "LC50 comment"  field.   Overall model
uncertainty resulting from this estimation can then be numerically quantified — these goodness of
fit data can be utilized within an iterative AQUATOX uncertainty analysis (see section 2.5).
                                          276

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                   CHAPTER 9
Internal Calculations

Toxicity is based on the internal concentration of the toxicant in the specified organism. Many
compounds,  especially those with higher octanol-water partition coefficients, take appreciable
time to accumulate in the tissue.  Therefore, length of exposure is critical in determining toxicity.
The same principles apply to organic toxicants and to both plants and animals.

The internal  lethal concentration for a given period of exposure can be computed from reported
lethal toxicity data based on the simple relationship  suggested by an algorithm in the FGETS
model (Suarez and Barber, 1992):
                               InternalLCSO = BCF • LC50
where:
                                                          (410)
       InternalLCSO
       BCF
       LC50
 internal concentration that causes 50% mortality;
 bioconcentration factor (L/kg), see (342) to (349); and
 concentration of toxicant in water that causes 50% mortality (ug/L).
For compounds with a LogKOWin excess of 5 the usual 96-hr toxicity exposure does not reach
steady state, so a time-dependent BCF is used to account for the actual internal concentration at
the end of the toxicity determination.  This is applicable no matter what the length of exposure
(Figure 155, based on Figure 141).

               Figure 155: Bioconcentration factor as a function of time and KOW
                Q
                i
                CD
                         1E4
                         1E3
                             0 \ 200   400  600  800  1000 1200
                                              DAY
The internal concentration causing 50% mortality after an infinite period of exposure, LCInfmite,
can be computed by:
where:
       k2
LCInfmite = InternalLCSO • (l - e-k2'ob*

   elimination rate constant (1/d); and
                  277
                                                                                  (411)

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 9


       ObsTElapsed =      exposure time in toxicity determination (h).

Essentially this equation determines the asymptotic toxicity relationship and provides the model
with a constant toxicity parameter for a given compound.

The model estimates k2, see (364) and (354),  assuming that this k2 is the same as that measured
in bioconcentration tests; good agreement has been reported between the two (Mackay et al.,
1992). The user may then override that estimate by entering an observed value. The k2 can be
calculated off-line based on the observed half-life:
                                                                                  (412)
                                            tl/2
where:
       t'/2    =     observed half-life.
Based on the Mancini (1983) model, the lethal internal concentration of a toxicant for a given
exposure period can be expressed as (Crommentuijn et al. (1994):
                               LethalConc =            ,                           (413)
                                             7   -k2» TElapsed                           ^   '

where:
       LethalConc   =     tissue-based concentration of toxicant that causes  50% mortality
                          (ppb or ug/kg);
       LCInfinite    =     ultimate internal lethal  toxicant concentration after  an infinitely
                          long exposure time (ppb);
       TElapsed     =     period of exposure (d).
The longer the exposure the lower the internal concentration required for lethality.

Exposure is limited to the lifetime of the organism:

                     if TElapsed > LifeSpan then TElapsed = LifeSpan                (414)
where:
       LifeSpan     =      user-defined mean lifetime for given organism (d).

Based on an estimate of time to reach equilibrium (Connell and Hawker, 1988),
                                •*• TTT/     ^^-+u
                                it 1 Elapsed > - then
                                      P       k2                                 (415)
                                LethalConc = LCInfinite

The fraction killed by a given internal concentration of toxicant is best estimated using the time-
dependent LethalConc in the cumulative form of the Weibull distribution (Mackay et al., 1992;
see also Christensen and Nyholm, 1984):
                                          278

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                CHAPTER 9
                            CumFracKilled = 1-e
                                                  PPB
                                                 LethalConc
                                                       (416)
where:
       CumFracKilled  =
       PPB
       Shape           =
fraction of organisms killed per day (g/g d),
internal concentration of toxicant (ug/kg), see (310); and
parameter expressing variability in toxic response (unitless).
As a practical matter, if CumFracKilled exceeds 95%, then it is set to 100% to avoid complex
computations with small numbers. By setting organismal loadings to very small numbers, seed
values can be maintained in the simulation.

This formulation is preferable to the empirical probit and logit equations because it is simple and
yet based on mechanistic relationships. The Shape parameter is important because it controls the
spread of mortality.  The larger the value, the greater the distribution of mortality  over toxicant
concentrations and time. Mackay et al. (1992) found that a value of 0.33 gave the best fit to data
on toxicity of 21 narcotic chemicals  to fathead minnows.  This value is used as a default  in
AQUATOX, but it can be changed by  the user.  Although mercury is not currently modeled, data
on MeHg toxicity shows that the Shape parameter may take a value less than 0.1 (Figure 154).

            Figure 156. The effect of Shape in fitting the observed (McKim et al., 1976)
                 cumulative fraction killed following continued exposure to MeHg
                                               - LetfialCoTicHg for Day 63 /
                      0    200   400  600
                  800  1000 120014001600
                 DAYS
The biomass killed per day is computed by disaggregating the cumulative mortality. Think of
the biomass at any given time as consisting of two types: biomass that has already been exposed
to the toxicant previously, which is called Resistant because it represents the fraction that was
not killed; and new biomass that has formed through growth, reproduction, and migration and
has not been exposed to a given level of toxicant and therefore is referred to as Nonresistant.
Then think of the cumulative distribution as being the total CumFracKilled, which includes the
FracKilled that is in excess of the  cumulative  amount on the previous  day if the internal
                                          279

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 9


concentration of toxicant increases.  A conservative estimate of the biomass killed at a given
timeis computed as:

          Poisoned = Resistant • Biomass • FracKilled + Nonresistant • CumFracKill     (417)
where:
       Poisoned       =    biomass of given organisms killed by exposure to toxicant at given
                           time (g/m3 d);
       Resistant       =    fraction of biomass not killed by previous exposure (frac);
       FracKilled     =    fraction killed per day in excess of the previous fraction (g/g d);
       Nonresistant    =    biomass not previously exposed; the biomass  in  excess  of the
                           resistant biomass (g/m3) = (1-Resistant)-Biomass.

New biomass is considered vulnerable, ignoring  the possibility of  inherited tolerance.   It  is
assumed for purposes of risk analysis that resistance is not conferred for an indefinite period.  In
animals elapsed exposure time is capped at the  average  life  span,  which is a parameter in the
animal record. However, it is assumed that resistance persists in the population until the end  of
the growing  season.   Macrophytes  can  live for an  entire growing  season,  and algae  usually
reproduce asexually as long as conditions are favorable. However, winter die-back does occur in
most  macrophytes, and many  algae will switch to sexual  reproduction under  unfavorable
conditions, especially triggered by light and temperature.  As a simplifying assumption for  both
animals and plants, in the northern  hemisphere  January  1  is taken as being the date at which
exposure and resistance are reset; in the southern hemisphere (denoted by negative latitude in the
site record) July 1  is the reset date.  On this date, the variables Resistant, FracKilledPrevious, and
TElapsed are all set to zero.

9.2 Sublethal Toxicity

Organisms usually have adverse reactions to toxicants at levels significantly below those that
cause death.  In fact, the lethal to sublethal ratio  is commonly used to quantify this relationship.
The user supplies observed EC50 values, which can then be used to  compute AFs (application
factors).  For example:

                                ..,„     .   ECSOGrowth
                               AFGrowth= 	                          (418)
                                               LC50
where:
       ECSOGrowth     =   external concentration  of  toxicant  at  which  there is  a  50%
                           reduction in growth (ug/L);
       AFGrowth       =   sublethal to lethal ratio for growth (unitless); and
       LC50           =   external concentration of toxicant at which 50% of population is
                           killed (ug/L).

If the  user enters  an observed EC50 value, the  model provides the option of applying the
resulting AF to  estimate ECSOs for other organisms.   The computations for AFPhoto and
AFRepro are similar:

                                          280

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                 CHAPTER 9
                                 AFPhoto =
                 ECSOPhoto
                    LC50
                                  .„„       ECSORepro
                                 AFRepro =	—
                                               LC50
(419)
                                                        (420)
where:
       ECSOPhoto

       AFPhoto
       ECSORepro

       AFRepro
external  concentration  of toxicant  at  which  there  is a  50%
reduction in photosynthesis (ug/L);
sublethal to lethal ratio for photosynthesis (unitless);
external  concentration  of toxicant  at  which  there  is a  50%
reduction in reproduction (ug/L); and
sublethal to lethal ratio for reproduction (unitless).
Because of the nature of these application factors, sublethal effects cannot be calculated (using
internal calculations) unless LC50 parameters are included in the model.

Similar to computation  of lethal toxicity in the model,  sublethal  toxicity is based on internal
concentrations of a toxicant.  Often sublethal effects form a continuum with lethal effects and the
difference is merely one of degree (Mackay et al., 1992).  Regardless of whether or not the mode
of action is  the same, the computed factors relate the observed effect to the lethal effect and
permit efficient computation of sublethal effects factors in conjunction with computation of
lethal effects. Because AQUATOX simulates biomass, no distinction is made between reduction
in a process in an individual and the fraction  of the population exhibiting that response. The
commonly measured reduction in photosynthesis is a good example: the data only indicate that a
given reduction takes place  at a given concentration, not whether all individuals  are affected.
The factor enters into the Weibull equation to estimate reduction factors  for photosynthesis,
growth, and reproduction:
                                               PPB
                         FracPhotO — e\LethalConc • AFPhoto
                                                           I/Shape
                                                        (421)
                     RedGrOWth — 1 - e\ LethalConc-AFGrowth
                                                 PPB
                                                              1/Shape
                                                        (422)
where:
                      RedRepro — 1 - e\ LethalConc • AFRepro
                                                 PPB
                                                              I/Shape
                                                        (423)
       FracPhoto   =   reduction factor for effect of toxicant on photosynthesis (unitless);
       RedGrowth  =   factor for reduced growth in animals (unitless);
       RedRepro   =   factor for reduced reproduction in animals (unitless);
       PPB        =   internal concentration of toxicant (ug/kg), see (310);
       LethalConc  =   tissue-based cone, of toxicant that causes mortality (ug/kg), see (413);
                                          281

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 9


       AFPhoto    =   sublethal to lethal ratio for photosynthesis (unitless, default of 0.10);
       AFGrowth   =   sublethal to lethal ratio for growth in animals (unitless, default of 0.10);
       AFRepro    =   sublethal to lethal ratio for reproduction in animals (unitless, default of
                       0.05);
       Shape       =   parameter expressing variability in  toxic response (unitless, default of
                       0.33).

The reduction factor for photosynthesis, FracPhoto, enters into the photosynthesis equation (Eq.
(35)) and it also appears in the equation for the acceleration of sinking of phytoplankton due to
stress (Eq. (69)).

The  variable for reduced growth, RedGrowth,  is  arbitrarily  split  between  two  processes,
ingestion (Eq. (91)), where it reduces consumption by 20%:

                           ToxReduction = 1-(0.2- RedGrow th)                       (424)
and defecation (Eq. (97)), where it increases the amount of food that is not assimilated by 80%:

                    IncrEgest = (1 - EgestCoeff prey pred) -0.8- RedGrowth               (425)
These have indirect  effects on the rest of the ecosystem through reduced predation and increased
production of detritus in the form of feces.

Embryos are often  more  sensitive to toxicants,  although reproductive  failure may occur for
various reasons.  As a simplification, the factor for reduced reproduction, RedRepro, is used
only to increase gamete mortality (Eq. (126)) beyond what would occur otherwise:

                            IncrMort = (1- GMort) • RedRepro                        (426)

By modeling sublethal and lethal effects, AQUATOX makes the link between chemical fate and
the functioning of the aquatic ecosystem-  a pioneering approach that has been refined over the
past twenty years, following the first publications (Park et al.,  1988; Park,  1990).

Sloughing of periphyton and drift of invertebrates also can be elicited by toxicants. For example,
sloughing can be caused by a  surfactant that disrupts the adhesion  of the periphyton, or  an
invertebrate may release its hold on the substrate when irritated by a  toxicant. Often the response
is immediate so that these responses can be modeled as dependent on dissolved concentrations of
toxicants with  an  available sublethal toxicity parameter, as in the equation for periphyton
sloughing:
           Dislodgep .T  = MaxToxSlough	^^	Biomassp        (427)
                                          ToxicantWater+EC50Dlsli
where:
                      =   periphyton sloughing due to given toxicant (g/m3 d);
                                          282

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 9


       MaxToxSlough  =   maximum fraction of periphyton biomass lost by sloughing due to
                          given toxicant (fraction/d, 0.1);
       Toxiccmtwater   =   concentration of toxicant dissolved in water (ug/L); see (300);
       ECSOpisiodge     =   external concentration of toxicant at which there is 50% sloughing
                          (ug/L); and
       Biomassperi     =   biomass of given periphyton (g/m3); see (33).

Likewise, drift is greatly increased when zoobenthos are subjected to stress by sublethal doses of
toxic chemicals (Muirhead-Thomson, 1987), and that  is represented by a saturation-kinetic
formulation that utilizes an analogous sublethal toxicity parameter :

               Dislodge,  =Y   	Toxrcant^-DnftThreshold	
                            ^tmToxicantWater-DriftThreshold + EC5QGrowth

where:

       Toxicant water     =     concentration of toxicant in water (ug/L);
       DriftThreshold    =     the concentration of toxicant that initiates drift (ug/L); and
       EC50'Growth       =     concentration at which half the population is affected (ug/L).

These  terms are incorporated in the  respective periphyton washout (72) and zoobenthos drift
(130) equations.

9.3 External Toxicity

Chemicals that  are taken up very rapidly and those that have an external mode of toxicity, such
as affecting the gills directly, are best simulated with an external toxicity construct.  AQUATOX
has an alternative computation for CumFracKilled, when calculating toxic effects based  on
external  concentrations, using the two-parameter Weibull distribution as  in Christiensen and
Nyholm (1984):

                             CumFracKilled = 1 - exp(-kzEta)                        (429)

where:
                z   =  external concentration of toxicant (ug/L);
   CumFracKilled   =  cumulative fraction of organisms killed for a given period of exposure
                        (fraction/d), applied to equation (417);
         k and Eta   =  fitted parameters describing the dose response curve.

Rather than require the user to fit toxicological bioassay data to determine the parameters for k
and Eta,  these parameters are derived to fit the LC50 and the slope of the cumulative mortality
curve at  the LC50 (in  the manner of the RAMAS Ecotoxicology model, Spencer and Person,
1997):
                                          283

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                     CHAPTER 9
                                      k =
                                          -ln(0.5)
                                          LC50
                                               Eta
                                            (430)
                                 Eta =
-2-LC50-slope
     ln(0.5)
            (431)
where:      slope   =   slope of the cumulative mortality curve at LC50 (unitless).
            LC50   =   concentration where half of individuals are affected (ng/L).

AQUATOX assumes that each chemical's dose response curve has a distinct shape, relevant to
all organisms modeled. In this manner, a single "slope factor" parameter describing the shape of
the Weibull curve can be entered in the chemical record rather than requiring the user to derive
slope parameters for each organism modeled.  (Note, this is different than the shape parameter
used for internal toxicity.)

As shown  below, the slope of the  curve  at the LC50  is both a function of the shape of the
Weibull distribution and  also the magnitude of the  LC50 in question.  Figure 157 shows two
Weibull distributions with identical shapes, but with  slopes that are significantly different due to
the scales of the x axes.

            Figure 157. Weibull distributions with identical shapes, but different slopes.

        Weibull Distribution, LC50=1, Slope=1        Weibull Distribution, LC50=100, Slope=0.1

      100% i                                100%
      50%
   3

   3
   O
       0%
                                        HI
 3

 3
 O
    50%
                0.5     1     1.5
                  Concentration
              50     100    150
               Concentration
200
For this reason, rather than have a user enter "the slope at LC50" into the chemical  record,
AQUATOX asks the user to enter a "slope factor" defined as "the slope at LC50 multiplied by
LC50." In the above example, the user would enter a slope factor of 1.0 and then, given an
LC50 of 1 or an LC50 of 100, the above two curves would be generated.
                                          284

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 9


When modeling toxicity based on external concentrations, organisms are assumed to come to
equilibrium with external concentrations (or the toxicity is assumed to  be based  on external
effects to the organism).

Unlike the internal model, application factors  are not used to estimate sublethal effects when
calculating external toxicity.  Therefore, EC50 parameters do not need to be paired with LC50
values to calculate sublethal effects with this model.
                                         285

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                            CHAPTER 10
                              10.  ESTUARINE SUBMODEL

The estuarine  version  of AQUATOX  is intended  to  be an
exploratory model for evaluating the possible fate and effects
of toxic chemicals and other pollutants in estuarine ecosystems.
The model is not  intended to represent detailed,  spatially
varying  site-specific conditions,  but  rather  to be  used in
representing the potential behavior of chemicals under average
conditions.   Therefore,  it is best used as  a screening-level
model  applicable   to   data-poor  evaluations  in  estuarine
ecosystems.  However,  it  can  be  calibrated  for   different
estuaries.

Hourly tidal fluctuations are not included in the model;  the
native AQUATOX time-step is one day.  Because of this,  the
overall water volume of the estuary may be assumed to remain
constant  over  the  entire  simulation.    The  simplifying
assumption is  that  the  water  volume  of the  estuary  is  not
sensitive to the freshwater inflow.  The volumes and depths of
the fresh layer and the salt wedge do vary as a function of the
daily average tidal range and freshwater flows.

If  simulation   of  spatially-explicit,  site-specific  estuarine
conditions is  desired,  then a  multi-segment model can be
implemented with flow  among segments  provided by an external  hydrodynamic model.  The
numerous effects of salinity described below in  the context of the  estuarine submodel  are also
applicable to the classic and multi-segment versions of AQUATOX.
             AQUATOX Estuarine Submodel:
             Simplifying Assumptions

              • Estuary is a single segment that
                always has two well-mixed layers
              • The estuary has freshwater inflow
                from upstream and saltwater inflow
                from the seaward end (salt-wedge)
              • Water flows at the seaward end are
                estimated using the salt-balance
                approach
              • Effects of salinity on sorption are
                minor and are not modeled
              • Hourly tidal fluxes are not modeled
              • Daily average volume of the
                estuary is assumed to remain
                constant over time
              • The surface area of the lower layer
                is the same as the upper layer
              • Nutrient concentrations in
                inflowing seawater are assumed to
                be constant
              • Possible salinity effects on
                microbial degradation, hydrolysis,
                and photolysis are ignored.
10.1 Estuarine Stratification
As a general case, the estuarine system is assumed to always have two layers, although at times
the layers may be essentially identical because of respective thicknesses and turbulent diffusion.
The two layers are assumed to be a function of and  to vary with freshwater loadings and daily
tidal ranges.  The fraction of depth in the upper layer is adjusted to account for changing volumes
due to entrainment (flow of water from lower to upper layer;  see section 10.4), with a value of
1.5 based on inspection of published observations. IfResidFlow > 0 then:
                               Freshwater Head =
ResidFlow
   Area
                                                                                      (432)
                    FracUpper = 1.5
                                              FreshwaterHead
                                      TidalAmplitude + FreshwaterHead
where:
                                            286

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    CHAPTER 10
       Freshwater Head
       ResidFlow

       Area
       FracUpper
       TidalAmplitude
      height of freshwater (m/d);
      inflow  residual  flow of fresh  water minus daily evaporation,
      (m3/d) user inputs;
      area of the estuary taken at mean tide (m ).
      fraction of mean depth that is upper layer (unitless).
      tidal amplitude (m), see (434);
If ResidFlow <= 0 then FracUpper is taken as having a nominal value of 0.05.

The thicknesses of the two layers, and therefore the volumes of the two layers, may be calculated
as a function of FracUpper.
where:
       ThickUpper
       FracLower
       ThickLower
       MeanDepth
       VolumeUpper
       VolumeLower
       Area
    ThickUpper = FracUpper • MeanDepth
   ThickLower = MeanDepth - ThickUpper
      VolumeUpper = FracUpper • Area
      VolumeLower = FracLower • Area

=   thickness of the upper layer (m);
=   1 -FracUpper; see (432);
=   thickness of the lower layer (m);
=   mean depth of the estuary (m);
=   volume of the upper layer (m3);
=   volume of the lower layer (m3);
=   area of the estuary taken at mean tide (m2).
                                                                                 (433)
As shown in the formulations above, layer thicknesses are a function of the daily predicted tidal
range.  Given that the estuary's average daily volume is assumed to remain constant, to maintain
mass-balance of water AQUATOX  moves water from one layer to the next when thicknesses
change.  (This same movement of water occurs when the user specifies a variable thermocline
depth in a stratified lake or reservoir, see section 3.4 on "Modeling Reservoirs and Stratification
Options.")  In order to maintain biomass, nutrient, and toxicant mass-balance AQUATOX also
transfers  state variables located in the moving water from one layer to the next.  This transfer can
cause minor fluctuations that are visible in some estuarine-version results (e.g. wave-like patterns
in fish biomass predictions.) Such fluctuation is predominantly an artifact of the simple manner
in which  AQUATOX models estuarine water volume.

10.2 TidalAmplitude

Tidal amplitude is calculated using the general equation found in the Manual of Harmonic
Analysis and Prediction of Tides (U.S. Department of Commerce 1994):
       TidalAmplitude =
                        JCon.
             cos((SpeedCon • Hours) + EquilConJear - EpochCon),
(434)
                                         287

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 10


where:
       TidalAmplitude =   one-half the range of a constituent tide  (m);
       Con.           =   eight constituents of tidal range listed below;
       Ampcon.        =   user-input amplitude for each constituent (m);
       Node/actor     =   node factor for each constituent for  each year,  hard-wired  into
                          AQUATOX for 1970-2037 (deg.);
       Speed          =   speeds  of  each  constituent  in   (deg./hour),  hard-wired  into
                          AQUATOX for each relevant constituent;
       Hours          =   time since the start of the year (hours);
       Equil          =   equilibrium argument for each constituent for each year in degrees
                          for the  meridian of Greenwich, hard-wired into  AQUATOX for
                          1970-2037 (deg.);
       Epoch          =   user input phase lag for each constituent (deg.).
AQUATOX requires Amplitudes and Epochs for the following eight constituents of tidal range
for the modeled esturary, generally  available for download from NOAA databases.   These
"primary" constituents were found to have the largest effect on tidal range and will predict tidal
range to the precision as required by the estuarine submodel:

             M2 - Principal lunar semidiurnal constituent
             S2 - Principal solar semidiurnal constituent
             N2 -  Larger lunar elliptic semidiurnal constituent
             Kl -  Lunar diurnal constituent
             Ol -  Lunar diurnal constituent
             SSA - Solar semiannual constituent
             SA -  Solar annual constituent
             PI- Solar diurnal constituent
10.3 Water Balance

Water balance is computed using the salt balance approach (Ibafiez et al. 1999):
                    07      T  r,               ResidFlow                          ,.~-^
                    Saltwater Inflow =	-	                (435)
                                    Salinity Lower I Salinity Uppper -1

                                           ResidFlow
                        Outflow =	                    (436)
                                 1 - SalinityUpper I Salinity Lower
where:
       Saltwaterlnflow  =   water entering estuary from mouth of estuary, usually into lower
                           level but may be into upper level if evaporation exceeds freshwater
                           inflow (m3/d);
       Outflow         =   water leaving estuary at mouth (m3/d);
       ResidFlow       =   residual  flow of  fresh water;  may  be negative if evaporation

                                          288

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 10


                           exceeds freshwater inflow (m3/d);
       SalinityLower    =   salinity of lower layer at mouth of estuary (psu or %o);
       SalinityUpper    =   salinity of upper layer at mouth of estuary (psu or %o);

Programmatically, the system is modeled as a single constant-volume segment with two layers
and with freshwater inflow from upstream and saltwater inflow from the seaward end.  Ice cover
is not assumed on top of estuaries unless the average water temperature falls below -1.8 deg.C.


10.4 Estuarine Exchange

Saltwater inflow occurs to replace water that is admixed (entrained) from one layer (usually the
lower) to the other layer, producing the observed salinities of the two layers  at the mouth of the
estuary. (Note that this use of the term "entrainment" differs from the downstream entrainment
of organisms, e.g.  (132).)  This circulation  is much greater than any longitudinal mixing (see
Thomann  and  Mueller  1987).    Therefore,  effectively,  entrainment  is  the  equivalent  of
Saltwaterlnflow, but its derivation is informative:


                            „      Tr ,   Saltwaterlnflow
                           EntrainVel =	
                                            Area

                           VertAdvection = EntrainVel • Thick                        (437)


                            „             VertAdvection • Area
                           Lntramment =	
                                               Thick

where:
       EntrainVel           =      entrainment velocity of lower layer into upper layer (m/d);
       VertAdvectiveDisp    =      vertical advective dispersion (m /d);
       Entrainment         =      vertical flow as derived above (m3/d).

Transport of suspended and dissolved substances from the lower layer to the upper layer can then
be computed.  In a truly stratified estuary turbulent diffusion will be minimal, so we will set the
bulk mixing coefficient (BulkMixCoeff) to  0.1 m2/d  following the example of Koseff et  al.
(1993). However, when wind exceeds 3 m/s Langmuir circulation sets up with downwelling and
upwelling extending to about 3 m.  Therefore, if the thickness of the upper layer is less than 3 m
and the wind  speed is  greater than 3  m/s, then bulk mixing  is increased by a factor of 5.
Turbulent diffusion can then be computed for each dissolved and suspended compartment:
                                          289

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 10
                     BulkMixCoeff         .  i                                \
        TurbDiff   = ——	Langmmr • (Concco    en    - Cone           }
                       Volumeupper

                                                                                  (438)
                     BulkMixCoeff         .  t                                \
        TurbDifflower = ——	Langmmr • (Cone           - Cone     mentjower)
                       Volumelmer
        If ThickUpper < 3 and Wind > 3 then Langmuir = 5  else Langmuir = 1

where:
           TurbDiff       =  turbulent diffusion (g/m3-d);
           BulkMixCoeff  =  bulk mixing coefficient (0.1 m2/d);
           Langmuir       =  factor for greater mixing when wind equals or exceeds 3 m/s
                             (unitless);
           Volumeupper     =  volume of the upper layer (m3);
           Volumeiower     =  volume of the lower layer (m3);
           Cone          =  concentration of given compartment in a given layer (g/m3).
10.5 Salinity Effects

Mortality and Gamete Loss

Salinity that is less than or greater than threshold values increases mortality and gamete loss:


                if SalMin < Salinity < SalMax then SaltMort = 0
                if Salinity < SalMin then SaltMort = SalCoeffl • eSalMm-Sa"mty             (439)
                if Salinity > SalMax then SaltMort = SalCoeffl • Q^""'^"^
where:
       SalMin     =   minimum salinity below which effect is manifested (%o);
       Salinity     =   ambient salinity (%o);
       SalMax     =   maximum salinity above which effect is manifested (%o);
       SaltMort   =   mortality due to salinity (1/d);
       SalCoeffl   =   coefficient for effect of low salinity (unitless);
       SalCoeff2   =   coefficient for effect of high salinity (unitless);
       e           =   the base of natural logarithms (2.71828, unitless).

SaltMort is then applied  to mortality (112)  and gamete loss  (126). The model  assumes
reproduction is affected because eggs and sperm are not viable in abnormal salinities.

Other Biotic Processes

Salinity beyond the range of tolerance  for  a particular  process, including  photosynthesis,
ingestion, and respiration, will reduce the process:

                                         290

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                                      CHAPTER 10
where:
          if SalMin < Salinity < SalMax then SalEffect = 1
          if Salinity < SalMin then SalEffect = SalCoeffl • eSa»">"
          if Salinity > SalMax then SalEffect = SalCoeffl • eSalMa*-SaUm»

SaltEffect     =      effect of salinity on given process (unitless).
                                                                                     (440)
In general, the ranges of tolerance of abnormal salinities in animals, going from least tolerant to
most tolerant, affects reproduction,  ingestion, respiration, and mortality  in that order (Figure
158).     Respiration decreases because gill ventilation is depressed.  SaltEffect is applied to
ingestion (91), respiration (100), and photosynthesis (35) as appropriate.

                   Figure 158.  Effects of salinity on various animal processes.




1 9 -,

Effects of Salinity


= 1t^
o
-s n R I
| 0.8^
^ 0.6 { 	
£
3 0.4
re
CC



0.2

n
tl

4^
V y
"











•






/
(

V v
0 10 20 30 40 50 60 70





Salinity (ppt)
estion GameteLoss Respiration
n


^ C
-
h











4.5
4
3.5
3
2.5
2
1.5
-I

0.5
n
80

lortality




•a
T—
>
4-1
t
O
5







Sinking

Sinking of phytoplankton and suspended detritus also is affected by salinity, more so than by
temperature (Figure 159, Figure 160). However, because ambient salinity and temperature affect
sinking by controlling density, we will compute a density factor based on the effects of both
compared to the salinity and temperature of the observed  sinking rate (Thomann and Mueller
1987):
                                           291

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 10
     WaterDensity = 1 + < 1CT
(28.14 - 0.0735 • Temperature - 0.00469 • Temperature2}
+ (0.802 - 0.002 • Temperature)-(Salinity - 35)
(441)
                            DensityFactor =
                                            WaterDensity referen(
                           2/erence
                                             WaterDensity amhjent
                                                       (442)
where:
     Water Density reference

     WaterDensity ambient
     Temperature
    DensityFactor

    Sink
    KSed
     Thick
                                       KSed
                                Sink =	Densityr actor
                                       rr\-j •  -j        •/
                                       Thick
                                                       (443)
density of water at temperature and salinity of observed sinking
rate (kg/L);
density of water at ambient temperature and salinity (kg/L);
temperature of water (°C);
correction factor for water densities other than those at which
sinking rates were observed (unitless);
sinking rate of given suspended compartment (g/m3-d);
intrinsic settling rate (m/d);
thickness of water layer (m).
Figure 159. Correction factor for sinking as
a function of temperature.
                    Figure 160. Correction factor for sinking as a
                    function of salinity.
 1.01
0.99
       Effect of Temperature on Sinking
    0   5   10   15   20  25  30   35   40   45
                   Temperature
                                                 0.93
                            Effect of Salinity on Sinking
                                            292

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 10
Sorption

The influence of seawater or "salting out" does not cause major changes in sorption of organic
compounds (Schwarzenbach et al. 1993). It varies with the compound, with greater effect on
polar compounds, but is seldom measured. Therefore, it will be ignored at the present time.
Volatilization

Volatilization is affected by salinity, and can be represented by a linear increase in the Henry's
Law constant (Eqn. 330).  At 35%o salinity the average increase in the constant across tested
organic compounds is  1.4 compared to that of distilled water  (Schwarzenbach et al. 1993).
Applying this relationship:
                          HLCSaltFactor = 1 + 0.01143 • Salinity                      (444)
Estuarine Reaeration

Reaeration is affected by salinity, especially through calculation of the saturation level (O2Saf).
Salinity is included in the present formulation for O2Sat.  Computation of the depth-averaged
reaeration coefficient (KReaer) requires determination of the effects of both tidal  velocity and
wind velocity. Thomann and Fitzpatrick (1982, see also (Chapra 1997)) combine the two in one
equation:
                        ^Velocity  0.728 -JWind -0.317 -Wind+ 0.0372-Wind2
                         Thick312                     Thick
KReaer = 3 93 *      '  + "•'^"••v"""«   "•-»-' ~" "«* -r^.^ '^-" "«*        (445)
              TV •  1 3 / 2                     T~T7   7                        ^   *
The daily average tidal velocity can be computed by a variation of a formulation presented by
(Thomann and Mueller, 1987), substituting the spring tide harmonic for the diurnal harmonic:
                 Velocity =
                          ResidFlow Vel + TidalVel •  1 + 0. 5 • sin

                                                               12
                                             86400
                                                                       (446)
                                D  WZ7/   T7 1
                               ResidFlow Vel =	                           (447)
                                              XSecArea
                                XSecArea = Depth • Width                           (448)


                                          293

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                               CHAPTER 10
                                 TidalVel =
                TidalPrism
                 XSecArea
                          (449)
                           TidalPrism = 2.0- Amplitude • Area
                                                      (450)
where:
       Velocity      =
       Wind
       ResidFlowVel =
       Outflow      =
       TidalVel
       Day
       XSecArea     =
       Depth
       Width
       TidalPrism    =
       Amplitude     =
       Area         =
water velocity (m/s);
wind velocity (m/s), see (29);
residual flow velocity of fresh water (m/d);
water leaving estuary at mouth (m3/d), see (436);
mean tidal velocity (m/d);
day of year (d);
cross-sectional area of estuary (m );
mean water depth (m);
width of estuary (m);
the difference in water volume between low and high tides (m3);
tidal amplitude (m), see (434);
area of site (m2).
         Figure 161. Daily average water velocities based on freshwater flow and tidal flow.
                          Freshwater and Tidal Velocities
               0.25
               0.00
                                     10
                   15
                  Days
20
25
30
Salinity can significantly affect atmospheric exchange of carbon dioxide. For saline systems, the
equilibrium parameters of the CO2 system should be obtained using CO2SYS (Yuan, 2006) or
CO2calc (USGS, 2010) and the results used as inputs for CO2Equil in the estuarine AQUATOX
                                         294

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
CHAPTER 10
simulation. See section 5.6 for more information about this implementation.
Migration

Fish and pelagic invertebrates will also migrate vertically when the salinity level is not favorable.
Favorable salinity is defined as the range of salinity in which no ingestion effects occur for the
animal (from the minimum to the maximum salinity tolerances for ingestion).  If the salinity of
the current segment is outside that range, and the salinity  of the other segment is within the range
of favorable salinity,  the animal is predicted to migrate vertically  to  the other  segment.
Entrainment for pelagic invertebrates (movement due to water movement from the lower layer to
the upper layer as predicted by the salt balance model, see (437)) will also be set to zero if the
salinity in the upper layer is outside of the favorable range.  This can have significant effects on
shrimp populations, for example.
10.6 Nutrient Inputs to Lower Layer

Nutrient concentrations in ocean water flowing into the lower layer are set to temporally constant
levels, the assumption being that the  chemical composition of seawater remains relatively
uniform.  Nutrients and gasses in seawater may be edited using a button available in the initial
conditions  and loadings  screen for  each relevant  variable.   The  default nutrient and  gas
composition of seawater are set as follows:
       Ammonia:  0.02 mg/L    (Data from Galveston Bay, TX)
       Nitrate:     0.05 mg/L    (Data from Galveston Bay, TX)
       Phosphate:  0.03 mg/L    (Data from Galveston Bay, TX)
       Oxygen:    7.0  mg/L    (Default oxygen inflow to lower segment)
       CO2 :      90.0 mg/L    (Anthoni, 2006)
                                          295

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11
                                  11. REFERENCES
Allan,  J.D., 1995,  Stream  Ecology: Structure  and Function of Running Waters: London,
       Chapman & Hall, 388 p.
Anthoni,   Dr.    J.   Floor,      2006,   The    Chemical   Composition   of   Seawater,
       www.seafriends.org.nz/oceano/seawater.htm
APHA.  1995.  Standard  methods.   19th  Edition.  American  Public  Health  Association,
       Washington, DC.
Barber, M. C.  2001. Bioaccumulation and Aquatic System Simulator (BASS) User's Manual,
       Beta Test Version 2.1. Pages 76. U.S. Environmental Protection Agency, Athens, GA.
Barber, M. C.  2003.  A Review and  Comparison of Models for Predicting Dynamic Chemical
       Bioconcentration in Fish.  Environmental Toxicology and Chemistry, 22(9): 1963-1992.
Barbour, M. T., J. Gerritsen, B.  D.  Snyder,  and J. B. Stribling.  1999. Rapid  Bioassessment
       Protocols  for  Use   in   Streams   and   Wadeable   Rivers:   Periphyton,   Benthic
       Macroinvertebrates and Fish,  Second Edition. EPA 841-B-99-002, U.S.  Environmental
       Protection Agency; Office of Water, Washington, D.C.

Bartell, S. M., G. Lefebvre, G. Kaminski, M. Carreau, and K. R. Campbell.  1999. An Ecological
       Model  for  Assessing Ecological Risks  in Quebec Rivers, Lakes,  and Reservoirs.
       Ecological Modelling 124: 43-67.
Bartell, S.  M.,  R.  H. Gardner, and  R.  V.  O'Neill. 1992.  Ecological Risk Estimation. Lewis
       Publishers, Boca Raton, Florida.
Berry,  W., N. Rubinstein, B. Melzian, and B. Hill. 2003.  The Biological Effects of Suspended
       and Bedded  Sediments  (SABS) in  Aquatic  Systems:  A  Review.  Pages  58.  U.S.
       Environmental Protection Agency, Narragansett, RI.
Biggs,  B. J. F. 1996. Patterns in Benthic Algae of Streams. Pages 31-56 in R. J.  Stevenson, M. L.
       Bothwell,  and R. L.  Lowe, eds. Algal Ecology: Freshwater  Benthic  Ecosystems.
       Academic Press, San Diego.
Bowie, G. L., W. B. Mills, D. P. Porcella, C. L. Campbell,  J. R. Pagenkopf, G. L. Rupp, K. M.
       Johnson,  P.  W. H.  Chan,  and  S.  A. Gherini.  1985.  Rates, Constants, and Kinetics
       Formulations  in  Surface  Water Quality Modeling.  U.S.  Environmental  Protection
       Agency, Athens GA.
Broekhuizen, N., S. Parkyn,  and D. Miller. 2001. Fine Sediment Effects on Feeding and Growth
       in the Invertebrate Grazers Potamopyrgus antipodarum (Gastropoda, Hydrobiidae) and
       Deleatidium sp. (Ephemeroptera, Leptophlebiidae). Hydrobiologia 457:  125-132.
Cahill, T. M., I. Cousins, and D. Mackay. 2003. General Fugacity-Based Model to Predict The
       Environmental Fate of Multiple  Chemical Species.  Environ. Toxicology Chemistry 22:
       483-493.
                                         296

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Canham, C. D., J. S. Denslow, W. J. Platt, J. R. Runkle, T. A. Spies, and P. S. White. 1990. Light
       regimes beneath closed canopies and tree-fall gaps in temperate and tropical  forests.
       Canadian Jour. Forest Research 20:620-631.
CH2M HILL, Eco Modeling, Warren Pinnacle Consulting, and Boise City Public Works. 2008.
       Numeric Nutrient  Criteria Development for  the Lower Boise  River Using  the
       AQUATOX  Model.  U.S.  Environmental  Protection  Agency,  Office of  Water,
       Washington, DC.
Chapra, S. C. 1997. Surface Water Quality Modeling. McGraw-Hill, New York NY.
Chapra, S. C.,  G.  J.  Pelletier,  and  H. Tao. 2007. QUAL2K:  A  Modeling Framework for
       Simulating River and Stream Water  Quality, Version 2.07:  Documentation and Users
       Manual. Civil and Environmental Engineering Dept, Tufts University, Medford, MA.
Clarke, D. G., and D. H. Wilber. 2000. Assessment of Potential Impacts of Dredging Operations
       Due  to  Sediment  Resuspension.  Pages  14.   U.S.  Army Engineer Research  and
       Development Center, Vicksburg, MS.
Collins, C. D., and J. H. Wlosinski.  1983.  Coefficients for Use in the U.S. Army Corps of
       Engineers  Reservoir  Model,  CE-QUAL-R1.  Tech.  Rept.  E-83-,  Environmental
       Laboratory, U.S. Army Engineer Waterways Experiment Station, Vicksburg, Miss.
Connolly,  J. P.  1991.  Application  of  a  food  chain model  to polychlorinated biphenyl
       contamination of the lobster and winter flounder food chains in New Bedford Harbor.
       Environ. Sci. Technol. 25:760-770.
Crowe,  A., and J. Hay. 2004. Effects of Fine Sediment on River Biota. Pages 35.  Cawthron
       Institute, prepared for  Motueka Integrated Catchment Management Program, Nelson,
       New Zealand.
DeAngelis, D. L., S. M. Bartell, and A. L. Brenkert. 1989. Effects of nutrient cycling and food-
       chain length on resilience. American Naturalist 134: 778-805.
Di Toro, D. M. 2001. Sediment Flux Modeling. Wiley-Interscience, New York.
Di Toro, D. M., J. J. Fitzpatrick, and R. V. Thomann. 1983. Water Quality Analysis Simulation
       Program  (WASP)   and Model  Verification   Program  (MVP)  -  Documentation.
       Hydroscience, Inc. for U.S. EPA, Duluth, MN.
Di Toro, D. M., P. Paquin, K.  Subburamu, and D. A. Gruber.  1990. Sediment oxygen demoand
       model: methane and ammonia oxidation. Journal Environmental Engineering ASCE 116:
       945-986.
Doisy, K. E., and C. F. Rabeni. 2004. Effects  of Suspended Sediment on Native Missouri Fishes:
       A Literature Review and Synthesis. Department of Fisheries and Wildlife,  University of
       Missouri, Columbia MO.
Donigian,  A. S., J. T.  Love, J. S.  Clough, R. A. Park,  J. N. Carleton, P. A. Cocca, and J. C.
       Imhoff 2005. Nutrient Criteria Development with a Linked Modeling System: Watershed
       and Ecological Model  Application and  Linkage,  in 2005  TMDL Conference.  Water
       Environment Federation, Philadelphia  PA.
                                         297

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 11


Effler, S. W., C. T. Driscoll, S. M. Doerr, C. M. Brooks, M. T. Auer, B. A. Wagner, J. Address,
      W. Wang, D. L. Johnson, J. Jiao, and S. G. Dos Santos. 1996. 5. Chemistry. Pages 263-
      283 in S. W.  Effler, ed. Limnological and Engineering Analysis of a Polluted Urban
      Lake. Springer, New York.
Egglishaw, HJ. 1972. An Experimental Study of the Breakdown of Cellulose in Fast-Flowing
      Streams.  In Melchiorri-Santolini, U.,  and J.W. Hopton, Eds.,  Detritus and Its Role in
      Aquatic Ecosystems, Proceedings  of  an  IBP-UNESCO Symposium, Mem.  1st. Ital.
      Idrobiol. 29 Suppl.:405-428.
Elser, J.  J.,  R. W. Sterner, A. E. Galford,  T. H. Chrzanowski, M.  P.  Stainton,  and D.  W.
      Schindler.  2000. Pelagic C:N:P  Stoichiometry in a  Eutrophied Lake: Responses to a
      Whole-Lake Food- Web Manipulation. Ecosystems 3: 293-307.
Engelund, F.  and E. Hansen.  1967. A Monograph of Sediment Transport in Alluvial Streams.
      Teknisk Vorlag, Copenhagen, Denmark.
Environmental Laboratory.  1982.  CE-QUAL-R1:  A Numerical One-Dimensional Model of
      Reservoir Water Quality;  A User's Manual.  Instruction Report E-82-1, U.S. Army
      Engineers Waterways Experiment Station,  Vicksburg,  Miss.
Fagerstrom, T., and B. Asell. 1973. Methyl Mercury Accumulation in an Aquatic Food Chain,
      A Model and Some Implications for Research  Planning. Ambio,  2(5): 164-171.
Fagerstrom,  T., R. Kurten  ,and B. Asell. 1975. Statistical  Parameters as Criteria in Model
      Evaluation: Kinetics of Mercury Accumulation in Pike&ox lucius.  Oikos 26:109-116.
Flosi, G., S. Downie, J. Hopelain, M. Bird, R. Coey,  and B. Collins. 1998. California Salmonid
      Stream Habitat Restoration Manual, Third Edition. Pages 495. CA Department of Fish
      and Game, Inland Fisheries Division, Sacramento, CA.
Fogg, G.E.,  C. Nalewajko, and  W.D.  Watt.  1965. Extracellular Products of  Phytoplankton
      Photosynthesis. Proc. Royal Soc. Biol, 162:517-534.
Ford, D.E., and K.W. Thornton.   1979.  Time  and Length Scales for the One-Dimensional
      Assumption  and  Its  Relation  to Ecological  Models.   Water  Resources Research
Freidig, A.P., E.A. Garicano, and F.J.M. Busser.  1998. Estimating Impact of Humic Acid on
       Bioavailability and Bioaccumulation  of Hydrophobic  Chemicals  in Guppies Using
       Kinetic Solid-Phase Extraction. Environmental Toxicology and Chemistry, 17(6):998-
       1004.
Frey, H.C. ,  and S. R. Patil. 2001. Identification and Review of Sensitivity Analysis Methods.
       Paper read at Sensitivity Analysis Methods, June 11-12, 2001, at North  Carolina State
       University, Raleigh NC.
Ganf, G.G, and  P. Blazka.   1974.   Oxygen Uptake,  Ammonia and Phosphate Excretion by
       Zooplankton in a Shallow Equatorial Lake (Lake Goerge, Uganda). Limnol. Oceanog.
       19(2):313-325.
Giesy, J. P., and K. Kannan. 2001. Global Distribution of Perfluorooctane  Sulfonate in Wildlife.
       Environ. Sci. Technol. 35: 1339-1342.
                                         298

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Gilek, M., M. Bjork, D. Broman, N. Kautsky, and C. Naf. 1996. Enhanced Accumulation of PCB
       Congeners by Baltic Sea Blue Mussels, Mytilus edulis, with Increased Algae Enrichment.
       Environmental Toxicology and Chemistry, 15(9):1597-1605.
Gobas, F.A.P.C. 1993. A Model  for Predicting  the Bioaccumulation Hydrophobic Organic
       Chemicals in Aquatic Food-webs: Application to Lake Ontario. Ecological Modelling.,
       69:1-17.
Gobas, F.A.P.C., E.J. McNeil, L. Lovett-Doust, and G.D. Haffner.  1991. Bioconcentration of
       Chlorinated  Aromatic Hydrocarbons in Aquatic Macrophytes (Myriophyllum spicatum).
       Environmental Science & Technology, 25:924-929.
Gobas, F. A. P. C., and H.  A. Morrison. 2000. Bioconcentration and Biomagnification in the
       Aquatic Environment. Pages 189-231 in R.  S. Boethling and D. Mackay, eds. Handbook
       of Property Estimation Methods for Chemicals. Lewis, Boca Raton, Florida
Gobas, F.A.P.C., M.N. Z-Graggen, X. Zhang.  1995.  Time response  of the  Lake  Ontario
       Ecosystem to Virtual Elimination  of PCBs. Environmental  Science & Technology,
       29(8):2038-2046.
Gobas, F.A.P.C., Xin Zhang,  and Ralph  Wells.  1993. Gastrointestinal Magnification: The
       Mechanism  of Biomagnification and Food  Chain Accumulation of Organic Chemicals.
       Environmental Science & Technology, 27:2855-2863.
Gobas, F. A. P. C.,  and X. Zhang. 1994. Interactions of Organic Chemicals with Particulate and
       Dissolved Organic Matter in the Aquatic Environment.  Pages 83-91 in J. L. Hamelink, P.
             F. Landrum,  H.  L.  Bergman,  and W. H.  Benson,  editors. Bioavailability:
       Physical,
       Chemical, and Biological Interactions. Lewis Publishers, Boca Raton FL.
Godshalk, G.L., and J.W. Barko.  1985.  Chapter 4,  Vegetative Succession and Decomposition in
       Reservoirs.  In D. Gunnison (ed.), Microbial Processes in Reservoirs,  Dordrecht: Dr. W.
       Junk Publishers, pp. 59-77'.
Groden, W.T.  1977. Modeling Temperature and Light Adaptationm of Algae. Report 2, Cenyer
       for Ecological Modeling, Rensselaer Polytechnic Institute, Troy, New York, 17 pp.
Gunnison, D., J.M. Brannon, and  R.L. Chen.   1985.  Chapter  9,  Modeling Geomicrobial
       Processes in Reservoirs.  In D. Gunnison (ed.), Microbial Processes  in Reservoirs.
       Dordrecht: Dr. W. Junk Publishers, pp. 155-167.
Hanna, M.  1990.  Evaluation of Models Predicting Mixing Depth.  Can. J. Fish. Aquat. Sci.,
       47:940-947.
Harris, G.P.  1986.  Phytoplankton Ecology: Structure, Function and Fluctuation.  Chapman and
       Hall, London, 384 pp.
Hawker,  D.W.  and D.W. Connell.   1985.  Prediction of  Bioconcentration Factors  Under
       Non-Equilibrium Conditions. Chemosphere 14(11/12):1835-1843.
Hemond, H. F., and E. J. Fechner. 1994.  Chemical Fate and Transport in the Environment.
       Academic Press, New York.
                                         299

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Henley, W. E., M. A. Patterson, R. J. Neves, and A. D. Lemly. 2000. Effects of Sedimentation
       and Turbidity on Lotic Food Webs: A Concise Review for Natural Resource Managers.
       Reviews in Fisheries Science 8: 125-139.
Hessen, D. O.  and L. J. Tranvik, editors.  1998.  Aquatic Humic Substances: Ecology and
       Biochemistry. Springer-Verlag, Berlin.
Hewett, S.W., and  B.L. Johnson.  1992. Fish Bioenergetics 2 Model.  Madison, Wisconsin:
       University of Wisconsin Sea Grant Institute, 79 pp.
Hill, I.R., and P.L. McCarty.  1967. Anaerobic Degradation of Selected Chlorinated Pesticides.
       Jour. Water Poll. Control Fed. 39:1259.
Hill, W.R., and Napolitano, G.E. 1997. PCB Congener Accumulation by Periphyton, Herbivores,
       and Omnivores. Archives Environmental Contamination Toxicology., 32:449-455.
Hoggan, D.H.  1989. Computer-Assisted Floodplain Hydrology and Hydraulics, McGraw-Hill
       New York, 518pp.
Home, A.J., and C.R. Goldman.  1994.  Limnology - 2nd edition. McGraw-Hill, New York, 576
       pp.
Howick, G.L., F. deNoyelles, S.L.  Dewey, L. Mason, and D. Baker.  1993.  The Feasibility of
       Stocking  Largemouth  Bass  in  0.04-ha  Mesocosms  Used  for Pesticide Research.
       Environmental Toxicology and Chemistry, 12:1883-1893.
Hrbacek, J. 1966. A Morphometrical Study of Some Backwaters and Fish Ponds in Relation to
       the Representative Plankton Samples. In Hydrobiological Studies 7,  J.  Hrbacek, Ed.,
       Czechoslovak Academy of Sciences, Prague, p. 221-257.
Hudon, C., S. Lalonde, and P. Gagnon.  2000.  Ranking the Effects of Site Exposure,  Plant
       Growth Form, Water Depth, and Transparency on Aquatic Plant Biomass. Can. J.  Fish.
       Aquat. Sci. 57(Suppl. l):31-42.
Hutchinson,  G.E.   1957.   A  Treatise  on  Limnology,  Volume I, Geography, Physics,  and
       Chemistry. John Wiley & Sons, New York, 1015 pp.
Hutchinson, G.E. 1967. A Treatise on Limnology, Volume II, Introduction to Lake Biology and
       the Limnoplankton. Wiley & Sons, New York, 1115 pp.
Ibafiez, C., J.  Saldafia,  and N. Prat. 1999. A Model to Determine the Advective Circulation in a
       Three  Layer, Salt Wedge  Estuary: Application to the Ebre River Estuary. Estuarine,
       Coastal and Shelf Science 48: 271-279.
Iman, R. I, and W. J. Conover. 1982. A distribution-free approach to inducing rank correlation
       among input variables. Communications in Statistics Ell: 311-334.
Imboden, D.M.   1973.  Limnologische Transport- und Nahrstoffmodelle. Schweiz. Z. Hydrol.
       35:29-68.
Johanson,  R.C.,  J.C. Imhoff,  and  H.H. Davis, Jr.  1980.   Users Manual for Hydrological
       Simulatiuon Program Fortran (HSPF).  U.S. Environmental Protection Agency, Athens
       Environmental Research Laboratory, EPA-600/9-80-015, 678 pp.
Jorgensen, L.  A., S. E. Jorgensen, and S. N. Nielsen. 2000. ECOTOX: Ecological Modelling and
       Ecotoxicology. in. Elsevier Science.
                                         300

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


J0rgensen, S.E.  1976.  A Eutrophication Model for a Lake. Ecol. Modelling, 2:147-165.
J0rgensen,  S.E., H.F. Mejer,  M.  Friis,  L.A.  J0rgensen,  and J.  Hendriksen  (Eds.).  1979.
      Handbook  of Environmental  Data  and  Ecological Parameters.    Copenhagen:
      International Society of Ecological Modelling.
J0rgensen, S. E. 1986. Fundamentals of Ecological Modelling. Elsevier, Amsterdam.
Junge, C.O. 1966. Depth  distributions for  quadratic  surfaces and other configurations.  In:
      Hrbacek, J. (Ed.): Hydrobiological Studies. Vol.  1, Academia, Prague, pp. 257-265.
Jupp, B.P., and D.H.N. Spence.  1977a. Limitations on Macrophytes in a Eutrophic Lake, Loch
      Lev en I.  Effects of Phytoplankton.  Journal Ecology, 65:175-186.
Jupp, B.P., and D.H.N. Spence.  1977b. Limitations on Macrophytes in a Eutrophic Lake, Loch
      Lev en II. Wave Action, Sediments, and Waterfowl Grazing. Journal Ecology, 65:431-
      446.

Kaller, M.  D.,  and K. J. Hartman. 2004. Evidence of a Threshold Level of Fine  Sediment
      Accumulation for Altering Benthic Macroinvertebrate Communities. Hydrobiologia 518:
      95-104.
Kannan, K., J. Koistinen, K. Beckman, T. Evans, J. F.  Gorzelany, K. J. Hansen, P. D. Jones, E.
      Helle,  M. Nyman, and J. P.  Giesy. 2001.  Accumulation of Perfluorooctane Sulfonate in
      Marine Mammals. Environ. Sci. Technol. 35: 1593-1598.

Karickhoff, S.W., and K.R. Morris. 1985.  Sorption  Dynamics of Hydrophobic Pollutants in
      Sediment Suspensions. Environmental Toxicology and Chemistry, 4:469-479.

Key, B. D., R.  D.  Howell, and C.  S. Criddle. 1998. Defluorination of Organofluorine Sulfur
      Compounds by Pseudomonas sp. Strain D2. Environ. Sci. Technol. 32: 2283-2287.

Kitchell, J.F.,  J.F. Koonce, R.V. O'Neill, H.H. Shugart, Jr., J.J Magnuson, and R.S. Booth.  1972.
      Implementation of a Predator-Prey Biomass Model for Fishes. Eastern Deciduous Forest
      Biome , International Biological Program,  Report 72-118. 57 pp.
Kitchell, J.F.,  J.F. Koonce, R.V. O'Neill, H.H. Shugart, Jr., J.J Magnuson, and R.S. Booth.  1974.
      Model offish biomass dynamics. Trans. Am.  Fish. Soc.  103:786-798.
Koelmans, A.A., S.F.M. Anzion, and L. Lijklema.  1995. Dynamics of Organic Micropollutant
      Biosorption  to Cyanobacteria  and Detritus.  Environmental  Science  & Technology,
      29(4):933-940.

Koelmans,  A.A., and E.H.W.  Heugens. 1998. Binding Constants  of Chlorobenzenes  and
      Polychlorobiphenyls for Algal Exudates. Water Science Technology, 37(3):67-73.
Koelmans,  A. A.,  A.  Van der Heidje, L. M. Knijff,  and R.  H. Aalderink. 2001. Integrated
      Modelling of Eutrophication and  Organic  Contaminant Fate &   Effects  in Aquatic
      Ecosystems. AReview. Water Research 35: 3517-3536.
Koseff, J. R.,  J. K. Holen, S. G. Monismith,  and  J.  E. Cloern. 1993. Coupled effects of vertical
      mixing and benthic grazing  on phytoplankton  populations in shallow, turbid estuaries.
      Journal of Marine Research 51: 843-868.
Kremer, J.N., and S.W. Nixon.  1978. A Coastal Marine Ecosystem.  Springer-Verlag, New
      York, N.Y., 217pp.
                                         301

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Krenkel,  P.A., and G.T. Orlob.  1962.  Turbulent Diffusion and the Reaeration Coefficient.
      Proc. ASCE, Jour. San. Eng. Div., 88 (SA 2):53-83.
Krone, R. B.  1962.   Flume Studies  of The  Transport of Sediment in Estuarial Shoaling
      Nrocesses: Final Report, Hydraulic Engr. and San. Engr., Research Lab., University of
      California at Berkeley.
Lam, R.K., and B.W. Frost.  1976.  Model of Copepod Filtering Responses to Changes in Size
      and Concentration of Food. Limnol. Oceanogr. 21:490-500.
Landrum, P. F., S. R. Nihart, B. J. Eadie, and W. S. Gardner. 1984. Reverse-phase Separation
             Method for Determining Pollutant Binding to Aldrich Humic Acid and Dissolved
             Organic Carbon of Natural Waters. Environ. Sci. Technol. 18:187-192.
Lange, C. C. 2000. The Aerobic Biodegradation of N-EtFOSE Alcohol by the Microbial Activity
      Present in  Municpal Wastewater Treatment Sludge. 3M Environmental Laboratory, St.
      Paul, MN.
Larkin, G. A., and P. A.  Slaney. 1996. Calibration of a Habitat Sedimentation Indicator for Use
      in Measuring the Effectiveness of Watershed Restoration Treatments. Pages  14. Province
      of British Columbia, Ministry of Environment, Lands and Parks, and Ministry of Forests,
      Vancouver, B.C., Canada.
Larsen, D.P., H.T. Mercier, and K.W. Malueg.   1973.  Modeling Algal  Growth Dynamics in
      Shagawa Lake, Minnesota, with Comments Concerning Nrojected Restoration of the
      Lake. In EJ. Middlebrooks, D.H. Falkenborg, and T.E. Maloney (Eds.). Modeling the
      Eutrophication Process. Logan, Utah: Utah State University, pp. 15-32.
Le  Cren, E.P.,  and R.H.  Lowe-McConnell (Eds.).  1980.   The Functioning of Freshwater
      Ecosystems. Cambridge: Cambridge University Press, 588 pp.
Legendre, P., and L. Legendre. 1998. Numerical Ecology. Elsevier Science BV, Amsterdam.
Lehman,  J.T.,  D.B.  Botkin,  and  G.E.  Likens.  1975.  The Assumptions  and Rationales of a
      Computer  Model  of Phytoplankton  Population  Dynamics.   Limnol.  and Oceanogr.
      20(3):343-364.
Leidy,  G.R., and R.M.  Jenkins.   1977.   The Development of Fishery Compartments and
      Population Rate Coefficients for Use in Reservoir Ecosystem Modeling.  Contract Rept.
      CR-Y-77-1, U.S. Army Engineer Waterways Experiment Station, Vicksburg Mississippi,
      134pp.
Leung, D.K. 1978. Modeling the Bioaccumulation of Pesticides in Fish. Report N. 5,  Center for
      Ecological Modeling, Resselaer Polytechnic Institute, Troy, N. Y.
Liss, P.S., and P.G. Slater. 1974. Flux of Gases  Across the Air-Sea Interface. Nature, 247:181-
      184.
Lyman,  W.J.,  W.F. Reehl, and D.H.  Rosenblatt.   1982.   Handbook of Chemical Property
      Estimation Methods. McGraw-Hill, New York.
Maaret, K., K, Leif, and H. Bjarne.  1992. Studies on the Partition Behavior of Three Organic
      Hydrophobic Pollutants in Natural Humic Water. Chemosphere, 24(7):919-925.
                                         302

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Mabey, W., and T. Mill. 1978. Critical Review of Hydrolysis of Organic Compounds in Water
       Under Environmental Conditions.  J. Phys. Chem. Ref. Data, 7:383-415.
Macek, K.J., M.E. Barrows, R.F. Frasny, and B.H. Sleight III. 1977.  Bioconcentration of 14C-
       Pesticides  by Bluegill Sunfish During Continuous  Aqueous  Exposure.  In Structure-
       Activity  Correlations  in  Studies  of  Toxicity and Bioconcentration  with  Aquatic
       Organisms, G.D. Veith and D. Konasewick, eds.
Mackay, D., H. Puig, and L.S. McCarty.  1992. An Equation Describing the Time Course and
       Variability in Uptake  and Toxicity of Narcotic  Chemicals to Fish.  Environmental
       Tixcology and Chemistry, 11:941 -951.
Maloney, K.  O.,  and J. W. Feminella. 2006.  Evaluation of single- and multi-metric  benthic
       macroinvertebrate  indicators of catchment disturbance over time at the  Fort Benning
       Military Installation, Georgia, USA. Ecological Indicators  6:469-484.

Mancini, J.L. 1983.  A  Method for Calculating Effects on Aquatic Organisms of  Time Varying
       Concentrations.  Water Res. 10:1355-1362.
Marmorek,  D. R., R. M. MacQueen, C. H. R.  Wedeles, J. Korman, P. J. Blancher, and D. K.
       McNicol.  1996.  Improving pH and Alkalinity Estimates for Regional-scale Acidification
       Models: Incorporation of Dissolved  Organic Carbon. Can. J. Fish. Aquat.  Sci 53: 1602-
       1608.
Martin, J. L., R. A. Ambrose, and T. A. Wool. 2006. WASP7Benthic Algae -Model Theory and
       User's Guide. U.S. Environmental Protection Agency, Athens, Georgia.
Martin, J. W., S. A. Mabury, K. R. Solomon, and D. C. G. Muir. 2003a. Bioconcentration and
       Tissue Distribution of Perfluorinated Acids in Rainbow Trout (Oncorhyncus mykiss).
       Environmental Toxicology and Chemistry 22: 196-204.
Martin, J. W., S.  A.  Mabury, K. R. Solomon, and D. C. G. Muir. 2003b. Dietary  Accumulation
       of   Perfluorinated  Acids  in  Juvenile   Rainbow   Trout  (Oncorhynchus   mykiss).
       Environmental Toxicology and Chemistry 22: 189-195.
Mauriello, D. A.,  and R. A. Park. 2002. An adaptive framework for ecological assessment and
       management.  Pages 509-514  in  A. E. Rizzoli  and A. J. Jakeman,  eds. Integrated
       Assessment and Decision  Support. International Environmental Modeling and Software
       Society, Manno  Switzerland.
Mayer, F. L., Jr., and M. R. Ellersieck.  1986. Manual of Acute Toxicity: Interpretation  and
       Data Base for 410 Chemicals and 66 Species of Freshwater Animals: U.S. Department of
       Interior Fish and Wildlife Service, Resource Publication 160; Wasjington, D.C.
Mayio, A.E., and G.H. Grubbs. 1993.  Nationwide Water-Quality Reporting to the Congress as
       Required Under Section 305(b) of the Clean Water Act.  In National  Water Summary
       1990-91, Water Supply Paper 2400; Washington, D.C.: U.S. Geological Survey, pp. 141-
       146.
McCarty, L.S., G.W. Ozburn,  A.D. Smith,  and D.G. Dixon.  1992. Toxicokinetic Modeling of
       Mixtures  of  Organic  Chemicals.  Environmental Toxicology and Chemistry,  11:1037-
       1047.

                                         303

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


McConnaughey,  T. A., J. W. LaBaugh, D. O. Rosenberry, R. G. Striegl, M. M. Reddy, P. F.
       Schuster,  and V. Carter.  1994. Carbon Budget for a Groundwater-fed Lake: Calcification
       Supports Summer Photosynthesis. Limnol. Oceanog. 39: 1319-1332.
Mclntire, C.D.  1968. Structural Characteristics of Benthic  Algal Communities in Laboratory
       Streams. Ecology 49(3):520-537.
Mclntire, C.D. 1973.  Periphyton Dynamics in Laboratory Streams: a Simulation Model and Its
       Implications. Ecological Monographs 43(3):399-419.
Mclntire, C.D., and J.A. Colby.  1978.  A Hierarchical Model of Lotic Ecosystems.  Ecological
       Monographs 48:167-190.
McKay, M.D., WJ. Conover, and RJ.  Beckman.  1979. A  Comparison of Three Methods for
       Selecting  Values of Input Variables in the Analysis of Output from a Computer Code.
       Technometrics 211:239-245.
McKim,  J.M., G.F.  Olson, G.W. Holcombe,  and E.P.  Hunt.  1976. Long-Term Effects  of
       Methylmercuric Chloride on Three Generations of Brook Trout (Salvelinus fontinalis):
       Toxicity,  Accumulation, Distribution, and Elimination. Journal Fisheries Research
       Board Canada, 33(12):27226-2739.
McKim, J.M., P. Schmeider, and G. Veith.  1985. Absorption Dynamics of Organic Chemical
       Transport  Across   Trout  Gills  as  Related  to Octanol-Water  Partition  Coefficient.
       Toxicology and Applied Pharmacology, 77:1-10.
Megard, R.O., W.S. Comles, P.O. Smith, and A.S. Knoll. 1979. Attenuation of Light and Daily
       Integral Rates  of Photosynthesis  Attained by Planktonic Algae.   Limnol.  Oceanogr.,
       24:1038-1050.

Moloney, C. L. and J. G. Field. 1989. General allometric equations for rates of nutrient uptake,
       ingestion, and  respiration in plankton organisms. Limnol. Oceanogr. 35:1290-1299.

Moody, C.  A.,  and  J.  A. Field.  2000. Perfluorinated  Surfactants  and  the Environmental
       Implications of their Use in Fire-Fighting Foams. Environ.  Sci. Technol. 34: 3864-3870.
Muirhead-Thomson, R.C.   1987. Pesticide Impact on Stream Fauna with Special Reference to
       Macroinvertebrates. Cambridge: Cambridge University Press, 275 pp.
Mulholland,  P. J., J. W. Feminella, B. G. Lockaby, and G.  L. Hollon. 2007. Riparian Ecosystem
       Management  at Military Installations:  Determination of Impacts  and  Evaluation  of
       Restoration and Enhancement Strategies. Oak  Ridge National Laboratory,  Oak Ridge
       TN.
Mullin, M.M.  1963.   Some Factors Affecting the Feeding of Marine Copepods of the Genus
       Calanus.  Limnol. Oceanogr. 8:239-250.
Mullin, M.M., E.F. Stewart, and FJ. Foglister.  1975. Ingestion by Planktonic Grazers  as a
       Function of Concentration of Food. Limnol. Oceanog. 20:259-262.
Murphy, T. P., K. J. Hall, and I. Yesaki. 1983. Coprecipitation of Phosphate with Calcite in a
       Naturally Eutrophic Lake. Limnol. Oceanog. 28: 58-69.

Nalewajko, C.   1966.  Photosynthesis and Excretion in  Various Planktonic Algae.   Limnol.
       Oceanogr., 11:1-10.

                                         304

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Neal, C. 2001. The potential for phosphorus pollution remediation by calcite precipitation in UK
       freshwaters. Hydrology and Earth System Sciences, 5:  119-131.
Newcombe,  C. P. 2003.  Impact Assessment  Model  for  Clear Water Fishes  Exposed to
       Excessively Cloudy Water.  Journal of the American Water  Resources Association
       (JAWRA) 39: 529-544.
Nichols, D.S., and D.R. Keeney.  1976.  Nitrogen Nutrition of Myriphyllum spicatum: Uptake
       and Translocation of 15N by Shoots and Roots. Freshwater Biology 6:145-154.
O'Connor, D.J., and J.P. Connolly.  1980. The Effect of Concentration of Adsorbing Solids on
       the Partition Coefficient. Water Research, 14:1517-1523.
O'Connor, D.J., and W.E. Dobbins.  1958. Mechanism of Reaeration in Natural Streams. ASCE
       Transactions, pp. 641-684, Paper No. 2934.
O'Connor, D.J., J.L. Mancini,  and J.R. Guerriero.  1981.  Evaluation of Factors Influencing the
       Temporal Variation of Dissolved Oxygen in the New York Bight, Phase II.  Manhattan
       College, Bronx, New York
O'Neill, R.V.  1969.  Indirect Estimation of Energy Fluxes in Animal Food Webs.  Jour.
       Theoret. Biol, 22:284-290.
Odum, E. P.  1971. Fundamentals of Ecology. Third edition. W.B. Saunders Co., Philadelphia.
Odum, E.P.,  and A. A. de la Cruz.  1963. Detritus as a Major Component of Ecosystems. Amer.
       Inst. Biol. Sci. Bull, 13:39-40.
OECD.   2002.    Co-operation   On   Existing    Chemicals:    Hazard   Assessment  of
       Perfluorooctaneulfonate (PFOS) and Its Salts.  Organization for Economic Co-operation
       and Development.
Oliphant, A. J., J. W. Rose III, C. S.  B. Grimmond, and H.-P. Schmid.  2006. Observations of
       Canopy Light Penetration and Net Ecosystem Exchange of CO2 under different sky
       conditions   in  a  Mid-Western Mixed  Deciduous  Forest,  in  27th Conference  on
       Agricultural and Forest Meteorology. American Meteorological Society, San Diego.
Oliver, B. G., and A. J. Niemi. 1988. Trophodynamic Analysis of Polychlorinated Biphenyl
       Congeners and Other Chlorinated Hydrocarbons in the Lake Ontario Ecosystem. Environ.
       Sci. Technol. 22:388-397.
O'Neill, R.V., D.L. DeAngelis, J.B. Waide, and T.F.H. Allen.  1986.  A Hierarchical Concept of
       the Ecosystem.  Princeton University Press, Princeton, N. J.
O'Neill, R.V., R.A. Goldstein, H.H. Shugart,  and  J.B. Mankin.  1972.   Terrestrial Ecosystem
       Energy Model.  Eastern Deciduous Forest Biome,  International Biological  Program
       Report 72-19.
Osmond, D.  L., D.E.  Line, J.A.  Gale, R.W. Gannon, C.B. Knott,  K.A. Bartenhagen,  M.H.
       Turner, S.W. Coffey, J. Spooner, J. Wells, J.C. Walker, L.L. Hargrove, M.A. Foster, P.O.
       Robillard,  and D.W.  Lehning.  1995.  WATERSHEDSS:  Water,   Soil and  Hydro-
       Environmental Decision Support System.
Otsuki, A., and R. G.  Wetzel.  1972.  Coprecipitation  of Phosphate with Carbonates in a Marl
       Lake. Limnology & Oceanography 17: 763-767.

                                         305

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Owens,  M., R.W. Edwards, and J.W. Gibbs.  1964.   Some Reaeration Studies ion Streams.
      Internal. Jour. Air Water Poll. 8:469-486.
Palisade Corporation. 1991. Risk Analysis and Simulation Add-In for Lotus 1-2-3. Newfield New
      York, 342 pp.

Park, K., A. Y. Kuo, J.  Shen, and J. M.  Hamrick. 1995. A Three-Dimensional Hydrodynamic-
      Eutrophication Model (HEM-3D): Description of Water Quality and Sediment Process
      Submodels. Special Report in Applied Marine Science and Ocean Engineering No. 327.
Park, R.A.  1978. A Model for Simulating Lake Ecosystems.  Center for Ecological Modeling
      Report No. 3, Rensselaer Polytechnic Institute, Troy, New York, 19 pp.
Park, R. A., and C. D. Collins. 1982. Realism in Ecosystem Models. Perspectives in Computing
      2: 18-27.
Park, R.A.   1984.   TOXTRACE:  A Model  to Simulate the Fate and Transport of  Toxic
      Chemicals in Terrestrial and Aquatic Environments. Acqua e Aria, No. 6, p. 599-607 (in
      Italian).
Park, R.A. 1990. AQUATOX, a Modular Toxic Effects Model for Aquatic Ecosystems.  Final
      Report, EPA-026-87; U.S. Environmental Protection Agency, Corvallis, Oregon.
Park, R.A. 1999.  Evaluation of AQUATOX for Predicting Bioaccumulation ofPCBs in the Lake
      Ontario Food Web. In: AQUATOX for Windows: A Modular Fate and Effects Model for
      Aquatic Ecosystems-Volume 3:  Model Validation  Reports.    U.S.  Environmental
      Protection Agency 2000. EPA-823-R-00-008

Park, R.A., B.B. MacLeod, C.D. Collins,  J.R. Albanese, and D. Merchant.  1985. Documentation
      of the  Aquatic Ecosystem MINI. Cleaner, A Final Report for Grant No. R806299020.
      U.S. Environmental Protection Agency, Environmental Research Laboratory,  Athens,
      Georgia. 85 pp.
Park, R.A., B.H. Indyke,  and G.W.  Heitzman.  1981.  Predicting  the Fate of Coal-Derived
      Pollutants in Aquatic Environments.   Paper presented  at Energy  and  Ecological
      Modelling symposium,  Louisville, Kentucky, April 2023,  1981.   Developments in
      Environmental Modeling 1.  7 pp.

Park,  R.A.,  C.D.   Collins,  C.I.  Connolly,  J.R.  Albanese,  and B.B.  MacLeod.  1980.
      Documentation  of the Aquatic Ecosystem Model MS.CLEANER, A Final Report for
      Grant No. R80504701, U.S. Environmental Protection Agency, Environmental Research
      Laboratory, Athens, Georgia. 112  pp.
Park, R.A., C.D. Collins, O.K. Leung,  C.W. Boylen, J.R. Albanese, P. deCaprariis, and H.
      Forstner.  1979.  The Aquatic Ecosystem Model MS.CLEANER. In State-of- the-Art in
      Ecological Modeling, edited  by  S.E. Jorgensen, 579-602.   International Society for
      Ecological Modelling, Denmark.
Park, R.A., C.I. Connolly, J.R. Albanese,  L.S. Clesceri, G.W. Heitzman, H.H. Herbrandson, B.H.
      Indyke, J.R.  Loehe,  S.  Ross, D.D.  Sharma,  and W.W. Shuster.   1980.   Modeling
      Transport and Behavior of Pesticides and Other Toxic Organic Materials in  Aquatic
      Environments. Center for Ecological Modeling Report No.  7.  Rensselaer Polytechnic
      Institute, Troy, New York.  163 pp.

                                         306

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Park, R.A., C.I. Connolly, J.R. Albanese, L.S. Clesceri, G.W. Heitzman, H.H. Herbrandson, B.H.
       Indyke, J.R. Loehe,  S. Ross, D.D. Sharma, and W.W. Shuster. 1982. Modeling the Fate
       of Toxic Organic Materials in Aquatic Environments.  U.S. Environmental Protection
       Agency Rept. EPA-6OO/S3-82-028, Athens, Georgia.
Park, R.A., D. Scavia, and N.L.  Clesceri.  1975.  CLEANER, The Lake George Model.   In
       Ecological Modeling in a Management  Context.   Resources  for  the  Future, Inc.,
       Washington, D.C.
Park, R.A., JJ. Anderson, G.L.  Swartzman, R. Morison, and J.M. Emlen.  1988. Assessment of
       Risks  of Toxic Pollutants  to Aquatic Organisms  and Ecosystems Using a Sequential
       Modeling Approach.  In  Fate  and Effects of Pollutants on Aquatic Organisms and
       Ecosystems, 153-165.  EPA/600/9-88/001.  Athens, Ga.: U.S. Environmental Protection
       Agency
Park, R.A., R.V. O'Neill, J.A.  Bloomfield, H.H.  Shugart, Jr., R.S. Booth, J.F. Koonce, M.S.
       Adams, L.S.  Clesceri, E.M. Colon, E.H. Dettman, R.A. Goldstein, J.A. Hoopes, D.D.
       Huff, S. Katz, J.F. Kitchell, R.C. Kohberger, E.J. LaRow, D.C, McNaught, J.L. Peterson,
       D. Scavia, J.E.  Titus, P.R. Weiler, J.W.  Wilkinson,  and  C.S.  Zahorcak.   1974.   A
       Generalized  Model  for  Simulating  Lake Ecosystems.   Simulation.,  23(2):30-50.
       Reprinted in Benchmark Papers in Ecology.
Park, R.A., T.W. Groden, and C.J. Desormeau.  1979.  Modifications to  the Model CLEANER
       Requiring Further Research. In Perspectives on Lake Ecosystem Modeling, edited by D.
       Scavia and A. Robertson. Ann Arbor Science Publishers, Inc., 22 pp.
Park, R.  A., E. C. Blancher, S. A. Sklenar, and  J. L.  Wood. 2002. Modeling the Effects of
       Multiple Stressors on a Use-Impaired River, in Society of Environmental Toxicology and
       Chemistry, Salt Lake City.
Park, R.  A., and J. S. Clough. 2005. Validation of AQUATOX with Nonylphenol Field Data
       (Unpublished Report). U.S. Environmental Protection Agency, Washington, DC.
Park,  R.  A.,  J.  S. Clough,  M.  C.  Wellman, and A. S. Donigian.  2005. Nutrient Criteria
       Development with  a  Linked Modeling System: Calibration of  AQUATOX Across a
       Nutrient Gradient.  Pages  885-902 in TMDL  2005. Water Environment Federation,
       Philadelphia, Penn.
Park, R. A., J. S. Clough, and M. C. Wellman. 2008. AQUATOX: Modeling Environmental Fate
       and Ecological Effects in Aquatic Ecosystems. Ecological Modelling 213: 1-15.
Parker, R.A.   1972.  Estimation of Aquatic  Ecosystem Parameters.  Verh. Internal. Verein.
       Limnol. 18:257-263.
Parsons, T.R., R.J. LeBresseur, J.D. Fulton, and O.D. Kennedy.  1969. Production Studies in the
       Strait of Georgia II. Secondary Production Under the Fraser River Plume, February to
       May, 1967. Jour. Exp. Mar. Biol. Ecol. 3:39-50.
Partheniades, E.  1965. Erosion and Deposition of Cohesive Soils.  ASCE Jour. Hydrol. Div.  pp.
       105-138.
Partheniades, E. 1971.  "Erosion and Deposition of Cohesive Materials".  In River Mechanics, H.
       W. Shen Ed. Chapter 25. Water Resources Publications, Littleton,  Colorado.

                                         307

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Pastorok, R. A., S. M. Bartell, S. Person, and L. R. Ginzburg, editors. 2002. Ecological Modeling
       in Risk Assessment. Lewis, Boca Raton, Florida.
Patten, B.C., D.A. Egloff, and T.H. Richardson.  1975.  Total Ecosystem Model for a Cove in
       Lake Texoma.  In B.C. Patten (Ed.) Systems Analysis and Simulation in Ecology. Vol. III.
       New York: Academic Press, pp. 205-241.
Plafkin, J. L., M. T.  Barbour, K. D.  Porter,  S. K.  Gross, and  R.  M. Hughes.  1989.  Rapid
       bioassessment protocols for use in streams  and rivers: Benthic macroinvertebrates and
       fish EPA  440-4-89-001,  U.S.  Environmental  Protection Agency,  Office of  Water
       Regulations and Standards, Washington, D.C.

Press, W.H., B.P.  Flannery, S.A.  Teukolsky, and W.T. Vetterling.  1986.  Numerical Recipes:
       The Art of Scientific Computing. Cambridge University Press, Cambridge, U.K. 818 pp.
Purcell, A. H., C. Friedrich, and  V.  H. Resh. 2002. An Assessment of a Small Urban Stream
       Restoration Project in Northern California. Restoration Ecology 10:685-694.

Quantitative   Environmental   Analysis.  2001.  Documentation:   Bioaccumulation  Model
       QEAFDCHN v. 1.0. Pages 21. QEA, LLC, Montvale, NJ.
Raimondo, S., Vivian, D.N.,  Barren, M.G.,  2007.  Web-based Interspecies  Correlation
       Estimation (Web-ICE) for Acute Toxicity: User Manual. Version 1.1. EPA/600/R-07/071,
       U.S. Environmental Protection Agency, Gulf Breeze, FL.
Redfield,  A.C.  1958.  The Biological  Control of  Chemical Factors in  the  Environment.
       American Scientist 46:205-222.
Riley, G.A. 1963. Theory of Food-Chain Relations in the Ocean. The Sea., 2.
Rode, M., U.  Suhr,  and G. Wriedt.  2007. Multi-objective calibration of a river water quality
       model—Information content of calibration data. Ecological Modelling 204: 129-142.
Rosemond, A.D.  1993. Seasonality and Control of Stream Periphyton: Effects of Nutrients,
       Light, and Herbivores.  Dissertation, Vanderbilt University, Nashville, Tenn., 185 pp.
Rowe, M., D.  Essig, and B. Jessup.  2003.  Guide to Selection of Sediment Targets for Use in
       Idaho TMDLs.  Pages 46. Idaho Department of Environmental Quality, Boise, Idaho.
Rykiel,  E. J., Jr. 1996. Testing ecological  models:  the meaning  of validation.  Ecological
       modelling 90:229-244.
Saltelli, A. 2001. Sensitivity Analysis  for Importance Assessment. Paper read at Sensitivity
       Analysis Methods, June  11-12, 2001, at North Carolina State University, Raleigh NC.
Sand-Jensen, K.  1977. Effects of Epiphytes on Eelgrass (Zostera marina L.) in Danish Coastal
       Waters. Marine Technology Society Journal 17:15-21.
Saunders,  G.W. 1980. 7. Organic Matter and Decomposers. In E.P. Le Cren and R.H. Lowe-
       McConnell (Eds.), The Functioning of Freshwater Ecosystems.  Cambridge: Cambridge
       University Press, pp. 341-392.

Scavia, D. 1979.  Chapter  6 The  Use of Ecological Models of Lakes in Synthesizing Available
       Information and Identifying Research Needs.   In D. Scavia and A. Robertson  (Eds.)

                                         308

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


      Perspectives on Lake Ecosystem Modeling.  Ann Arbor, Michigan: Ann Arbor Science,
      pp. 109-168.
Scavia, D. 1980. An Ecological Model of Lake Ontario. Ecological Modelling 8:49-78.
Scavia, D., and R.A. Park.  1976.  Documentation of Selected Constructs and Parameter Values
      in the Aquatic Model CLEANER. Ecological Modelling 2(l):33-58.
Scavia, D., BJ.  Eadie, and A. Robertson.  1976. An Ecological Model for Lake Ontario-Model
      Formulation, Calibration, and Preliminary Evaluation.  Tech. Report ERL 371-GLERL
       12, National Oceanic and Atmospheric Administration, Boulder, Colorado.
Schnoor, J.  E. 1996. Environmental Modeling: Fate and Transport of Pollutants in Water, Air,
      and Soil. John Wiley & Sons, Inc., New York.
Schol, A., V. Kirchesch, T. Bergfeld, and D. Miiller. 1999. Model-based analysis of oxygen
      budget and biological processes in the  regulated rivers Moselle and Saar: modelling the
      influence of benthic filter feeders on phytoplankton. Hydrobiologia 410:  167-176.
Schol, A., V. Kirchesch, T. Bergfeld, F. Scholl, J. Borcherding, and D. Muller.  2002. Modelling
      the chlorophyll  content of the River Rhine - interaction between riverine algal production
      and  population biomass of grazers, rotifers  and zebra mussel, Dreissena polymorpha.
      IntyernationalReview of Hydrobiology  87: 295-317.
Schwarzenbach, R., P. M.  Gschwend,  and  D.  M. Imboden.  1993. Environmental Organic
      Chemistry. John Wiley & Sons, New York.
Sedell, J.R., F.J. Triska, and N.S. Triska.  1975.  The Processing of Conifer and Hardwood
      Leaves in Two  Coniferous Forest Streams: I. Weight Loss and Associated Invertebrates.
      Herh. Internal.  Verein. LimnoL, 19:1617-1627.
Sijm, D.T.H.M., K.W.  Broersen, D.F de Roode, and  P. Mayer.  1998.  Bioconcentration Kinetics
      of Hydrophobic  Chemicals  in Different  Densities of  Chlorella  Opyrenoidosa.
      Environmental  Toxicology and Chemistry 17:9:1695-1704.
Skoglund, R.S.,  K. Stange, and D.L. Swackhamer.  1996.  A Kinetics Model for Predicting the
      Accumulation  of PCBs in Phytoplankton.   Environmental  Science and Technology
      30:7:2113-2120.
Small, M. J., and M. C. Sutton. 1986. A Regional pH-Alkalinity Relationship. Water Research
      20: 335-343.
Smayda, T.J.  1971.  Some Measurements of the Sinking Rate of Fecal Pellets.  Limnology and
      Oceanography  14:621-625.
Smayda, T.J.  1974.   Some Experiments on the Sinking  Characteristics  of  Two Freshwater
      Diatoms. Limnology and Oceanography 19:628-635.
Smejtek, P., and S. Wang.  1993.  Distribution of Hydrophobic lonizable Xenobiotics Between
      Water and Lipid Membranes: Pentachlorophenol and Pentachlorophenate.  A Comparison
      with Octanol-Water Partition. Archives  of Environmental Contamination and Toxicology,
      25(3):394.
                                         309

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Smith, DJ. 1978.  WQRRS,  Generalized Computer Program for River-Reservoir Systems. U.S.
      Army  Corps of Engineers, Hydrologic Engineering Center (HEC), Davis, California
      Users Manual 401-100, 100A, 210 pp.
Southworth, G.R., JJ.  Beauchamp, and P.K. Schmieder.  1978.  Bioaccumulation Potential of
      Polycyclic Aromatic Hydrocarbons in Daphniapulex.  Water Res.., 12:973-977.
Spacie, A., and J.L. Hamelink. 1982.  Alternative Models for Describing the Bioconcentration of
      Organics in Fish. Environmental Toxicology and Chemistry, 1:309-320.
Stange,  K., and  D.L.  Swackhamer.  1994.  Factors Affecting Phytoplankton  Species-Specific
      Differences  in  Accumulation  of 40  Polychlorinated Biphenyls (PCBs). Environmental
      Toxicology and Chemistry, 13(11): 1849-1860.
Steele, J.H.  1962.   Environmental Control  of Photosynthesis in the Sea.  Limnol. Oceanogr.,
      7:137-150.
Steele, J.H. 1974.  The Structure of Marine  Ecosystems. Harvard University Press, Cambridge,
      Massachusetts,  128 pp.
Steele, J.H., and M.M.  Mullin. 1977. Zooplankton Dynamics.  In E.D. Goldberg, IN. McCave,
      JJ. O=Brien, and J.H. Steele (Eds.),  The Sea Vol.  6: Marine Modeling, New York:
      Wiley-Interscience, p. 857.
Stefan,  H.G., and  X.  Fang. 1994.  Dissolved Oxygen Model  for Regional Lake Analysis.
      Ecological Modelling 71:37-68.
Sterner,  R. W., and J.  J. Elser. 2002. Ecological Stoichiometry: The Biology of Elements from
      Molecules to the Biosphere. Princeton University Press, Princeton NJ.
Sterner,  R.W., and N.  B. George. 2000.  Carbon, Nitrogen,  and Phosphorus Stoichiometry of
      Cyprinid Fishes. Ecology 81: 127-140.
Stewart, D.C.  1975.  Mathematical Modelling  of the Ecosystem of Lough Neagh.   Ph.D.
      Dissertation, Queen's University, Belfast, Northern Ireland.
Straskraba, M. 1973.  Limnological Basis for Modeling Reservoir Ecosystems.  In Ackermann,
      W.C.,  G.F. White, and E.B. Worhtington  (eds.) Man-Made Lakes: Their Problems and
      Environmental Effects.  Geophys. Momogr. Series Vol. 17, London, pp. 517-538.
Straskraba, M. and A.H. Gnauck.  1985. Freshwater Ecosystems: Modelling and Simulation.
      Developments in Environmental Modelling, 8.  Elsevier Science Publishers, Amsterdam,
      The Netherlands.  309pp.
Stream Bryophyte Group. 1999. Role of bryophytes in stream ecosystems. Journal of the North
      American Benthological Society, 18, 151-184.
Stumm,  W., and J.  J.  Morgan. 1996. Aquatic Chemistry: Chemical Equilibria and Rates in
      Natural Waters. John Wiley & Sons, New York.
Suarez,  L.A.,  and M.C. Barber.  1992.  PIRANHA Version 2.0, FGETS Version 3.0-11 User's
      Manual,  In  PIRANHA Pesticide and  Industrial Chemical Risk Analysis and  Hazard
      Assessment.  Athens,  Georgia: U.S. Environmental Protection Agency.
Suren, A., M., and  J. G.  Jowett. 2001. Effects of Deposited Sediment on Invertebrate Drift: an
      Experimental Study. New Zealand Journal of Marine and Freshwater Research 35: 725-
                                         310

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION _ CHAPTER 11


       737.Suren, A., M. 2005. Effects of Deposited Sediment on Patch Selection by Two
       Grazing Stream Invertebrates. Hydrobiologia 549: 205-218.
Suter, G.W., II, A.E. Rosen, and E. Linder.  1986. 4. Analysis of Extrapolation Error.  User's
      Manual for Ecological Risk Assessment.  Oak Ridge National Laboratory, ORNL-6251,
       pp. 49-81.
Swackhamer, D.L., and R.S. Skoglund. 1991. The Role of Phytoplankton in the Partitioning of
       Hydrophobic Organic Contaminants in Water.  In Baker, R.A., ed., Organic Substances
       and Sediments in  Water  Vol.  2 C Processes and Analytical, Lewis: Chelsea MI, pp.  91-
       105.
Swackhamer, D.L., and R.S. Skoglund. 1993.  Bioaccumulation of PCBs by Algae: Kinetics
       versus Equilibrium. Environmental Toxicology & Chemistry, 12:831-838.
Tetra Tech Inc. 2002. Draft User's  Manual  for Environmental Fluid Dynamics Code Hydro
       Version (EFDC-Hydro). U.S.  Environmental Protection Agency, Atlanta, GA.
Thomann, R. V.,  and J. Mueller. 1987.  Principles  of Surface Water Quality Modeling and
       Control. HarperCollins, New York, NY.
Thomann, R.V.  1989. Bioaccumulation Model of Organic Chemical Distribution in  Aquatic
       Food Chains. Environmental Science & Technology, 23:699-707.
Thomann, R.V., and JJ. Fitzpatrick.  1982.  Calibration and Verification of a Mathematical
      Model of the  Eutrophication of the Potomac Estuary.   Prepared for Department of
       Environmental Services, Government of the District of Columbia, Washington, D.C.
Thomann, R.V.,  D.M. Di  Toro, R.P.  Winfield, and D.J.  O'Connor.    1975.  Mathematical
      Modeling of  Phytoplankton  in Lake  Ontario,  Part  1. Model  Development  and
       Verification. Manhattan College, Bronx, New York, for U.S.  Environmental Protection
       Agency EPA-600/3-75-005.
Thomann, R.V., J. Segna, and R. Winfield.   1979.  Verification Analysis of Lake Ontario and
      Rochester Embayment Three-Dimensional Eutrophication Models.  Manhattan College,
       Bronx, New York, for U.S. Environmental Protection Agency.
Thomann, R.V., J.A. Mueller, R.P.  Winfield, and C.-R. Huang.  1991.  Model  of Fate  and
       Accumulation of  PCB Homologues  in Hudson Estuary.  Jour. Environ. Engineering,
Thompson, J. B., S. Schultze-Lam, T. J. Beveridge, and D. J. Des Marais. 1997. Whiting Events:
      Biogenic  Origin Due to the Photosynthetic Activity of Cyanob acted al Picoplankton.
      Limnology and Oceanography 42: 133-141.
Titus, I.E., M.S. Adams, P.R. Weiler, R.V. O'Neill, H.H. Shugart, Jr., and J.B.  Mankin.  1972.
      Production Model for Myriophyllum spicatum L.  Memo Rept. 72-19, U.S. International
      Biological Program  Eastern  Deciduous  Forest  Biome,   University  of  Wisconsin,
      Madison,  17 pp.
Toetz, D.W.  1967.   The  Importance  of Gamete Losses in Measurements of Freshwater Fish
      Production. Ecology. 48:1017-1020.
Traas, T. P.,  J.  A. Stab,  P. R. G. Kramer, W. P. Cofino, and T. Aldenberg. 1996. Modeling and

                                        311

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


       Risk Assessment of Tributyltin Accumulation in the Food Web of a Shallow Freshwater
       Lake Environ. Sci.  Technol. 30: 1227-1237.
Traas, T. P., J. H. Janse, T. Aldenberg, and J. T. Brock. 1998. A Food Web Model for Fate and
       Direct and Indirect Effects of Dursban 4E (Active Ingredient Chlorpyrifos) in Freshwater
       Microcosms. Aquatic Ecology 32: 179-190.
Traas, T. P., J. H. Janse, P. J. Van den Brink, and T. Aldenberg. 2001. A Food Web Model for
       Fate and Effects of Toxicants and Nutrients in Aquatic Mesocosms.  Model Description.
       RIVM, Bilthoven, The Netherlands.
U.S. Department of Commerce,  1994, Manual of Harmonic Analysis and Prediction of Tides.
       Special Publication No.  98, Revised (1940) Edition (reprinted 1958 with corrections;
       reprinted again 1994). United States Government Printing Office, 1994.
U.S. Environmental Protection Agency,  1977. Various reports on Lake Chemung and Lake
       Allegan, MI; White Bear Lake, MN; Saratoga Lake, NY; Sebasticook Lake, ME; and
       Bantam Lake, Aspinook Pond, and Hanover Pond, CT. National Eutrophication Survey
       Working Napers. U.S. Environmental Protection Agency, Washington, D.C.
U.S. Environmental Protection Agency, 1986.  Quality Criteria for Water, 1986.  Environmental
       Protection Agency, Washington,  D.C.,  Office of  Water Regulations  and Standards,
       EPA/440/5-86/001, 398 pp.
U.S. Environmental Protection Agency,  1988.   The Effects of Chloropyrifos on a Natural
       Aquatic System: A Research Design for Littoral Enclosure Studies and Final Research
       Report.  U.S. Environmental Protection Agency, Environmental Research Laboratory,
       Duluth, Minnessota, 194 pp.
U.S. Environmental Protection Agency. 1989. Green Bay/Fox River Mass Balance Study U.S.
       Environmental Protection Agency Great Lakes National Program Office, Chicago, IL.
U.S. Environmental Protection Agency, 1991. Hydrological Simulation Program - FORTRAN -
       User's Manual for Release 10 (Pre-release Draft  Version).  U.S. EPA Technology
       Development  and  Applications Branch  in  cooperation with USGS  Water Resources
       Division,  Office of Surface Water.  By Bicknell, B.R., J.C. Imhoff,  J.L. Kittle, A.S.
       Donigian, and -R.C. Johanson.
U.S. Environmental Protection Agency, 1995.  Great Lakes Water Quality Initiative Technical
       Support Document for the Procedure to Determine Bioaccumulation Factors. EPA-820-
       B-95-005, U.S. Environmental Protection Agency, Washington, D.C.
U.S. Environmental Protection Agency.  1997. Guiding Principles for Monte  Carlo Analysis.
       Risk Assessment Forum. Washington, DC: U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency, 1999,  Update of Ambient Water Quality Criteria for
       Ammonia, September 1999, U.S. EPA Office of Water,  U.S. EPA Office of Science and
       Technology Washington, D.C.
U.S. Environmental Protection Agency, 2000a.  Ambient Aquatic Life Water Quality Criteria for
       Dissolved Oxygen (Saltwater)
U.S. Environmental Protection Agency, 2000b. Progress in  Water Quality: An Evaluation of the
       National Investment in Municipal  Wastewater Treatment: Appendix B.  EPA-832-R-00-

                                         312

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


       008, http://www.epa.gov/owtnitnet/wquality/app-b.pdf.
U.S. Environmental Protection Agency, 2002. National Water Quality Inventory: 2000 Report.
       EPA-841-R-02-001.
U.S. Environmental Protection Agency, 2005. AQUATOX For Windows: A Modular Fate and
       Effects Model For Aquatic Ecosystems Release 2.1 Addendum To Release 2 Technical
       Documentation.
USGS, 2010.  CO2calc: A User-Friendly Seawater Carbon Calculator for Windows, Mac OS X,
       andiOS (iPhone), by L.L. Robbins, M.E. Hansen, J.A. Kleypas, and S.C. Meylan. Open-
       File Report 2010-1280.
Van Rees, K. C. I, K. R. Reddy, and P.  A. Moore, Jr. 1991. Lake Okeechobee Phosphorus
       Dynamics Study: Biogeochemical Processes in the Sediments,  Chapter  7: Phosphorus
       Exchange Between  Sediment and Overlying Water.  Pages 7-1 to 7-26. University
       Florida, Soil Science Department, Gainesville, Fl.
Velleux, M., S. Westenbroek, J.  Ruppel, M.  Settles, and D. Endicott. 2000. A User's Guide to
       IPX, the In-Place Pollutant Export  Water Quality Modeling Framework,  Version 2.7.4.
       U.S. Environmental Protection Agency, Office of Research and Development, National
       Health  and Environmental  Effects  Research  Laboratory,  Mid-Continental  Ecology
       Division-Duluth, Large Lakes Research Station, Grosse He, Michigan. 179 pp.
Verduin,  1982.  Components Contributing to Light Extinction in  Natural Waters: Method of
       Isolation. Arch. Hydrobiol, 93(3):303-312.
Verscheuren, K. 1983. Handbook of Environmental Data on Organic Chemicals, Second edition.
       Van Nostrand Reinhold, New York.
Ward 1963, ASCE 1989, 6:1-16
Watt, W.D.  1966.  Release of Dissolved Organic Material From the  Cells of  Phytoplankton
       Species in Natural and Mixed Populations. Proceedings of the Royal Society, London, B
       164:521-525.
Weininger, D.   1978.   Accumulation of PCBs by Lake Trout  in Lake Michigan.   Ph.D.
       Dissertation, University of Wisconsin, Madison, 232 pp.
Westlake,  D.F.  1967. Some Effects of Low Velocity Currents on the Metabolism of  Aquatic
       Macrophytes.  Journal Experimental Botany 18:187-205.
Wetzel, R.G., P.H. Rich, M.C. Miller, and H.L.  Allen.  1972. Metabolism of  Dissolved and
       Particulate Detrital Carbon in a Temperate Hard-water Lake,  in U. Melchiorri-Santolinii
       and J.W. Hopton (eds.) Detritus and Its Role  in Aquatic Ecosystems, Mem. 1st. Ital.
       Idobiol, 29(Suppl): 185-243.
Wetzel, R.G.  1975. Limnology, W.B. Saunders, Philadelphia, 743 pp.
Wetzel, R.G.  2001. Limnology: Lake and River Ecosystems. San Diego: Academic Press, 1006
       pp.
Whitman, W.G. 1923. The two-film theory of gas absorption. Chem. Metal. Eng.  29:146-148.
Winberg,  G.  G. 1971.  Symbols, Units and Conversion  Factors  in  Studies  of Freshwater
       Productivity. Pages 23. International Biological Programme Central Office, London.
                                         313

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION	CHAPTER 11


Wlosinski, J. H., and C. D. Collins. 1985. Confirmation of the Water Quality Model CE-QUAL-
      Rl Using Data from Eau Galle Reservoir, Wisconsin.  Army Engineers Waterways
      Experiment Station, Vicksburg, Mississippi.
Wood, L.W.,  P.  O.Keefe, and B.  Bush.  1997.  Similarity Analysis  of PAH  and PCB
      Bioaccumulation  Patterns   in   Sediment-Exposed  Chironomus  tentans   Larvae.
      Environmental Toxicology and Chemistry, 16(2):283-292.
Wool, T. A., R. B. Ambrose, J. L. Martin, and E. A. Comer. 2004. Water Quality Analysis
      Simulation Program (WASP) Version 6.0 DRAFT:  User's Manual. US Environmental
      Protection Agency - Region 4, Atlanta GA.
Yuan, 2006.  The Development of the Web Based CO 2 SYS Program. Masters Thesis. University
      of Montana, Autumn 2006.
                                        314

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    APPENDICES
                       APPENDIX A.  GLOSSARY OF TERMS

Taken in large part from: The Institute of Ecology.  1974. An Ecological Glossary for Engineers
and Resource Managers. TIE Publication #3, 50 pp.
Abiotic
Adsorption

Aerobic
Algae

Allochthonous

Algal bloom
Alluvial
Alluvium
Ambient
Anaerobic
Anoxic
Aphotic
Assimilation
Autochthonous
Benthic

Benthos
Biodegradable

Biochemical oxygen
 demand (BOD)

Biomagnification

Biomass

Biota
Chlorophyll
Colloid

Consumer
Copepods

Crustacean
Decomposers
Detritus
Diatom
Diurnal
nonliving, pertaining to physico-chemical factors only
the adherence of substances to the surfaces of bodies with which they
are in contact
living, acting, or occurring in the presence of oxygen
any of a group of chlorophyll-bearing aquatic plants with no true leaves,
stems, or roots
material  derived   from  outside  a  habitat  or environment  under
consideration
rapid and flourishing growth of algae
of alluvium
sediments deposited by running water
surrounding on all sides
capable of living or acting in the absence of oxygen
pertaining to conditions of oxygen deficiency
below the level of light penetration in water
transformation of absorbed nutrients into living matter
material derived from within a habitat, such as through plant growth
pertaining to the bottom of a  water body; pertaining to  organisms that
live on the bottom
those  organisms that live on the bottom of a body of water
can be broken down  into simple inorganic substances by the action of
decomposers (bacteria and fungi)

the amount of oxygen required to decompose a given amount of organic
matter
the step by step concentration of chemicals in successive levels of a food
chain  or food web
the total  weight  of  matter  incorporated into  (living  and/or  dead)
organisms
the fauna and flora of a habitat or region
the green, photosynthetic pigments of plants
a dispersion of particles  larger than small molecules and that do not
settle  out of suspension
an organism that consumes another
a  large  subclass  of usually  minute,  mostly free-swimming aquatic
crustaceans
a large class of arthropods that bear a horny shell
bacteria and fungi that break down organic detritus
dead organic matter
any of class of minute algae with cases of silica
pertaining to daily occurrence
                                         315

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    APPENDICES
Dynamic equilibrium
Ecology

Ecosystem

Emergent

Environment
Epilimnion
Epiphytes
Equilibrium
Euphotic

Eutrophic
Fauna
Flood plain
Flora
Fluvial
Food chain

Food web
Forage fish
Habitat
Humic
Hydrodynamics
Hypolimnion
Influent
Inorganic

Invertebrate
Limiting factor

Limnetic zone

Limnology
Littoral zone

Macrofauna
Macrophytes
Nutrients
Omnivorous
Organic chemical
Overturn

Oxygen depletion
Parameter
a state of relative balance between processes having opposite effects
the study  of the interrelationships of organisms with and within their
environment
a  biotic  community and its  (living  and  nonliving)  environment
considered together
aquatic plants, usually rooted, which have portions above water for part
of their life cycle
the sum total of all the external conditions that act on an organism
the well mixed surficial layer of a lake; above the hypolimnion
plants that grow on other plants, but are not parasitic
a steady state in a dynamic system, with outflow balancing inflow
pertaining  to  the upper  layers of water in which sufficient light
penetrates to permit growth of plants
aquatic systems with high nutrient input and high plant growth
the animals of a habitat or region
that part of a river valley that is covered in periods of high (flood) water
plants of a habitat or region
pertaining to a stream
animals linked by  linear predator-prey  relationships with  plants or
detritus at the base
similar to food chain, but implies cross connections
fish eaten by other fish
the environment in which a population of plants or animals occurs
pertaining to the partial decomposition of leaves and other plant material
the study of the movement of water
the lower layer of a stratified water body, below the well mixed zone
anything flowing into a water body
pertaining to matter that is neither living nor immediately derived from
living matter
animals lacking a backbone
an environmental factor that limits the growth of an organism; the factor
that is closest to the physiological limits of tolerance of that organism
the open water zone of a lake or pond from the surface  to the depth of
effective light penetration
the study of inland waters
the shoreward zone of a water body in which the light penetrates to the
bottom, thus usually supporting rooted aquatic plants
animals visible to the naked eye
large (non-microscopic), usually rooted, aquatic plants
chemical elements essential to life
feeding on a variety of organisms and organic detritus
compounds containing carbon;
the complete  circulation or mixing of the upper and lower waters of a
lake when temperatures (and densities) are similar
exhaustion of oxygen by chemical or biological use
a measurable, variable quantity as distinct from a statistic
                                          316

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
                                                    APPENDICES
Pelagic zone
Periphyton

Oxidation
Photic zone

Phytoplankton
Plankton
Pond
Population
Predator
Prey
Producer

Production
Productivity
Productivity, primary
Productivity, secondary
Reservoir
Riverine
Rough fish
Sediment
Siltation

Stratification

Substrate

Succession
Tolerance
Trophic level
Turbidity

Volatilization

Wastewater
Wetlands

Zooplankton
open water with no association with the bottom
community of algae and associated organisms, usually small but densely
set, closely attached to surfaces on or projecting above the bottom
a reaction between molecules, ordinarily involves gain of oxygen
the region of aquatic environments  in  which the intensity of light is
sufficient for photosynthesis
small, mostly microscopic algae floating in the water column
small organisms floating in the water
a small, shallow lake
a group of organisms  of the same species
an organism, usually an animal, that kills and consumes other organisms
an organism killed and at least partially consumed by a predator
an organism that can synthesize organic matter using inorganic materials
and an external energy source (light or chemical)
the amount of organic material produced by biological activity
the rate of production of organic matter
the rate of production by plants
the rate of production by consumers
an artificially impounded body of water
pertaining to rivers
a non-sport fish, usually omnivorous in food habits
any mineral and/or organic matter deposited by water or air
the deposition of silt-sized  and clay-sized (smaller than sand-sized)
particles
division  of  a water  body  into two  or  more  depth  zones due to
temperature or density
the layer on which organisms grow; the organic substance attacked by
decomposers
the replacement of one plant assemblage with another through time
an organism's capacity to endure or adapt to unfavorable conditions
all organisms that  secure their food at a common step in the food chain
condition of water resulting from suspended matter, including  inorganic
and organic material and plankton
the act of passing into a  gaseous  state at ordinary temperatures  and
pressures
water derived from a municipal or industrial waste treatment plant
land saturated or nearly  saturated with water  for most  of  the year;
usually vegetated
small  aquatic  animals,  floating,  usually with  limited swimming
capability
                                          317

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
APPENDIX B. USER-SUPPLIED PARAMETERS AND DATA

       The model has many parameters and internal variables.  Most of these are linked to data structures such as ChemicalRecord, SiteRecord,
and ReminRecord, which in turn may be linked to input forms that the user accesses through the Windows environment. Although consistency
has been a goal, some names may differ between the code, the user interface, and the technical documentation
USER INTERFACE

Chemical
CAS Registry No.
Molecular Weight
Dissociation Constant
Solubility
Henry's Law Constant
Vapor Pressure
Octanol- water partition
coefficient
KPSED
KOMRe&DOM
Uptake Rate (Kl ) Detritus
Cohesives Kl
Cohesives K2
Cohesives Kp
Non-Cohesives Kl
Non-Cohesives K2
Non-Cohesives Kp
INTERNAL
ChemicalRecord
ChemName
CASRegNo
MolWt
pka
Solubility
Henry
VPress
LogKow
KPSed
KOMRefrDOM
KIDetritus
CohesivesKl
CohesivesK2
CohesivesKp
NonCohKl
NonCohK2
NonCohKp
TECH DOC
Chemical
Underlying Data
N/A
N/A
MolWt
pKa
N/A
Henry
N/A
LogKow
KPSed
KOMRe&DOM
Kloetr
Kl
K2
Kp
Kl
K2
Kp
DESCRIPTION
For each chemical simulated, the following
Parameters are required
Chemical's Name. Used for Reference only.
CAS Registry Number. Used for Reference only.
Molecular weight of pollutant
Acid dissociation constant
Not utilized as a parameter by the code.
Henry's law constant
Not utilized as a parameter by the code.
Log octanol- water partition coefficient
Detritus-water partition coefficient
Reftractory DOM to Water Partition Coefficient
Uptake rate constant for organic matter, default of 1 .39
Uptake rate constant for cohesives
Depuration rate constant for cohesives
Partition coefficient for cohesives
Uptake rate constant for non-cohesives class 1
Depuration rate constant for non-cohesives class 1
Partition coefficient for non-cohesives class 1
UNITS

N/A
N/A
g/mol
negative log
ppm
atm m3 mol-1
mmHg
unitless
L/kgOC
L/kgOM
L/kg dry day
L/kg dry day
day1
L/kg dry
L/kg dry day
day1
L/kg dry
                                                          318

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Non-Cohesives2 Kl
Non-Cohesives2 K2
Non-Cohesives2 Kp
Activation Energy for
Temperature
Rate of Anaerobic Microbial
Degradation
Max. Rate of Aerobic
Microbial Degradation
Uncatalyzed hydrolysis
constant
Acid catalyzed hydrolysis
constant
Base catalyzed hydrolysis
constant
Photolysis Rate
Oxidation Rate Constant
Weibull Shape Parameter
Weibull Slope Factor
Chemical is a Base
This Chemical is a PFA
Type of PFA
Perfluoralkyl Chain Length
Kom for Sediments (PFA)
INTERNAL
NonCoh2Kl
NonCoh2K2
NonCoh2Kp
En
KMDegrAnaerobic
KMDegrdn
KUnCat
KAcid
KBase
PhotolysisRate
OxRateConst
Weibull_Shape
WeibullSlopeF actor
ChemlsBase
IsPFA
PFAType
PFAChainLength
PFASedKom
TECH DOC
Kl
K2
Kp
En
KAnaerobic
KMDegrdn
KUncat
KAcid
KBase
KPhot
N/A
Shape (Internal Model)
Slope Factor (External
Model)
Compound is a base
Compound is a PFA
carboxylate / sulfonate
ChainLength
Kom for Sediments
DESCRIPTION
Uptake rate constant for non-cohesives class 2
Depuration rate constant for non-cohesives class 2
Partition coefficient for non-cohesives class 2
Arrhenius activation energy
Decomposition rate at 0 g/m3 oxygen
Maximum (microbial) degradation rate
The measured first-order reaction rate at ph 7
Pseudo-first-order acid-catalyzed rate constant for a given ph
Pseudo-first-order rate constant for a given ph
Direct photolysis first-order rate constant
Not utilized as a parameter by the code.
Parameter expressing variability in toxic response; default is 0.33
Slope at LC50 multiplied by LC50
True if the compound is a base
True if the compound is a perfluorinated surfactant
Sulfonate group and carboxylate group
Length of perfluoroalkyl chain
Organic matter partition coefficient for the PFA
UNITS
L/kg dry day
day1
L/kg dry
cal/mol
1/d
1/d
1/d
L/mol • d
L/mol • d
1/d
L/mold
unitless
Unitless
True/False
True/False
carboxylate /
sulfonate
Integer
L/kg
                                            319

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
BCF for Algae (PFA)
BCF for Macrophytes (PFA)
INTERNAL
PFAAlgBCF
PFAMacroBCF
TECH DOC
BCF for Algae
BCF for Macrophytes
Description
Bioconcentration Factor for the PFA to algae
Bioconcentration Factor for the PFA to macrophytes
UNITS
L/kg
L/kg

Site Name
Max Length (or reach)
Vol.
Surface Area
Estuary Site Width
Mean Depth
Maximum Depth
Ave. Temp, (epilimnetic or
hypolimnetic)
Epilimnetic Temp. Range (or
hypolimnetic)
Latitude
Altitude (affects oxygen sat.)
Average Light
Annual Light Range
Total Alkalinity
Hardness as CaCOS
Sulfate Ion Cone
Total Dissolved Solids
SiteRecord
SiteName
SiteLength
Volume
Area
SiteWidth
ZMean
ZMax
TempMean
TempRange
Latitude
Altitude
LightMean
LightRange
AlkCaCOS
HardCaCO3
SO4Conc
TotalDissSolids
Site Underlying
Data
N/A
Length
Volume
Area
Width
ZMean
ZMax
TempMean
TempRange
Latitude
Altitude
LightMean
LightRange
N/A
N/A
N/A
N/A
For each water body simulated, the following
Parameters are required
Site's Name. Used for Reference only.
Maximum effective length for wave setup
Initial volume of site (must be copied into state var.)
Site area
Width of estuary
Mean depth, (initial condition if dynamic mean depth is selected)
Maximum depth
Mean annual temperature of epilimnion (or hypolimnion)
Annual temperature range of epilimnion (or hypolimnion)
Latitude
Site specific altitude
Mean annual light intensity
Annual range in light intensity
Not utilized as a parameter by the code.
Not utilized as a parameter by the code.
Not utilized as a parameter by the code.
Not utilized as a parameter by the code.

N/A
km
m3
m2
m
M
M
°C
°C
Deg, decimal
m
Langleys/day
Langleys/day
mg/L
mg CaCOS /L
mg/L
mg/L
                                            320

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Enclosure Wall Area
Mean Evaporation
Extinct. Coeff Water
Extinct. Coeff Sediment
Extinct. Coeff DOM
Extinct. Coeff POM
Baseline Percent
Embeddedness
Minimum Volume Frac.
Auto Select Eqn. for
reaeration
Enter KReaer
Total Length
Watershed Area
M2, Amplitude & Epoch
S2, Amplitude & Epoch
N2, Amplitude & Epoch
Kl, Amplitude & Epoch
Ol, Amplitude & Epoch
SSA, Amplitude & Epoch
SA, Amplitude & Epoch
PI, Amplitude & Epoch
INTERNAL
EnclWallArea
MeanEvap
ECoeffWater
ECoeffSed
ECoeffDOM
ECoeffPOM
BasePercentEmbed
Min Vol Frac
UseCovar
KReaer
TotalLength
WaterShedArea
amplitudel,kl
amplitude2, k2
amplitudes, k3
amplitude4, k4
amplitudes, k5
amplitude6, k6
amplitude?, k7
amplitudes, k8
TECH DOC
EnclWallArea
MeanEvap
ExtinctmO
ECoeffSed
ECoeffDOM
ECoeffPOM
baseline embeddedness
Minimum Volume Frac.
Covar
KReaer
TotLength
Watershed
M2
S2
N2
Kl
Ol
SSA
SA
PI
DESCRIPTION
Area of experimental enclosures walls; only relevant to enclosure
Mean annual evaporation
Light extinction of wavelength 312.5 nm in pure water
Light extinction due to inorganic sediment in water
Light extinction due to dissolved organic matter in water
Light extinction due to particulate organic matter in water
Observed embeddedness that is used as an initial condition
Fraction of initial condition that is the minimum volume of a site
Boolean to determine whether user is entering reaeration
coefficient
Depth-averaged reaeration coefficient
Total river length for calculating Nhytoplankton retention
Watershed area for estimating total river length (above)
Estuary Only - principal lunar semidiurnal constituent
Estuary Only - principal solar semidiurnal constituent
Estuary Only - larger lunar elliptic semidiurnal constituent
Estuary Only - lunar diurnal constituent
Estuary Only - lunar diurnal constituent
Estuary Only - solar semiannual constituent
Estuary Only - solar annual constituent
Estuary Only - solar diurnal constituent
UNITS
m2
inches / year
1/m
l/(m-g/m )
l/(m-g/m3)
l/(m-g/m3)
percent (0-100)
frac. of Initial
Condition
boolean
1/d
km
km2
m, deg. Local
Siderial Time
(LST)
m, deg. LST
m, deg. LST
m, deg. LST
m, deg. LST
m, deg. LST
m, deg. LST
m, deg. LST
                                            321

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Channel Slope
Maximum Channel Depth
Before Flooding
Sediment Depth
Stream Type
use the below value
Mannings Coefficient
Percent Riffle
Percent Pool

Silt: Critical Shear Stress for
Scour
Silt: Critical Shear Stress for
Deposition
Silt: Fall Velocity
Clay: Critical Shear Stress
for Scour

INTERNAL
SiteRecord (Stream-
Specific)
Channel_Slope
Max_Chan_Depth
SedDepth
StreamType
UseEnteredManning
EnteredManning
PctRiffle
PctPool
SiteRecord (Sand-Silt-
Clay Specific)
ts_silt
tdep_silt
FallVel_silt
ts clay

TECH DOC
Site Underlying
Data
Slope
Max_Chan_Depth
SedDepth
Stream Type

Manning
Riffle
Pool
Site Underlying
Data
TauScourSed
TauDepSed
VTSed
TauScourSed

DESCRIPTION
For each stream simulated, the following
Parameters are required
Slope of channel
Depth at which flooding occurs
Maximum sediment depth
Concrete channel, dredged channel, natural channel
Do not determine Manning coefficient from streamtype
Manually entered Manning coefficient.
Percent riffle in stream reach
Percent pool in stream reach
For each stream with the inorganic sediments
model included, the following Parameters are
required
Critical shear stress for scour of silt
Critical shear stress for deposition of silt
Terminal fall velocity of silt
Critical shear stress for scour of clay

UNITS

m/m
m
m
Choice from List
true/false
s/m10
%
%

kg/m2
kg/m2
m/s
kg/m2

                                            322

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Clay: Critical Shear Stress
for Deposition
Clay: Fall Velocity
INTERNAL
tdep clay
FallVel_clay
TECH DOC
TauDepSed
VTSed
DESCRIPTION
Critical shear stress for deposition of clay
Terminal fall velocity of clay
UNITS
kg/m2
m/s

Max. Degrdn Rate, labile
Max Degrdn Rate, Refrac
Temp. Response Slope
Optimum Temperature
Maximum Temperature
Min. Adaptation Temp
Min pH for Degradation
Max pH for Degradation
KNitri, Max Rate of Nitrif
KDenitri Bottom (max.)
KDenitri Water (max.)
P to Organics, Labile
N to Organics, Labile
P to Organics, Refractory
N to Organics, Refractory
P to Organics, Diss. Labile
N to Organics, Diss. Labile
P to Organics, Diss. Refr.
ReminRecord
DecayMax Lab
DecayMax Refr
Q10
TOpt
TMax
TRef
pHMin
pHMax
KNitri
KDenitri_Bot
KDenitri_Wat
P20rgLab
N20rgLab
P2OrgRefr
N2OrgRefr
P2OrgDissLab
N2OrgDissLab
P2OrgDissRefr
Remineralization
Data
DecayMax
ColonizeMax
Q10
TOpt
TMax
TRef
pHMin
pHMax
KNitri
KDenitriBottom
KDenitriWater
P20rgLab
N20rgLab
P2OrgRefr
N2OrgRefr
P2OrgDissLab
N2OrgDissLab
P2OrgDissRefr
For each simulation, the following Parameters are
required (pertaining to organic matter)
Maximum decomposition rate
Maximum colonization rate under ideal conditions
Not utilized as a parameter by the code.
Optimum temperature for degredation to occur
Maximum temperature at which degradation will occur
Not utilized as a parameter by the code.
Minimum ph below which limitation on biodegradation rate
occurs.
Maximum ph above which limitation on biodegradation occurs.
Maximum rate of nitrification
Maximum rate of denitrification at the sed/water interface
Maximum rate of denitrification in the water column
Ratio of phosphate to labile organic matter
Ratio of nitrate to labile organic matter
Ratio of phosphate to refractory organic matter
Ratio of nitrate to refractory organic matter
Ratio of phosphate to dissolved labile organic matter
Ratio of nitrate to dissolved labile organic matter
Ratio of phosphate to dissolved refractory organic matter

g/g'd
g/g'd

°C
°C
°C
PH
PH
I/day
I/day
I/day
fraction dry weight
fraction dry weight
fraction dry weight
fraction dry weight
fraction dry weight
fraction dry weight
fraction dry weight
                                            323

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
N to Organics, Diss. Refr.
O2 : Biomass, Respiration
CBODu to BODS
conversion factor
O2 : N, Nitrification
Detrital Sed Rate (KSed)
Temperature of Obs. KSed
Salinity of Obs. KSed
PO4, Anaerobic Sed.
NH4, Aerobic Sed
Wet to Dry Susp. Labile
Wet to Dry Susp. Refr
Wet to Dry Sed. Labile
Wet to Dry Sed. Refr.
KD, P to CaC03

Animal
Animal Type
Taxonomic Type or Guild
Toxicity Record
Half Saturation Feeding
Maximum Consumption
Min Prey for Feeding
Sorting: selective feeding
INTERNAL
N2OrgDissRefr
O2Biomass
BOD5_CBODu
02N
KSed
KSedTemp
KSedSalinity
PSedRelease
NSedRelease
Wet2DrySLab
Wet2DrySRefr
Wet2DryPLab
Wet2DryPRefr
KDPCalcite
ZooRecord
AnimalName
Animal_Type
Guild Taxa
ToxicityRecord
FHalfSat
CMax
BMin
Sorting
TECH DOC
N2OrgDissRefr
O2Biomass
N/A
02N
KSed
TemperatureReference
Salinity Reference
N/A
N/A
Wet2DrySLab
Wet2DrySRefr
Wet2DryPLab
Wet2DryPRefr
KD_P_Calcite
Animal Underlying
Data
N/A
Animal Type
Taxonomic type or guild
N/A
FHalfSat
CMax
BMin
Sorting
DESCRIPTION
Ratio of nitrate to dissolved refractory organic matter
Ratio of oxygen to organic matter
Not utilized as a parameter by the code.
Ratio of oxygen to nitrogen
Intrinsic sedimentation rate
Reference temperature of water for calculating detrital sinking rate
Reference salinity of water for calculating detrital sinking rate
Not utilized as a parameter by the code.
Not utilized as a parameter by the code.
Wet weight to dry weight ratio for suspended labile detritus
Wet weight to dry weight ratio for suspended refractory detritus
Wet weight to dry weight ratio for particulate labile detritus
Wet weight to dry weight ratio for particulate refractory detritus
Partition coefficient for phosphorus to calcite
For each animal in the simulation, the following
Parameters are required
Animal's Name. Used for Reference only.
Animal type (fish, pelagic invert, benthic invert, benthic insect)
Taxonomic type or trophic guild
Associates animal with appropriate toxicity data
Half-saturation constant for feeding by a predator
Maximum feeding rate for predator
Minimum prey biomass needed to begin feeding
Fractional degree to which there is selective feeding
UNITS
fraction dry weight
unitless ratio
unitless ratio
unitless ratio
m/d
deg. c
%o
g/m2-d
g/m2-d
ratio
ratio
ratio
ratio
L/kg

N/A
Choice from List
Choice from List
Choice from List
g/m3
g/g'd
g/m3 or g/m2
Unitless
                                            324

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Susp. Sed. Affect Feeding
Slope for Sed. Response
Intercept for Sed. Resp.
Temp Response Slope
Optimum Temperature
Maximum Temperature
Min Adaptation Temp
Endogenous Respiration
Specific Dynamic Action
Excretion:Respiration
N to Organic s
P to Organic s
Wet to Dry
Gamete : Biomass
Gamete Mortality
Mortality Coefficient
Sensitivity to Sediment
Ortanism is Sensitive to
Percent Embeddedness
Percent Embeddedness
Threshold
Carrying Capacity
Average Drift
Trigger: Deposition Rate
INTERNAL
SuspSedFeeding
SlopeSSFeed
InterceptSSFeed
Q10
TOpt
TMax
TRef
EndogResp
KResp
KExcr
N20rg
P20rg
Wet2Dry
PctGamete
GMort
KMort
SensToSediment
SenstoPctEmbed
PctEmbedThreshold
KCap
AveDrift
Trigger
TECH DOC
Option to use eqn.
SlopeSS
InterceptSS
Q10
TOpt
TMax
TRef
EndogResp
KResp
KExcr
N20rg
P20rg
Wet2Dry
PctGamete
GMort
KMort
Sensitivity Categories
N/A
embeddedness threshold
value
KCap
Dislodge
Trigger
DESCRIPTION
Does suspended sediment affect feeding
Slope for sediment response
Intercept for sediment response
Slope or rate of change in process per 10°C temperature change
Optimum temperature for given process
Maximum temperature tolerated
Adaptation temperature below which there is no acclimation
Basal respiration rate at 0° C for given predator
Proportion assimilated energy lost to specific dynamic action
Proportionality constant for excretion:respiration
Ratio of nitrate to organic matter for given species
Ratio of phosphate to organic matter for given species
Ratio of wet weight to dry weight for given species
Fraction of adult predator biomass that is in gametes
Gamete mortality
Intrinsic mortality rate
Which equation to use for mortality due to sediment
If this checkbox is checked then the organism will be sensitive to
the sites calculated embeddedness as a function of TSS
If the site's calculated embeddedness exceeds this value, mortality
for the organism is set to 100%
Carrying capacity
Fraction of biomass subject to drift per day
deposition rate at which drift is accelerated
UNITS
Boolean
Unitless
Unitless
Unitless
°C
°C
°C
day1
Unitless
Unitless
fraction dry weight
fraction dry weight
Ratio
Unitless
1/d
1/d
"Zero," "Tolerant,"
"Sensitive,"
"Highly Sensitive"
Boolean
percent (0-100)
g/m2
fraction / day
kg/m2 day
                                            325

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Frac. in Water Column
VelMax
Removal due to Fishing
Mean lifespan
Fraction that is lipid
Mean Wet Weight
Low O2: Lethal Cone
LowO2: Pet. Killed
LowO2: EC50 Growth
LowO2: EC50 Reproduction
Ammonia Toxicity: LC50,
Total Ammonia (pH=8)
Salinity Ingestion Effects
Salinity Gamete Loss Effects
Salinity Respiration Effects
Salinity Mortality Effects
Percent in Riffle
Percent in Pool
INTERNAL
FracInWaterCol
VelMax
Fishing Frac
LifeSpan
FishFracLipid
MeanWeight
O2_LethalConc
O2_LethalPct
O2_EC50growth
O2_EC50repro
Ammonia_LC50
Salmin_Ing, SalMax_Ing,
Salcoeffl Ing, Salcoeff2 Ing
Salmin Gam, SalMax Gam,
Salcoeffl Gam, SalcoeffZ Gam
Salmin Rsp, SalMax Rsp,
Salcoeffl _Rsp, Salcoeff2_Rsp
Salmin_Mort, SalMax_Mort,
Salcoeffl _Mort,
Salcoeff2_Mort
PrefRiffle
PrefPool
TECH DOC
FraCWaterColumn
VelMax
fraction fished
LifeSpan
LipidFrac
WetWt
LCKnownduratlon
PctKilled K,,^,,
JiC50 duration
EC50 duration
LC50t8
SalMin, SalMax,
SalCoeffl,SalCoeff2
SalMin, SalMax,
SalCoeffl,SalCoeff2
SalMin, SalMax,
SalCoeffl,SalCoeff2
SalMin, SalMax,
SalCoeffl,SalCoeff2
PreferenceHabitat
PreferenceHabitat
DESCRIPTION
Fraction of organism in water column, differentiates from pore-
water uptake if the multi-layer sediment model is included
Maximum water velocity tolerated
Daily loss of organism due to fishing Pressure
Mean lifespan in days
Fraction of lipid in organism
Mean wet weight of organism
Concentration where there is a known mortality over 24 hours
The percentage of the organisms killed at the Icknown level above.
Concentration where there is 50% reduction in growth over 24
hours
Concentration where there is 50% reduction in reproduction over
24 hours
LC50total ammonla at 20 degrees centigrade and ph of 8
Parameters used to calculate the effects of the current level of
salinity on ingestion for the given animal
Parameters used to calculate the effects of the current level of
salinity on gamete loss for the given animal
Parameters used to calculate the effects of the current level of
salinity on respiration for the given animal
Parameters used to calculate the effects of the current level of
salinity on mortality of the given animal
Percentage of biomass of animal that is in riffle, as opposed to run
or pool
Percentage of biomass of animal that is in pool, as opposed to run
or riffle
UNITS
Fraction
cm/s
Fraction
Days
g lipid/g org. Wet
g wet
mg/L (24 hour)
Percentage
mg/L (24 hour)
mg/L (24 hour)
mg/L (ph=8)
%o, %o, unitless,
unitless
%o, %o, unitless
%o, %o, unitless,
unitless
%o, %o, unitless,
unitless
%
%
                                            326

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Fish spawn automatically,
based on temperature range
Fish spawn of the following
dates each year
Fish can spawn an unlimited
number of times. . .
Use Allometric Equation to
Calculate Maximum
Consumption
Intercept for weight
dependence
Slope for weight dependence
Use Allometric Equation to
Calculate Respiration
RA
RB
Use "Set 1 " of Respiration
Equations
RQ
RTL
ACT
RTO
RK1
BACT
INTERNAL
Auto Spawn
SpawnDatel..3
UnlimitedSpawning
UseAllom_C
CA
CB
UseAllom_R
RA
RB
UseSetl
RQ
RTL
ACT
RTO
RK1
BACT
TECH DOC










RQ
RTL
ACT
RTO
RK1
BACT
DESCRIPTION
Does AQUATOX calculate Spawn Dates
User entered spawn dates
Allow fish to spawn unlimited times each year
Use allometric consumption equation
Allometric consumption parameter
Allometric consumption parameter
Use allometric consumption respiration
Intercept for species specific metabolism
Weight dependence coefficient
Use "Set 1" of Allometric Respiration Parameters
Allometric respiration parameter
Temperature below which swimming activity is an exponential
function of temperature
Intercept for swimming speed for a Ig fish
Coefficient for swimming speed dependence on metabolism
Intercept for swimming speed above the threshold temperature
Coefficient for swimming at low temperatures
UNITS
true/false
Date
true/false
true/false
real number
real number
true/false
real number
real number
true/false
real number
°C
cm/s
s/cm
cm/s
1/°C
                                            327

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
RTM
RK4
ACT
Preference (ratio)
Egestion (frac.)
INTERNAL
RTM
RK4
ACT
TrophInt.Pref[ ]
TrophInt.Egest[ ]
TECH DOC

RK4

Prefprey,pred
EgestCoeffprey,pred
DESCRIPTION
Not currently used as a parameter by the code
Weight-dependent coefficient for swimming speed
Intercept of swimming speed vs. Temperature and weight
Initial preference value from the animal parameter screen
Fraction of ingested prey that is egested
UNITS

real number
real number
Unitless
Unitless

Plant
Plant Type
Plant is Surface Floating
Macrophyte Type
Taxonomic Group
Toxicity Record
Saturating Light
Use Adaptive Light
Max. Saturating Light
Min. Saturating Light
P Half-saturation
N Half-saturation
Inorg C Half-saturation
Temp Response Slope
Optimum Temperature
PlantRecord
PlantName
PlantType
SurfaceFloating
Macrophyte_Type
Taxonomic_Type
ToxicityRecord
LightSat
UseAdaptiveLight
MaxLightSat
MinLightSat
KP04
KN
KCarbon
Q10
TOpt
Plant Underlying
Data

Plant Type
SurfaceFloating
Macrophyte Type
Taxonomic Group
N/A
LightSat
Adaptive Light
user-entered maximum
user-entered minimum
KP
KN
KC02
Q10
TOpt
For each Plant in the Simulation, the following
Parameters are required
Plant's name. Used for reference only.
Plant type: (Phytoplankton, Periphyton, Macrophytes, Bryophytes)
Is this plant surface floating and therefore subject to a shallowlight
climate as well as excluded from the hypolimnion.
Benthic, rooted floating, free-floating
Taxonomic group
Associates plant with appropriate toxicity data
Light saturation level for photosynthesis
Choice whether to use adaptive light construct
Maximum light saturation allowed from adaptive light equation
Minimum light saturation allowed from adaptive light equation
Half-saturation constant for phosphorus
Half-saturation constant for nitrogen
Half-saturation constant for carbon
Slope or rate of change per 10°C temperature change
Optimum temperature

N/A
Choice from List
Boolean
Choice from List
Choice from List
Choice from List
ly/d
Boolean
ly/d
ly/d
gP/m3
gN/m3
gC/m3
Unitless
°C
                                            328

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Maximum Temperature
Min. Adaptation Temp
Max. Photosynthesis Rate
Photorespiration Coefficient
Resp Rate at 20 deg. C
Mortality Coefficient
Exponential Mort Coeff
P to Photosynthate
N to Photosynthate
Light Extinction
Wet to Dry
Fraction that is lipid
Phytoplankton:
Sedimentation Rate (KSed)
Phytoplankton: Temperature
ofObs. KSed
Phytoplankton: Salinity of
Obs. KSed
Phytoplankton: Exp.
Sedimentation Coeff
Macrophytes: Carrying
Capacity
Macrophytes: VelMax
Periphyton: Reduction in
Still Water
INTERNAL
TMax
TRef
PMax
KResp
Resp20
KMort
EMort
P20rg
N2Org
ECoeffPhyto
Wet2Dry
PlantFracLipid
KSed
KSedTemp
KSedSalinity
ESed
Carry_Capac
Macro VelMax
Red_Still_Water
TECH DOC
TMax
TRef
PMax
KResp
Resp20
KMort
EMort
P20rg
N2Org
EcoeffPhyto
Wet2Dry
LipidFrac
KSed
TemperatureReference
Salinity Reference
ESed
KCap
VelMax
RedStillWater
DESCRIPTION
Maximum temperature tolerated
Adaptation temperature below which there is no acclimation
Maximum photo synthetic rate
Coefficient of proportionality between. Excretion and
photosynthesis at optimal light levels
Respiration rate at 20°C
Intrinsic mortality rate
Exponential factor for suboptimal conditions
Ratio of phosphate to organic matter for given species
Ratio of nitrate to organic matter for given species
Attenuation coefficient for given alga
Ratio of wet weight to dry weight for given species
Fraction of lipid in organism
Intrinsic settling rate
Reference temperature of water for calculating Nhytoplankton
sinking rate
Reference salinity of water for calculating Nhytoplankton sinking
rate
Exponential settling coefficient
Macrophyte carrying capacity, converted to g/m3 and used to
calculate washout of free-floating macrophytes
Velocity at which total breakage occurs
Reduction in photosynthesis in absence of current
UNITS
°C
°c
1/d
Unitless
g/g'd
g/g'd
g/g'd
fraction dry weight
fraction dry weight
l/m-g/m3w
Ratio
g lipid/g org. Wet
m/d
deg.C
%o
Unitless
g/m2
cm/s
Unitless
                                            329

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Periphyton: Critical Force
(FCrit)
Percent Lost in Slough Event
Percent in Riffle
Percent in Pool
Salinity Photosyn. Effects
Salinity Mortality Effects
INTERNAL
FCrit
PctSloughed
PrefRiffle
PrefPool
Salmin Phot, SalMax Phot,
Salcoeffl_Phot,
Salcoeff2_Phot
Salmin_Mort, SalMax_Mort,
Salcoeffl_Mort,
Salcoeff2_Mort
TECH DOC
FCrit
FracSloughed
PrefRiffle
PrefPool
SalMin, SalMax,
SalCoeffl,SalCoeff2
SalMin, SalMax,
SalCoeffl,SalCoeff2
DESCRIPTION
Critical force necessary to dislodge given periphyton group
Fraction of biomass lost at one time
Percentage of biomass of plant that is in riffle, as opposed to run or
pool
Percentage of biomass of plant that is in pool, as opposed to run or
riffle
Parameters used to calculate the effects of the current level of
salinity on photosynthesis for the given plant
Parameters used to calculate the effects of the current level of
salinity on mortality for the given plant
UNITS
newtons (kg m/s2)
%
%
%
%o, %o, unitless,
unitless
%o, %o, unitless,
unitless
                                            330

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

LC50
LC50 exp time (h)
K2 Elim rate const
Kl Uptake const
BCF
Biotrnsfm rate
EC50 growth
Growth exp (h)
EC50 repro
Repro exp time (h)
Mean wet weight (g)
Lipid Frac
Drift Threshold (ug/L)


EC50 photo
EC50 exp time (h)
INTERNAL
AnimalToxRecord
LC50
LC50_exp_time
K2
Kl
BCF
BioRateConst
EC50_growth
Growth_exp_time
EC50_repro
Repro_exp_time
Mean_wet_wt
Lipid frac
Drift_Thresh

TPlantToxRecord
EC50_photo
EC 50 exp time
TECH DOC
Animal Toxicity
Parameters
LC50
ObsTElapsed
K2
Kl
BCF
BioRateConst
ECSOGrowth
ObsTElapsed
ECSORepro
ObsTElapsed
WetWt
LipidFrac
Drift Threshold

Plant Toxicity
Parameter
ECSOPhoto
ObsTElapsed
DESCRIPTION
For each Chemical Simulated, the following
Parameters are required for each animal simulated
Concentration of toxicant in water that causes 50% mortality
Exposure time in toxicity determination
Elimination rate constant
Uptake rate constant, only used if "Enter Kl" option is selected
Bioconcentration factor, only used if "Enter BCF" option is
selected
Percentage of chemical that is biotransformed to
Specific daughter products
External concentration of toxicant at which there is a 50%
reduction in growth
Exposure time in toxicity determination
External concentration of toxicant at which there is a 50%
reduction in reprod
Exposure time in toxicity determination
Mean wet weight of organism
Fraction of lipid in organism
Concentration at which drift is initiated

For each Chemical Simulated, the following
Parameters are required for each plant simulated
External concentration of toxicant at which there is 50% reduction
in photosynthesis
Exposure time in toxicity determination
UNITS

pg/L
H
1/d
L / kg dry day
L/kgdry
1/d
pg/L
H
PS/L
H
G
g lipid/g wet wt.
pg/L


pg/L
H
                                            331

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
EC50 dislodge
K2 Elim rate const
Kl Uptake const
BCF
Biotrnsfm rate
LC50
LC50 exp.time (h)
Lipid Frac


Initial Condition
Gas-phase cone.
Loadings from Inflow
Loadings from Point Sources
Loadings from Direct
Precipitation
Nonpoint-source Loadings
Biotransformation
INTERNAL
EC50_dislodge
K2
Kl
BCF
BioRateConst
LC50
LC50_exp_time
Lipid_frac

TChemical
InitialCond
Tox Air
Loadings
Alt LoadingsfPointsource]
Alt LoadingsfDirect Precip]
Alt_Loadings[NonPointsource]
BioTransf ]
TECH DOC
ECSODislodge
K2
Kl
BCF
BioRateConst
LC50
ObsTElapsed
LipidFrac

Chemical
Parameters
Initial Condition
Toxicantair
Inflow Loadings
Point Source Loadings
Direct Precipitation Load
Non-Point Source Loading
Biotransform
DESCRIPTION
For periphyton only: external concentration of toxicant at which
there is 50% dislodge of periphyton
Elimination rate constant
Uptake rate constant, only used if "Enter Kl" option is selected
Bioconcentration factor, only used if "Enter BCF" option is
selected
Percentage of chemical that is biotransformed to
Specific daughter products
Concentration of toxicant in water that causes 50% mortality
Exposure time in toxicity determination
Fraction of lipid in organism

For each Chemical to be simulated, the following
Parameters are required
Initial Condition of the state variable
Gas-phase concentration of the pollutant
Daily loading as a result of the inflow of water
Daily loading from point sources
Daily loading from direct precipitation
Daily loading from non-point sources
Percentage of chemical that is biotransformed to specific daughter
products
UNITS
pg/L
1/d
L / kg dry day
L / kg dry
1/d
PS/L
H
g lipid/g org. Wet


pg/L
g/m3
pg/L
g/d
g/m2 -d
g/dTox_AirGas-
phase
concentrationg/m3
%
                                            332

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Initial Condition

Loadings from Point Sources
Loadings from Direct
Precipitation
Non-point source loadings
Fraction of Phosphate
Available


Initial Condition
Initial Condition
Loadings from Inflow
(Toxicant) Loadings
INTERNAL
TRemineralize
InitialCond
Loadings
Alt_Loadings[Pointsource]
Alt_Loadings[Direct Precip]
Alt_Loadings[NonPointsource]
FracAvail

TSedDetr
InitialCond
TToxicant.InitialCond
Loadings
TToxicant.Loads
TECH DOC
Nutrient Parameters
Initial Condition
Inflow Loadings
Point Source Loadings
Direct Precipitation Loa
Non-Point Source Loading


Sed. Detritus
Parameters
Initial Condition
Toxicant Exposure
Inflow Loadings
Tox Exposure of Inflow L
DESCRIPTION
For each Nutrient to be simulated, O2 and CO2,
the following Parameters are required
Initial Condition of the state variable (TotP or TotN optional)
Daily loading as a result of the inflow of water (TotP or TotN
optional)
Daily loading from point sources
Daily loading from direct precipitation
Daily loading from non-point sources
Fraction of phosphate loadings that is available versus that which
is tied up in minerals

For the Labile and Refractory Sedimented
Detritus compartments, the following Parameters
are required
Initial Condition of the labile or refractory sedimented detritus
Initial Toxicant Exposure of the state variable, for each chemical
Daily loading of the sedimented detritus as a result of the inflow of
water
Daily parameter; Toxicant Exposure of each type of inflowing
detritus, for each chemical
UNITS

mg/L
mg/L
g/d
g/m2 -d
g/d
Unitless


g/m2
ug/kg
mg/L
ug/kg
                                            333

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Initial Condition
Initial Condition: %
Particulate
Initial Condition: %
Refractory
Inflow Loadings
Dissolved / Particulate
Breakdown
Labile / Refractory
Breakdown
Loadings from Point Sources
Nonpoint-source Loadings
(Associated with Organic
Matter)
(Toxicant) Initial Condition
(Toxicant) Loadings
(associated with Organic
Matter)

INTERNAL
TDetritus
InitialCond
Percent_Part_IC
Percent_Refr_IC
Loadings
Percent Part
Percent Refr
Alt_Loadings[Pointsource]
Alt_Loadings
TToxicant.InitialCond
TToxicant.Loads

TECH DOC
Susp & Dissolved
Detritus
Initial Condition


Inflow Loadings
Percent Particulate Inflow,
Point Source, Non-Point
Source
Percent Refractory Inflow,
Point Source, Non-Point
Source
Point Source Loadings
Non-Point Source Loading
Toxicant Exposure
Tox Exposure of Inflow
Loading

DESCRIPTION
For the Suspended and Dissolved Detritus
compartments, the following Parameters are
required
Initial Condition of suspended & dissolved detritus, as organic
matter, organic carbon, or biochemical oxygen demand
Percent of Initial Condition that is particulate as opposed to
dissolved detritus
Percent of Initial Condition that is refractory as opposed to labile
detritus
Daily loading as a result of the inflow of water
Three constant or time-series parameters; % of each type of
loading that is particulate as opposed to dissolved detritus
Three constant or time-series parameters; % of each type of
loading that is refractory as opposed to labile detritus
Daily loading from point sources
Daily loading from non-point sources
Initial Toxicant Exposure of the suspended and dissolved detritus
Daily parameter; Toxicant Exposure of each type of inflowing
detritus, for each chemical

UNITS

mg/L
Percentage
Percentage
mg/L
Percentage
Percentage
& organic matter'^
& organic matter'^
ug/kg
ug/kg

                                            334

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Initial Condition
(Toxicant) Initial Condition


Initial Condition
Loadings from Inflow
(Toxicant) Initial Condition
(Toxicant) Loadings


Initial Condition
Loadings from Inflow
(Toxicant) Initial Condition
(Toxicant) Loadings
Preference (ratio)
Egestion (frac.)
INTERNAL
TBuried Detritus
InitialCond
TToxicant.InitialCond

TPlant
InitialCond
Loadings
TToxicant.InitialCond
TToxicant.Loads

TAnimal
InitialCond
Loadings
Ttoxicant. InitialCond
TToxicant.Loads
TrophlntArray.Pref
TrophlntArray.ECoeff
TECH DOC
Buried Detritus
Initial Condition
Toxicant Exposure

Plant Parameters
Initial Condition
Inflow Loadings
Toxicant Exposure
Tox Exposure of Inflow L

Animal Parameters
Initial Condition
Inflow Loadings
Toxicant Exposure
Tox Exposure of Inflow L
Prefprey, pred
EgestCoeff
DESCRIPTION
For Each Type of Buried Detritus, the following
Parameters are required
Initial Condition of the labile and refractory buried detruitus
Initial Toxicant Exposure of the labile and refractory buried
detritus , for each chemical simulated

For each plant type simulated, the following
Parameters are required
Initial Condition of the plant
Daily loading as a result of the inflow of water
Initial Toxicant Exposure of the plant
Daily parameter; Toxicant exposure of the Inflow Loadings, for
each chemical simulated

For each animal type simulated, the following
Parameters are required
Initial Condition of the animal
Daily loading as a result of the inflow of water
Initial Toxicant Exposure of the animal
Daily parameter; toxic exposure of the Inflow Loadings, for each
chemical simulated
For each prey -type ingested, a preference value within the matrix
of preferences
For each prey -type ingested, the fraction of ingested prey that is
egested
UNITS

Kg/cu. m
Kg/cu. m


mg/L or g/m2 dry
mg/L or g/m2 dry
ug/kg
ug/kg


mg/L or g/sq.m
also expressed as
g/m2
mg/L or g/sq. m
ug/kg
ug/kg
Unitless
Unitless
                                            335

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Initial Condition
Water volume
Inflow of Water
Discharge of Water


Site Type
Frac. of Site that is Shaded
Water Velocity
Site Mean Depth


Initial Condition
Could this system stratify
Valuation or loading

INTERNAL
TVolume
InitialCond
Volume
InflowLoad
DischargeLoad

Site Characteristics
SiteType
Shade
Dyn Velocity
DynZMean

Temperature
InitialCond



TECH DOC
Volume Parameters
Initial Condition
Volume
Inflow of Water
Discharge of Water

Site Characteristics
Site Type
user input shade
user entered velocity
user entered mean depth

Temperature
Initial condition



DESCRIPTION
For each segment simulated, the following water
flow parameters are required
Initial Condition of the water volume .
Choose method of calculating volume; choose between Manning's
equation, constant volume, variable depending upon inflow and
discharge, or use known values
Inflow of water; daily parameter, can choose between constant and
dynamic loadings
Discharge of water; daily parameter, can choose between constant
and dynamic loadings

The following Parameters are required
Site type affects many portions of the model.
Fraction of site that is shaded, time-series
Optional, time series of run velocities
Optional, time series of mean depth for site

Temperature Parameters Required
Initial temperature of the segment or layer (if vertically stratified
could system vertically stratify
Temperature of the segment. Can use annual means for each
stratum and constant or dynamic values

UNITS

m3
cu. M
m3/d cu m/d
m3/d


Pond, Lake,
Stream, Reservoir,
Enclosure, Estuary
Fraction
cm/s
M


°C
true/false
°C

                                            336

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE

Initial Condition
Mean Value
Wind Loading


Initial Condition
Loading
Photoperiod


Initial Condition
State Variable Valuation
Mean alkalinity

Sand / Silt / Clay
Initial Susp. Sed.
Frac in Bed Seds
Loadings from Inflow
INTERNAL
Wind
InitalCond
MeanValue
Wind

Light
Light
Loadsrec
Photoperiod

pH
InitialCond
pH
alkalinity

TSediment
InitialCond
FracInBed
Loadings
TECH DOC
Wind


Wind

Light
Light

Photoperiod

pH

pH
alkalinity

Inorganic Sediment
Parameters
Initial Condition
FracSed
Inflow Loadings
DESCRIPTION
Wind parameters required
Initial wind velocity 10m above the water
Mean wind velocity
Daily parameter; wind velocity 10m above the water; 1, can
choose default time series, constant or dynamic loadings

Light Parameters Required

Daily parameter; avg. light intensity at segrment top; can choose
annual mean, constant loading or dynamic loadings
Fraction of day with daylight; optional, can be calculated from
latitude

pH Parameters Required
Initial pH value
pH of the segment; can choose constant or daily value.
mean Gran alkalinity (if dynamic pH option selected)

If the inorganic sediments model is included in
AQUATOX, the following Parameters are required
for sand, silt, and clay
Initial Condition of the sand, silt, or clay
Fraction of the bed that is composed of this inorganic sediment.
Fractions of sand, silt, and clay must add to 1 .0
Daily sediment loading as a result of the inflow of water
UNITS

m/s
m/s
m/s


ly/d

hr/d


PH
PH
ueq CaCO3/L


mg/L
Fraction
mg/L
                                            337

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Loadings from Point Sources
Loadings from Direct
Precipitation
Non-point source loadings

Multi-Layer Sediment
Model
Densities [Organic and
Inorganic Components]
Multi-Layer Sediment
Model
Max Thickness of Active
Layer
Min Thickness of Active
Layer
Cohesives, NonCohesives,
Daily Scour
Cohesives, NonCohesives,
Daily Deposition
Cohesives only, Erosion
Velocity
Cohesives only, Deposition
Velocity
Multi-Layer Sediment
Model
Thickness
Diffusion Coefficient for top
of sediment layer
INTERNAL
Alt LoadingsfPointsource]
Alt LoadingsfDirect Precip]
Alt LoadingsfNonPointsource]

Global SedData
Densities
Active Layer SedData
MaxUpperThick
BioTurbThick
LScour
LDeposition
LErodVel
LDepVel
Each Layer SedData
BedDepthIC
UpperDispCoeff
TECH DOC
Point Source Loadings
Direct Precipitation Loa
Non-Point Source Loading

Multi-layer
Sediment
Parameters
Density Sed
Multi-layer
Parameters
user defined maximum
thickness
user defined minimum
thickness
Erode Sed
DepositSed
ErodeVel
DepVel
Multi-layer
Parameters
thickness
DiffCoeff
DESCRIPTION
Daily loading from point sources
Daily loading from direct precipitation
Daily loading from non-point sources

If the multi-layer sediment model is included in
AQUATOX, the following general parameters are
required
Density of each organic and inorganic component of the sediment
bed.
If the multi-layer sediment model is included these
parameters are required/or the active layer only
Maximum thickness of the active layer before it becomes split into
multiple layers
Minimum thickness of active layer before it is added to the layer
below it
Scour of this sediment to the water column above
Deposit of this sediment from the water column
User input time-series of cohesives erosion velocities, used to
calculate scour of organics
User input time-series of cohesives deposition velocities, used to
calculate deposition of organics
If the multi-layer sediment model is included these
parameters are required/or each layer modeled
Initial thickness of each modeled layer
Dispersion coefficient
UNITS
g/d
Kg-d
g/d


g/cm3

M
M
g/d
g/d
m/d
m/d

M
m2/d
                                            338

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
Pore Water Init. Cond.
ROOM, LOOM PoreW,
Initial Cond
Cohesives, NonCohesives,
Initial Cond
R Detr Sed, L Detr Sed,
Initial Cond
Chemical Exposures

Trophic Interactions,
BCFs for Shorebirds
Preference (ratio)
Biomagnification Factor
Clearance Rate

Link Between Two
Segments
Type of Link
Link Name
FromSeg, ToSeg
Characteristic Length
Water Flow Data
Dispersion Coeff
INTERNAL
TPoreWater.InitialCond
TDOMPorewater.InitialCond
TBottomSediment.InitialCond
TBuriedSed.InitialCond
[ComponentJTox.InitialCond

Gull Parameters
GullPref
GullBMF
GullClear

TSegmentLink
LinkType
Name
FromlD, ToID
CharLength
WaterFlowData
DiffusionData
TECH DOC
Cone sed Initial Cond.
Cone sed Initial Cond.
Cone sed Initial Cond.
Cone sed Initial Cond.
ToxicantBottomSed Initial
Cond.

Shorebirds
Prpf
riciprey; pred
BMFTox
ClearTox

Multi-Segment
Model
two types of linkages


CharLength
Discharge
DiffusionThlsSeg
DESCRIPTION
Concentration of pore water initial condition
Concentration of refractory or labile DOM in pore water, initial
condition
Concentration of inorganic sediments in the layer, initial condition
Concentration of refractory and labile organic sediments in the
layer, initial condition
Concentration of relevant toxicant in element of sediment layer

If the shorebird model is included in a simulation,
the following Parameters are required
For each prey-type ingested, a preference value within the matrix
of preferences
Biomagnification factor for this chemical in gull
Clearance rate for the given toxicant in gulls

If the multi-segment model is used for a simulation,
the following Parameters are required for each link
between segments
Indicates whether linkage is unidirectional or bidirectional
Used for the user to keep track of linkages
Used for the model to keep track of linkages
Characteristic mixing length, feedback links only
Time-series of water flow from one segment to the next
Time-series of dispersion coefficients between two segments,
feedback links only
UNITS
m3 water / m2
g/m3
gm2
g/m2 dry
ug/L pore water,
ug/kg solids


Unitless
Unitless
day1


"cascade" or
"feedback"
String
existing segments
M
m3/d
m2/d
                                            339

-------
AQUATOX (RELEASE 3.1) TECHNICAL DOCUMENTATION
APPENDICES
USER INTERFACE
XSection of Boundary
BedLoadImogamcs
INTERNAL
XSectionData
BedLoad
TECH DOC
Area
BedloadUpstreamlmk
DESCRIPTION
Time-series of cross sectional areas between two segments,
feedback links only
Time-series of bedload from the upstream segment to the
downstream segment
UNITS
7
m
g/d
                                            340

-------