United States
Environmental Protection
Agency
Office of Water
(4305)
EPA-823-R-01-008
November 2001
&EPA AQUATOX FOR WINDOWS
A MODULAR FATE AND EFFECTS MODEL
FOR AQUATIC ECOSYSTEMS
RELEASE 1.1
VOLUME 3: MODEL VALIDATION REPORTS
ADDENDUM
-------
AQUATOX FOR WINDOWS
A MODULAR FATE AND EFFECTS MODEL
FOR AQUATIC ECOSYSTEMS
RELEASE 1.1
VOLUME 3: MODEL VALIDATION REPORTS
ADDENDUM: FORMULATION, CALIBRATION, AND
VALIDATION OF A PERIPHYTON SUBMODEL FOR
AQUATOX RELEASE 1.1
NOVEMBER 2001
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF WATER
OFFICE OF SCIENCE AND TECHNOLOGY
WASHINGTON DC 20460
11
-------
DISCLAIMER
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.
This document presents a calibration and validation for a new application for AQUATOX, an aquatic
ecosystem simulation model which was originally released in September 2000. It is not intended to
serve as guidance or regulation, nor is the use of this model in any way required. This document
cannot impose legally binding requirements on EPA, States, Tribes, or the regulated community.
ACKNOWLEDGMENTS
Daily flow data for Walker Branch were collected as part of the Walker Branch Watershed research
project supported by the Environmental Sciences Division, Office of Biological and Environmental
Research, U. S. Department of Energy and were provided by Patrick Mulholland, Oak Ridge National
Laboratory. This calibration and validation would not have been possible without the remarkable
factorial experiments and the rigorous data set compiled by Amy Rosemond for her dissertation.
Access to Dr. Rosemond's raw data is especially appreciated.
AQUATOX has been developed and documented by Dr. Richard A. Park of Eco Modeling; most
of the programming has been by Jonathan S. Clough under subcontract to Eco Modeling. It was
funded with Federal funds from the U.S. Environmental Protection Agency, Office of Science and
Technology. This validation study was funded under contract numbers 68-C-98-010 and 68-C-01-
0037 to AQUA TERRA Consultants, Anthony Donigian, Work Assignment Manager. AQUATOX
programming and other support was provided by Jonathan Clough, and his help is gratefully
acknowledged. Marjorie Coombs Wellman, as the EPA work assignment manager, provided
technical direction, encouragement, and editorial insights that were very helpful.
in
-------
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. In the National
Water Quality Inventory: 1996 Report to Congress, 36 percent of assessed river lengths and 39
percent of assessed lake areas were impaired for one or more of their designated uses (US EPA
1998). The most commonly reported causes of impairment in rivers and streams were siltation,
nutrients, bacteria, oxygen-depleting substances, and pesticides; in lakes and reservoirs the causes
also included metals and noxious aquatic plants. The most commonly reported sources of
impairment were agriculture, nonpoint sources, municipal point sources, atmospheric deposition,
hydrologic modification, habitat alteration and resource extraction. There were 2196 fish
consumption advisories, which may include outright bans, in 47 States, the District of Columbia and
American Samoa. Seventy-six percent of the advisories were due to mercury, with the rest due to
PCBs, chlordane, dioxin, and DDT (US EPA 1998). States are not required to report fish kills for
the National Inventory; however, available information for 1992 indicated 1620 incidents in 43
States, of which 930 were attributed to pollution, particularly oxygen-depleting substances,
pesticides, manure, oil and gas, chlorine, and ammonia.
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 offish. It has been implemented for streams, small rivers, ponds,
lakes, and reservoirs.
The AQUATOX model is described in these documents. Volume 1: User's Manual describes the
usage of the model. Volume 2: Technical Documentation provides detailed documentation of the
concepts and constructs of the model so that its suitability for given applications can be determined.
Volume 3: Model Validation Reports presents three model validation studies performed for different
environmental stressors and in different waterbody types. The validations were performed using test
versions of the model which had only very minor differences from Release Version 1; the specific
test version is noted in the title of each report. This Addendum to Volume 3 presents a calibration
and validation study for the use of AQUATOX to simulate periphyton in streams. Constructs were
added to AQUATOX in order to better represent periphyton dynamics; these are described in the
text. Readers should visit the AQUATOX website (http://www.epa.gov/ost/models/aquatox/) to
download the new executable program that contains the new constructs.
IV
-------
MODEL VALIDATION REPORTS
ADDENDUM
TABLE OF CONTENTS
Abstract [[[ v
Introduction
Modifications to Formulations in AQUATOX
Sloughing
Detrital Accumulation
Calibration Strategy
1
1
2
Calibration Results 5
Comparison of Maximum Values 5
Statistical Tests 8
Visual Inspection 10
Process-level Verification 27
Validation 28
Conclusions 30
Bibliography 31
Appendix—Biotic Parameters 33
List of Tables
Table 1. Loadings and Initial Conditions Used in Spring, 1989, Simulations 4
Table 2. Loadings and Initial Conditions Used in Spring, 1990, Simulations 4
Table 3. Loadings and Initial Conditions Used in Summer, 1989, Simulations 5
Table 4. Summary of Treatments Simulated 6
Table 5. Statistics of Overlap Between Predicted and Observed Distributions 9
-------
MODEL VALIDATION REPORTS ADDENDUM
Abstract
The AQUATOX periphyton submodel was successfully calibrated and validated with data from
Walker Branch, Tennessee. A single parameter set for diatoms, green algae, and gastropods was
developed to fit all twenty treatments in a factorial experiment involving low and high nutrient
levels, low and high light levels, and grazing and no grazing in stream-side channels and stream
enclosures. A two-year data set for periphyton on cobbles in the stream bed was used to validate the
model for continuous ambient conditions in this woodland stream. Nitrogen, phosphorus, light, and
grazing were all limiting, both singly and in combination. Inclusion of a sloughing term based on
senescence due to deteriorating environmental conditions and the drag force of currents on exposed
periphyton biomass provided a realistic response to limiting environmental factors and was
especially useful in modeling the effects of variable flow rates on diatoms and filamentous algae.
The model was shown to be calibrated and validated based on the weight of evidence of reproduction
of observed patterns of biomass change, concordance of maxima between predicted and observed
biomass, and equivalence in predicted and observed means and variances as confirmed by relative
bias and F tests.
VI
-------
MODEL VALIDATION REPORTS ADDENDUM
FORMULATION, CALIBRATION, AND VALIDATION OF A PERIPHYTON
SUBMODEL FOR AQUATOX RELEASE 1
Introduction
Periphyton are benthic algae and associated organic detritus that are attached to hard substrates and
macrophytes and that carpet stabilized sands. They are an important constituent of the aquatic
community, especially in shallow lakes, ponds, streams, and rivers (Stevenson, 1996). They also are
an important link for bioaccumulation of organic contaminants (Wang et al., 1999). Periphyton have
been shown to be sensitive to eutrophication of streams (Marcus, 1980; Bothwell, 1985; Hart and
Robinson, 1990; Van Nieuwenhuyse and Jones, 1996; Biggs, 2000). Although they are nominally
included in several ecosystem models, periphyton have been difficult to model. The purpose of this
study is to test alternate formulations, calibrate for several different combinations of environmental
factors, and validate with data from a woodland stream.
Modifications to Formulations in AQUATOX
Originally periphyton were modeled in AQUATOX Release 1 as if they were merely non-sinking
phytoplankton subject to current-induced scour. However, two processes distinguish periphyton
from their free-living counterparts. First, as periphyton die, detritus accumulates and remains a part
of the operational biomass (best characterized as ash-free dry weight). Second, as environmental
conditions deteriorate or water velocity increases, sloughing occurs and large proportions of biomass
are lost instantaneously. AQUATOX has been modified to account for both these processes.
Sloughing
Earlier attempts at calibration were unsuccessful because they were unable to reproduce the buildup
and then decline in biomass in 1989 at Walker Branch. As a result, the formulation for scour was
reexamined and the more 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.
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
where:
DragForce = drag force (kg m/s2);
Rho = density (kg/m3);
DragCoeff = drag coefficient (2.53E-4, unitless);
Vel = velocity (m/s);
BioVol = biovolume of algae (mm3/mm2);
Unit Area = unit area (mm2);
1E-6 = 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):
1
-------
MODEL VALIDATION REPORTS ADDENDUM
„. , Biomass
Bwvoln. -
Dia 2.08E-9
D. , Biomass
Biovolm, =
m 8.57E-9
where:
BiovolDia = biovolume of diatoms (mm3/mm2);
Biovolpn = biovolume of filamentous algae (mm3/mm2);
Biomass = biomass of given algal group (g/m2).
Suboptimal light and nutrients 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 suboptimal nutrients and light on photosynthesis, and which is multiplied
by 5 to desensitize its effect on sloughing. 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, then sloughing occurs:
If DragForce > Suboptimal o • FCrito then Slough - Biomass • FracSloughed
else Slough - 0
where:
Suboptimal0rg = factor for suboptimal nutrient and light effect on senescence of given
periphyton group (unitless);
FCrit0rg = critical force necessary to dislodge given periphyton group (kg m/s2);
Slough = biomass lost by sloughing (g/m3);
FracSloughed = fraction of biomass lost at one time (90%, unitless).
Suboptimal0 - NutrLimito • LtLimito • 5
If SuboptimalOr > 1 then SuboptimalOr - 1
where:
NutrLimit = nutrient limitation for given algal group (unitless) computed by
AQUATOX;
LtLimit0rg = light limitation for given algal group (unitless) computed by
AQUATOX.
Detrital Accumulation
In phytoplankton, mortality results in immediate production of detritus, and that transfer is modeled.
However, 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 model 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, with compensation through the high internal extinction rate
constant.
-------
MODEL VALIDATION REPORTS ADDENDUM
Calibration Strategy
One of the difficulties with modeling periphyton is that there are few parameter values that are based
on laboratory or field studies. Many studies have been conducted, but not with the goal of obtaining
process-level estimates that can be used in modeling. Therefore, an additional burden is placed on
the calibration effort.
The strategy for calibration involving several poorly defined parameters has been to use the results
of field experiments on Walker Branch, Tennessee, conducted by Rosemond (1993) involving a
factorial design (see also Mulholland and Rosemond, 1992; Rosemond, 1994; Rosemond, 1995;
Rosemond et al., 1996; Rosemond et al., 2000). Walker Branch is a small woodland stream with
naturally low discharge, nutrient-poor conditions, intense grazing, and low light levels (Rosemond,
1993). Nutrients were increased and grazers (gastropods) were removed in Spring, 1989, and
nutrients were increased, grazers were removed, and light levels were increased in Summer, 1989,
in stream-side channels; nutrients were increased and grazers were removed in Spring, 1990, in
stream enclosures (Tables 1-3). Poorly defined parameters such as light saturation levels, maximum
photosynthetic rates, and light extinction due to self-shading were calibrated for diatoms and green
algae; and the maximum consumption rate, minimum biomass for feeding, and carrying capacity
were calibrated for gastropods (see Appendix). The calibrations were then validated with a two-year
data set from Walker Branch (Rosemond, 1993).
All simulations were conducted with exactly the same set of parameter values so that only the site
characteristics (such as channel morphometry) and loadings (such as discharge, light, and nutrient
loadings) were changed among simulations. Calibrations were conducted on each successive
parameter in an iterative approach.
-------
MODEL VALIDATION REPORTS
ADDENDUM
Table 1. Loadings and Initial Conditions Used in Spring, 1989, Simulations of Walker Br.
Spring, 1989 (March 15 to May 3)
Initial gastropod
biomass (g/m2)*
Nutrients (mg/L)
Light (Ly/d)+
Velocity (cm/s)
Initial algal
biomass (g/m2)*
Treatment
Grazed
Ungrazed
NH4-N
NO3-N
P04-P
Diatoms
Greens
Control
2.5
0
0.006
0.0183
0.00194
57.5 to 10.5
9.44
0.0236
0.0118
High
Nutrients
2.5
0
0.0338
0.2078
0.0366
57.5 to 10.5
9.44
0.019
0.0096
HighN
2.5
0
0.0338
0.2078
0.00194
57.5 to 10.5
9.44
0.0324
0.0162
HighP
2.5
0
0.006
0.0183
0.0366
57.5 to 10.5
9.44
0.0324
0.0162
+ Light was reported by Rosemond (1993) as photosynthetic active radiation, which was multiplied
by 2 to provide total radiation values used by AQUATOX *A11 biomass is ash-free dry weight
Table 2. Loadings and Initial Conditions Used in Spring, 1990, Simulations of Walker Br.
Spring, 1990 (March 5 to April 24)
Initial gastropod biomass
(g/m2)*
Nutrients (mg/L)
Light (Ly/d)+
Velocity (cm/s)
Initial algal biomass (g/m2)*
differed between treatments
Treatment
Grazed
Ungrazed
NH4-N
NO3-N
PO4-P
Diatoms
Greens
Control
2.5
0.25
0.002
0.0101
0.0024
45 to 20
9.44
0.0116,0.0178
0.0058, 0.0089
High Nutrients
2.5
0.25
0.0441
0.2233
0.0441
45 to 20
9.44
0.019, 0.02
0.01,0.01
+ Light was reported by Rosemond (1993) as photosynthetic active radiation, which was multiplied
by 2 to provide total radiation values used by AQUATOX *A11 biomass is ash-free dry weight
-------
MODEL VALIDATION REPORTS
ADDENDUM
Table 3. Loadings and Initial Conditions Used in Summer, 1989, Simulations of Walker Br.
Summer, 1989 (July 21 to September 8)
Initial gastropod
biomass (g/m2)
Nutrients (mg/L)
Light (Ly/d)+
Velocity (cm/s)
Initial algal
biomass (g/m2)*
Treatment
Grazed
Ungrazed
NH4-N
NO3-N
PO4-P
Diatoms
Greens
Control
2.5
0
0.006
0.0183
0.00194
15.2 to 20
9.44
0.024, 0.019
0.012, 0.009
High
Nutrients
2.5
0
0.05338
0.2078
0.04366
15.2 to 20
9.44
0.02
0.01
High Light,
Control
2.5
0
0-0.0076
0.0296-
0.0392
0.0027-
0.0049
204
9.44
0.027
0.014
High Light
& Nutrients
2.5
0
0.05338
0.2078
0.04366
204
9.44
0.024
0.012
+ Light was reported by Rosemond (1993) as photosynthetic active radiation, which was multiplied
by 2 to provide total radiation values used by AQUATOX *A11 biomass is ash-free dry weight
Calibration Results
Altogether, twenty treatments were simulated, including controls and manipulated grazing, nutrient,
and light experiments. These are named, described, and keyed to the figures in Table 4.
Comparison of Maximum Values
Data from twenty factorial treatments were used in the calibration. Initially the objective was to
minimize the squared deviations of the observed and predicted maxima. A close fit was obtained
across almost all treatments; however, observed temporal patterns were not well represented (see
next section). Therefore, the criteria were relaxed so that the temporal patterns could be matched
as well. Because grazing usually prevents a buildup of biomass (ash-free dry weight or AFDW), the
maxima of the grazed treatments were fairly uniform (Figure 1). The observed grazed maximum
values were all higher than the predicted values, but often only one point exceeded the predicted.
The ungrazed treatments exhibit the opposite relationship: almost all predicted maxima exceed the
observed (Figure 2). However, the data were taken approximately every two weeks, and it is likely
that observed maxima would miss peak biomass buildup that was followed by sloughing (Rosemond,
pers. comm.). Therefore, it is to be expected that the predicted maxima often would exceed the
observed biomass for treatments where there was stimulation in the absence of grazing. Note the
difference in scale between the two graphs, which emphasizes the magnitude of the grazing pressure.
-------
MODEL VALIDATION REPORTS
ADDENDUM
Table 4. Summary of Treatments Simulated
File Name
WBCtlSpr89
WBCtlSpr90
WBCtlSmr89
WBNSpr89
WBPSpr89
WBN&PSpr89
WBN&PSpr90
WBN&PSmr90
WBHiLtSmr89
WBN&PHiLtSmr89
WBCtlwoGrzSpr89
WBCtllO%GrzSpr90
WBCtlwoGrzSmr89
WBNwoGrzSpr89
WBPwoGrzSpr89
WBN&PwoGrzSpr89
WBN&P 1 0%GrzSpr90
WBN&PwoGrzSmr89
WBHiLtwoGrzSmr89
WBN&PHiLtwoGrzSmr89
Description
control, Spring 1989
control, Spring 1990
control, Summer 1989
N-enriched, Spring 1989
P-enriched, Spring 1989
N- and P-enriched, Spring 1989
N- and P-enriched, Spring 1990
N- and P-enriched, Summer 1989
High light, Summer 1989
N- and P-enriched, high light, Summer 1989
Control, without grazers, Spring 1989
Control, 10% grazers, Spring 1990
Control, without grazers, Summer 1989
N-enriched, without grazers, Spring 1989
P-enriched, without grazers, Spring 1989
N- and P-enriched, without grazers, Spring 1989
N- and P-enriched, 10% grazers, Spring 1990
N- and P-enriched, without grazers, Summer 1989
High light, without grazers, summer, 1989
N- and P-enriched, high light, without grazers,
Summer 1989
Figure No.
22
23
24
16
17
3,4,5,6,11
7
8
28
29
18, 19
20
21
14
15
9, 10,11,19
12
13,27
25
26,27
-------
MODEL VALIDATION REPORTS
ADDENDUM
GRAZED
Figure 4. Comparison of observed (Rosemond, 1993) and predicted maxima for periphyton in
Walker Branch, Tennessee, for various grazed treatments.
UNGRAZED
Figure 5. Comparison of observed (Rosemond, 1993) and predicted maxima for periphyton in
Walker Branch, Tennessee, for various ungrazed treatments.
-------
MODEL VALIDATION REPORTS ADDENDUM
Statistical Tests
Two measures help answer the question: how much overlap is there between data and model
distributions? 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):
= (Pred - Obs)
Sobs
where:
rB = relative bias (standard deviation units);
Pred = mean predicted value;
Obs = mean observed value; and
Sobs = standard deviation of observations.
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:
Pred
obs
Very small F values suggest that the observed data may be too variable to determine the goodness
of fit; very large F values indicate that the predictions are imprecise (Bartell et al., 1992). Large F
values also may indicate that the model is predicting greater fluctuations than can be supported by
sparse data. Assuming normal distributions, the probability that the observed and predicted
distributions are the same can be evaluated. Putting the two tests together, if a comparison has rB
= 0 and F = 1, then the predicted and observed results are identical.
Table 5 provides statistics for the simulations. For purposes of computing the variances and
standard deviations, all observations were pooled, which tended to emphasize the sample variability.
Six out of the twenty simulations exhibited similar predicted and observed means and variances at
the 95% confidence level. Twelve other treatments had similar means, but the distributions were
indeterminate and could not be compared because the observed data were considerably more variable
than predicted. In one treatment, with high light and no grazing, the predictions and observations
were significantly different. In the remaining treatment the variances were similar but the means
were significantly different. In the following graphs, presented for visual inspection, the standard
deviations are plotted for each sample mean, aiding in the evaluation.
-------
MODEL VALIDATION REPORTS
ADDENDUM
Table 5. Statistics of Overlap Between Predicted and Observed Distributions
Fig.
3
7
8
9
12
13
14
15
16
17
18
20
21
22
23
24
25
26
28
29
Simulation
N&P, Spr, 1989
N&P, Spr, 1990
N&P, Smr, 1989
N & P, no Grz, Spr,
1989
N&P, 10% Grz, Spr,
1990
N & P, no Grz, Smr,
1989
N, no Grz, Spr, 1989
P, no Grz, Spr, 1989
N, Spr, 1989
P, Spr, 1989
Ctl,Nogrz, Spr, 1989
Ctl, 10%grz, Spr, 1990
Ctl, no Grz, Smr, 1989
Ctl, Spr, 1989
Ctl, Spr, 1990
Ctl, Smr, 1989
Ctl, Hi Lt, no Grz, Smr,
1989
N & P, Hi Lt, no Grz,
Smr, 1989
Ctl, HiLt, Smr, 1989
Nut, HiLt, Smr, 1989
Mean predicted
0.366
0.411
0.369
1.94
1.91
0.520
1.12
1.05
0.385
0.404
1.18
0.326
0.475
0.336
0.312
0.327
1.85
2.40
0.395
0.499
Mean observed
0.415
0.415
0.386
1.77
1.14
0.687
0.742
0.801
0.384
0.388
0.623
0.349
0.372
0.320
0.348
0.389
1.07
1.68
0.400
0.404
rB
-0.348+
-0.0313+
-0.137+
0.0978+
0.880+
-0.992+
0.810+
0.268+
0.0124+
0.147+
1.80
-0.0856+
0.871+
0.1 72+
-0.299+
-0.570+
2.41
0.609+
-0.0255+
0.580+
F
0.006017
0.02597
0.002717
0.121*
0.591*
0.579*
0.550*
0.09567
-0.01217
0.006477
1.03*
0.03547
0.290*
0.00087
0.002157
0.0009 •
19.3
1.28*
0.002077
0.09677
N - nitrogen enriched, P - phosphorus enriched, Grz - grazers, Ctl -
observed (95% CL), * predicted variance = observed (95% CL), 7
observed variance
control, + predicted mean =
= indeterminate due to high
-------
MODEL VALIDATION REPORTS ADDENDUM
Visual Inspection
Visual inspections of fits of predictions to observed data are useful in evaluating how well patterns
are represented, with allowance for the vagaries of widely spaced data points. Although not
qualitative, they contribute considerably to the weight of evidence that the model is representing the
periphyton dynamics realistically. In the graphs that follow, predictions are shown for diatoms and
other periphyton (nominally greens), and the sum of these is presented for comparison with the
observed data, which are shown with ± 1 standard deviation.
Simulation of the stream-side channel experiment in Spring, 1989, with both nitrogen (N) and
phosphorus (P) enrichment and with normal density of grazers represented a mean response similar
to the five periphyton observations, including the given initial condition (Figure 3). The gastropod
biomass is stable at 2.5 g/m2, compared to observed biomass of 1 to 3.3 (Rosemond, pers. comm.).
Examination of the process rates for diatoms shows that photosynthesis is balanced with grazing
(Figure 4). Although the parameter set was derived to give the best fit to all the treatments, it is
instructive to examine the sensitivity of the simulations to key parameters. There was little response
of diatoms to a normal distribution of values for the maximum photosynthetic rate (0.8 ± 0.24 g/g
d) as shown in Figure 5; this is almost certainly due to suppression by grazing (cf. Hill, 1992). In
another sensitivity analysis, gastropods exhibited only a 10% change in biomass by the end of the
seven weeks, but diatoms exhibited a large and varied response to a normal distribution of values
for the maximum consumption rate of gastropods (0.05 ± 0.015 g/g d) as shown in Figure 6.
Simulation of the comparable stream enclosure experiment in Spring, 1990, exhibited a very similar
response with a mean that was very close to the mean of the observations (Figure 7). Simulation
of the comparable stream-side channel experiment in Summer, 1989, also was close to the mean for
the five observed points (Figure 8).
10
-------
MODEL VALIDATION REPORTS
ADDENDUM
0.8
-Diatoms —"—Others
Periphyton X Observed
Gastropods
Figure 6. Comparison of predicted biomass of periphyton, constituent algae, observed biomass
of periphyton ± 1 standard deviation (Rosemond, 1993, pers. comm.), and gastropods in Walker
Branch, Tennessee, with addition of both N and P in Spring, 1989.
] Photsynthesis D Respiration D Excretion D Mortality D Predation D Sloughing
Figure 7. Predicted rates for diatoms in Walker Branch, Tennessee, with addition of both N and
P in Spring, 1989. Note the importance of photosynthesis and grazing.
11
-------
MODEL VALIDATION REPORTS
ADDENDUM
•Periphyton
1 std dev
1 std dev X. Observed
Figure 8. Sensitivity of periphyton predictions to ± 30% change in maximum photosynthetic
rate for diatoms with addition of N and P in Spring, 1989. Observed data from Rosemond (1993,
pers. comm.).
•Periphyton
+ 1 std dev 1 std dev X Observed Gastropods •
• 1 std dev •
--1 std dev
Figure 9. Sensitivity of periphyton and gastropod predictions to ± 30% change in maximum
consumption rate for gastropods with addition of N and P in Spring, 1989. Observed data from
Rosemond (1993, pers. comm.).
12
-------
MODEL VALIDATION REPORTS
ADDENDUM
I
r 0.4
-Diatoms—"—Others Periphyton X Observed
Figure 10. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993, pers. comm.) in Walker Branch, Tennessee, with
addition of both N and P in Spring, 1990.
0)
.2
m
0.7
0.6
0.5
0.4
0.3
0.2
0.1
-Diatoms
Others
Periphyton X Observed
Figure 11. 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 in Summer, 1989.
13
-------
MODEL VALIDATION REPORTS
ADDENDUM
Without grazers, the Spring, 1989, stream-side, N- and P-enriched experiment exhibited an
exponential buildup of biomass, followed by a decline due to sloughing (Rosemond, 1993). Given
the sampling interval, it is difficult to be sure that the model simulated the dynamics properly;
however, all five data points are matched quite well, and the predicted maximum is within one
standard deviation of the nearest observation (Figure 9). Examination of the constituent rates for
diatoms (?) reveals that sloughing occurs as a threshold biomass is reached or as photosynthesis
starts to decline; in the latter case, seen near the end of the experiment, deteriorating environmental
conditions caused senescence and sloughing. Comparison of grazed and ungrazed treatments
confirms the importance of grazing in depressing periphyton biomass even with nutrient enrichment
The fit to the Spring, 1990, channel enclosure, N- and P-enriched experiment with about 90%
grazers removed is equally as good, with a similar prediction of greater biomass buildup and
sloughing than can be confirmed with the data (Figure 12). Simulation of the comparable channel
experiment in Summer, 1989, with severe light limitation, did not exhibit the rapid buildup
suggested by the observations in Week 1 , but statistically it was a good fit to the observed mean and
variance (Figure 13). However, the model indicates that the growth is almost entirely diatoms,
while Rosemond (1993) showed that much of the buildup is green algae.
°
m
- D iato ms (g/m2 |
Oth alg (g/m2 )
Periphyton X Observed
Figure 12. 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.
14
-------
MODEL VALIDATION REPORTS
ADDENDUM
D Photosynthesis •Respiration Q Bscretion QMortality • Grazing QSloughing
Figure 13: 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.
•Periphyton
X Observed
•Periphyton-grazed
Observed-grazed
Figure 14. Grazed and ungrazed predicted and observed biomasses of periphyton in Walker
Branch, Tennessee, with addition of N and P in Spring, 1989. This figure combines Figure 3
and Figure 9 for purposes of comparison.
15
-------
MODEL VALIDATION REPORTS
ADDENDUM
.2
m
- D iato ms (g/m2 I
Oth alg (g/m2 )
Periphyton X Observed
Figure 15. 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 90% of grazers in Spring, 1990.
I
- Diatoms(g/m2) —•— Oth alg(g/m2) ^^Periphyton X Observed
Figure 16. 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 but with low light in Summer, 1989.
16
-------
MODEL VALIDATION REPORTS
ADDENDUM
Fits to data from enrichment studies involving the addition of either N or P, but not both, and with
removal of grazers are interesting because both the observed data and the simulations suggest that
addition of either nutrient led to moderate stimulation (Figure 14 and Figure 15). Simulations of
both the N-enrichment treatment and the P-enrichment seem to fit to the observed data well. With
grazers included, the simulations bound the higher observed results but do not exhibit the 0.1-0.2
g/m2 fluctuations that were observed temporally and within samples (Figure 16 and Figure 17).
Perhaps grazing is variable from one tile to another—a detail that is not simulated by a spatially
aggregated model such as AQUATOX.
°
m
3.5
2.5
1.5
0.5
\N" tf
^ ^ ^ ^
-Diatoms
-Others
Periphyton X Observed
Figure 17. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of N and
removal of grazers in Spring, 1989
17
-------
MODEL VALIDATION REPORTS
ADDENDUM
3.5
.2
m
- D iato ms (g/m2 | —•—Oth alg (g/m2 ) Periphyton X Observed
Figure 18. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of P and
removal of grazers in Spring, 1989.
0.6
0.5
0.4
0.2
0.1
D iato m s — • — Others ^^^Pe riphyton X Observed
Figure 19. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of N in
Spring, 1989
18
-------
MODEL VALIDATION REPORTS
ADDENDUM
.2
m
0.6
0.5
0.4
0.3
0.2
0.1
X
A A A
r r r ..°r r
/ / / / /
Diatoms
Others
Periphyton X Observed
Figure 20. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of P in
Spring, 1989.
The simulation of the control channel, with grazers removed, seems to fit the exponential buildup
of biomass in the first several weeks in the Spring, 1989 experiment (Figure 18). However, the
timing of the apparent sloughing is off and the maximum value cannot be confirmed by the data
(Figure 18). Nevertheless, it is instructive to compare the results from this experiment with those
of the companion experiment with addition ofN andP; without the addition of nutrients periphyton
growth is much slower and senescence and sloughing occurs at a lower biomass (Figure 19).
The exponential buildup of periphyton biomass in the Spring, 1990, experiment seems to fit the
observed data well (Figure 20). Statistically, the simulation of the summer experiment is similar
to the observed (Table 5); however, the data do not support the predicted simple exponential
increase in biomass over the seven weeks (Figure 21); possibly light is more limiting for part of the
experiment than represented by the model.
Similar to the simulations of the grazed nutrient enrichment experiments, the predicted biomasses
in the control channels with grazers provide mean responses that are similar to the observations but
do not represent the 0.1 to 0.2 g/m2 temporal and within-sample fluctuations (Figure 22 - Figure
24).
19
-------
MODEL VALIDATION REPORTS
ADDENDUM
2.5
1.5
0.5
-Diatoms(g/m2)
-Oth alg(g/m2)
Periphyton X Observed
Figure 21. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993, pers. comm.) in control channel in Walker Branch,
Tennessee, with removal of grazers in Spring, 1989.
Periphyton-N&P X Observed-N&P
•Periphyton-Ctl
» Observed-Ctl
Figure 22. Control and nutrient-enriched predicted and observed periphyton biomass for Walker
Branch, Tennessee, without grazers in Spring, 1989. This figure combines Figure 9 and Figure
18 for purposes of comparison.
20
-------
MODEL VALIDATION REPORTS
ADDENDUM
.2
m
-D iatoms
-Others
Periphyton X Observed
Figure 23. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in control channel in Walker Branch, Tennessee, with
removal of 90% of grazers in Spring, 1990.
o
in
-Diatoms —"—Others Periphyton X Observed
Figure 24 Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in control channel in Walker Branch, Tennessee, with
removal of grazers in Summer, 1989.
21
-------
MODEL VALIDATION REPORTS
ADDENDUM
.2
m
0.6
0.5
0.4
0.3
0.2
0.1
t
Diatoms
Others
Periphyton X Observed
Figure 25. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in control channel in Spring, 1989.
0.8
0.7
0.6
Diatoms
Others
Periphyton X Observed
Figure 26. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in control channel in Walker Branch, Tennessee, in
Spring, 1990.
22
-------
MODEL VALIDATION REPORTS
ADDENDUM
.2
m
0.8
0.7
0.6
0.5
0.3
0.2
0.1
,-? ,-? ,-? ,-? ,-? ,-? >•? A*
^ ^ ,/ ./ ^ ./• ^ #
Diatoms
Others Periphyton X Observed
Figure 27. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in control channel in Walker Branch, Tennessee, in
Summer, 1989.
Simulations of experiments with high light levels and no grazing in Summer, 1989, are difficult to
evaluate because the predicted dynamic responses, with buildup of biomass followed by sloughing,
cannot be confirmed by the data. In fact, statistically the predicted and observed distributions for
the unenriched experiment are quite different (Table 5). The predicted patterns are intriguing in that
they predict stimulation of algal growth and sloughing both without (Figure 25) and with (Figure
26) addition of nutrients and in the absence of grazers. The addition of nutrients decreases the
predicted doubling time from 6.7 days to 5.7 days. This indicates that light and not nutrients are
limiting in the summer. The nitrate and phosphate concentrations in the stream are twice those of
the spring experiments, probably because the nutrients are not being assimilated as quickly by
periphyton in the heavily shaded stream in the summer. Unfortunately, carbon fixation rates
observed by Rosemond (1993, pers. comm.) at the beginning and end of the experiment do not
confirm high productivity in the unenriched experiment. However, based on the increase in observed
biomass from the initial value to that eight days later (Figure 25), a doubling time of as little as 3
days is possible. As shown in Figure 27, the model predicts a substantial algal response to high
light, especially when compared to the comparable treatment with natural, low light.
23
-------
MODEL VALIDATION REPORTS
ADDENDUM
- Diatoms(g/m2)
Oth alg(g/m2) ^
X Observed
Figure 28. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, channel without
addition of nutrients but with high light levels and removal of grazers in Summer, 1989.
-Diatoms(g/m2) —•— Oth alg(g/m2) ^^Periphyton X Observed
Figure 29. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of N and
P, high light levels, and removal of grazers, Summer, 1989.
24
-------
MODEL VALIDATION REPORTS
ADDENDUM
Periphyton-High Lt X Observed-High Lt ^—Periphyton-Low Lt —•— Observed-Low Lt
Figure 30. Low-light and high-light predicted and observed biomass of periphyton in Walker
Branch, Tennessee, with addition of N and P and removal of grazers in Summer, 1989. This
figure combines Figure 13 and Figure 26 for purposes of comparison.
With grazers present the results are quite different in the high-light experiments. Without the
addition of nutrients the grazers are able to consume most of the algal production (Figure 28); the
model gives an acceptable fit to all the observed data by passing within one standard deviation of
all the points. However, the model predicts a slight increase in biomass of greens, in contrast to
Rosemond's (1993) observation that greens decreased due to the competitive advantage of diatoms.
With the addition of nutrients the grazers are unable to keep pace with algal production in both the
simulation and in the experiment (as Rosemond, 1993, concluded), and the fit of the prediction to
the observed data point on the last day is almost exact, although the model overestimates biomass
in the intervening weeks (Figure 29).
25
-------
MODEL VALIDATION REPORTS
ADDENDUM
0.7
0.6
0.5
0.4
.2
m
0.3
-Diatoms — • — Others
e riphyton X Observed
Figure 31. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, channel without
addition of nutrients but with high light levels, Summer, 1989.
0.9
0.:
E
u
I
-Diatoms —"—Others Periphyton X Observed
Figure 32. Comparison of predicted biomass of periphyton, constituent algae, and observed
biomass of periphyton (Rosemond, 1993) in Walker Branch, Tennessee, with addition of N and P
and high light levels, Summer, 1989.
26
-------
MODEL VALIDATION REPORTS ADDENDUM
Process-level Verification
Walker Branch is a simple spring-fed woodland stream ecosystem with periphyton, grazers (almost
entirely one species of gastropod), detrital loadings from upstream, low nutrient levels, spring-
moderated temperatures, and low light levels. Through careful factorial experiments conducted in
stream-side channels and stream enclosures, nutrients, light, and grazing were manipulated in Spring
and Summer (Rosemond, 1993). Twenty control and perturbed treatments form the basis for
calibration of the periphyton submodel in AQUATOX. The ability of one set of parameters to
represent all twenty treatments is evidence of the robustness of the model for representing woodland
streams. The fact that the model demonstrates bottom-up and top-down controls on the periphyton
with multiple limiting factors suggests that there is a sound scientific basis for the formulations.
Rosemond (1993) came to the conclusion that
"the biomass, productivity, and taxonomy of the periphyton community in this study
were strongly limited by nutrients and light and controlled by herbivory. The
addition of nutrients and light and the removal of snails resulted in a shift in the algal
community from one composed of chlorophytes and cyanophytes, and characterized
by low biomass... to a community of filamentous diatoms, with high biomass....
Removal of herbivores, alone, resulted in taxonomic shifts towards diatoms, but with
little increase in biomass or productivity. When nutrients were added alone, biomass
and productivity were not strongly affected. However, it was the interaction between
all three factors that produced the greatest effects on biomass and productivity,
indicating simultaneous limitation by light, nutrients, and herbivory."
The simulations presented in the previous section provide a rigorous, mathematically coherent
confirmation of most of Rosemond's (1993) conclusions.
Grazing by large, long-lived populations of gastropods was shown to suppress periphyton biomass
under all circumstances except the combination of high nutrients and high light. Both N and P,
singly and especially in combination, were shown to stimulate periphyton growth, resulting in either
buildup of periphyton biomass (to higher levels than observed by Rosemond, 1993) or trophic
transfer to gastropods. Periphyton growth can occur under the low light conditions provided by the
dense forest canopy; however, elevated light levels, approaching those of an unshaded stream, were
shown to stimulate growth, especially if coupled with enriched nutrient conditions.
As periphyton biomass, including associated detritus, accumulates with low or no grazing it is
vulnerable to sloughing. AQUATOX is formulated to represent sloughing as a function of biomass
and deteriorating environmental conditions including decreasing nutrients and decreasing light, the
latter as a function of both incident solar radiation and self-shading. Generally the predicted buildup
of biomass and sloughing were shown to be consistent with the observed data and the experience of
the field investigator (Rosemond, pers. comm.). Sloughing is also a function of water velocity,
which could not be invest!gated using the experimental data with near-constant velocities. However,
in the validation exercise described in the next section, daily varying discharge was a factor in a
simulation of periphyton sloughing from natural cobbles in the stream.
27
-------
MODEL VALIDATION REPORTS
ADDENDUM
Validation
A validation for a woodland stream was performed using a two-year bimonthly data set from Walker
Branch, Tennessee (Rosemond, 1993). Periphyton biomass and gastropod densities were measured
on cobbles in the stream channel. Because it involved continuous, ambient stream conditions (Table
6), in contrast to the seven-week stream-side and channel enclosure studies used for calibration, a
successful simulation can be considered a validation of the periphyton submodel for this system.
Table 6. Loadings and Initial Conditions Used in 2-year Simulations of Walker Br.
October 1, 1988 to November 1, 1990
Initial gastropod biomass
(g/m2)*
Nutrients (mg/L)
Light (Ly/d)+
Velocity (cm/s)
Initial algal biomass (g/m2)*
Treatment
NH4-N
NO3-N
PO4-P
Diatoms
Greens
Constant Flow
8
0.0001 to 0.0065
0.0009 to 0.0435
0.0008 to 0.0053
8.4 to 115
13.0
0.022
0.023
Daily Flow
8
0.0001 to 0.0065
0.0009 to 0.0435
0.0008 to 0.0053
8.4 to 115
7.0 to 23.0
0.022
0.023
+ Light was reported by Rosemond (1993) as photosynthetic active radiation, which was multiplied
by 2 to provide total radiation values used by AQUATOX * Ash-free dry weight
First, the model was run for the two-year period using a constant flow rate and a velocity of 13 cm/s.
This corresponded closely to the conditions under which the model was calibrated except that the
velocity was greater than in the experiments. The predictions fell below all the observed points, but
passed within one standard deviation of all but one point (?). Mean predicted biomass was 0.280
g/m2 compared to observed biomass of 0.413 g/m2, with rB = -0.148 and F = 0.589, indicating that
the means and variances are similar at the 95% confidence level. The gastropod biomass increased
to 15 g/m2, which probably is above the maximum biomass in the stream.
Then the model was run using observed daily flow rates. The results are somewhat different in this
simulation that more accurately represents the dynamics of the natural stream system (?). The
predictions pass within one standard deviation of all thirteen points, passing very close to some of
the observations. The mean predicted periphyton biomass was 0.328 g/m2, compared to the
observed biomass of 0.413 g/m2, with an rB of-0.942; and the F value is 1.13, confirming that the
predicted and observed distributions are similar at the 95% confidence level.
28
-------
MODEL VALIDATION REPORTS
ADDENDUM
1.4
1.2
°
m
Periphyton XC Obs —*—Gastropods
Figure 33 Comparison of predicted and observed (Rosemond, 1993) biomass of periphyton and
predicted biomass of gastropods on cobbles in Walker Branch, Tennessee, with constant flow in
1989 and 1990.
o
m
Periphyton X Obs —*— Gastropods
Figure 34. Comparison of predicted and observed (Rosemond, 1993) biomass of periphyton and
predicted biomass of gastropods in Walker Branch, Tennessee, with daily varying flow rates in
1989 and 1990.
29
-------
MODEL VALIDATION REPORTS
ADDENDUM
After a decrease in the biomass of greens at the beginning of the simulation, perhaps due to an
inappropriate initial condition, the fluctuations in periphyton are due to diatoms (Figure 32). Long
periods with only base flow allow predicted diatom biomass to accumulate until finally a small
runoff event initiates massive sloughing. Gastropods exhibit a stable fluctuation in predicted
biomass between 8 and 9 g/m2, and grazing pressure on diatoms is intense when diatoms reach a
biomass of 0.20 mg/cm2, the minimum biomass for feeding by gastropods (Figure 33). Loss due
to photorespiration also increases with biomass buildup due to self-shading.
0.45
0.35
I
r 0.25
0.15
0.05
•Diatoms Others Inflow
Figure 35. Predicted biomass of periphytic diatoms and other algae on cobbles in Walker
Branch, Tennessee, with daily varying inflow in 1989 and 1990.
Conclusions
The periphyton submodel, with the newly formulated sloughing function, provides enhanced
capability for simulating periphyton growth as a result of eutrophication or nutrient enrichment. The
weight of evidence based on predicted and observed maximum and mean periphyton biomass,
biomass variance, visual inspection, and process rate plots, strongly suggests that the model provides
a good fit to the periphyton data in twenty treatments. The model also satisfactorily simulated
gastropod biomass. Using a single parameter set, the model performed well in simulating nutrient-
poor and nutrient-enriched, low- and high-light conditions, and grazing and lack of grazing in a
woodland stream. An independent, two-year data set from Walker Branch was used to validate the
model for continuous ambient woodland stream conditions, including low light, low nutrients,
variable temperature, intense grazing, and daily varying stream flow. Visual inspection, statistics,
and consideration of process plots all confirm that the model provides a realistic representation of
the woodland stream ecosystem.
30
-------
MODEL VALIDATION REPORTS
ADDENDUM
1 Photosynthesis •Respiration D Excretion D Mortality • Grazing • Sloughing
Figure 36. Predicted process rates affecting diatoms on cobbles in Walker Branch, Tennessee,
with daily inflow in 1989 and 1990.
Bibliography
Asaeda, T., and D.H. Son. 2000. Spatial structure and populations of a periphyton community: a
model and verification. Ecological Modelling, 133:195-207.
Bartell, S.M., R.H. Gardner, and R.V. O'Neill. 1992. Ecological Risk Estimation. Boca Raton,
Florida: Lewis Publishers, 252 pp.
Biggs, B.J.F. 2000. Eutrophication of streams and rivers: dissolved nutrient-chlorophyll
relationships for benthic algae. Jour. North American Benthological Society, 19(1): 17-31.
Bothwell, M.L. 1985. Phosphorus limitation of lotic periphyton growth rates: an intersite
comparison using continuous-flow troughs (Thompson River system, British Columbia).
Limnol. Oceanogr., 30(3):527-542.
Hart, D.D., and C.T. Robinson. 1990. Resource limitation in a stream community: phosphorus
enrichment effects on periphyton and grazers. Ecology, 71(4): 1494-1502.
Hill, W.R. 1992. Food Limitation and Interspecific Competition in Snail-Dominated Streams.
Canadian Jour. Fish. Aquat. Sci., 49:1257-1267.
31
-------
MODEL VALIDATION REPORTS ADDENDUM
Marcus, M.D. 1980. Periphytic community response to chronic nutrient enrichment by a reservoir
discharge. Ecology, 61(2):387-399.
Mulholland, P. J., andA.D.Rosemond. 1992. Periphyton response to longitudinal nutrient depletion
in a woodland stream: evidence of upstream-downstream linkage. Jour. North American
Benthological Society, 11 (4):405-419.
Rosemond, A.D. 1993. Seasonally and Control of Stream Periphyton: Effects of Nutrients, Light,
and Herbivores. Dissertation, Vanderbilt University, Nashville, Tenn., 185 pp.
Rosemond, A.D. 1994. Multiple Factors Limit Seasonal Variation in Periphyton in a Forest Stream.
Jour. North American Benthological Society, 13(3):333-344.
Rosemond, A.D. 1995. Indirect Effects of Herbivores Modify Predicted Effects of Resources and
Consumption on Plant Biomass. In G.A. Polis and K.O. Winemiller (Eds.) Food Webs:
Integration of Patterns and Dynamics. New York: Chapman and Hall, pp. 149-159.
Rosemond, A.D., and S.H. Brawley. 1996. Species-specific Characteristics Explain the Persistence
of Stigeoclonium tenue (Chlorophyta) in a Woodland Stream. Jour. Phycology, 32(1): 54-63.
Rosemond, A.D., PJ. Mulholland, and S.H. Brawley. 2000. Seasonally shifting limitation of stream
periphyton: response of algal populations and assemblage biomass and productivity to
variation in light, nutrients, and herbivores. Canadian Jour. Fish. Aquat. Sci., 57:66-75.
Stevenson, RJ. 1996. An introduction to algal ecology in freshwater benthic habitats. In RJ.
Stevenson, M.L. Bothwell, and R.L. Lowe (eds.), Algal Ecology: Freshwater Benthic
Ecosystems, San Diego, Academic Press, pp. 3-20.
Van Nieuwenhuyse, E.E., and J.R. Jones. 1996. Phosphorus-chlorophyll relationship in temperate
streams and its variation with stream catchment area. Canadian Jour. Fish. Aquatic Sci.,
53:99-105.
Wang, H., J.A. Kostel, A.L. St Amand, and K.A. Gray. 1999. The response of a laboratory stream
system to PCB exposure: study of periphytic and sediment accumulation patterns. Water
Research, 33(18):3749-3761.
32
-------
MODEL VALIDATION REPORTS
ADDENDUM
Appendix—Biotic Parameters
Periphyton, Diatoms
&OU&TOK!- Edit Want
Load From Lib. |E Save to Library Q; OK ;|E Cancel E; Print
Plant Periphyton, Diatoms
Plant Type: iPeriphyton
ToKicity Record: Diatoms
Plant Data:
Deferences:
Saturating Light 64 Ly/ti
P Ha If-saturation f 0.0012 mg/L
N Half-saturation 0.013 mg/L
Inorg. C Ha If-saturation 0.054 mg/L
Temp. Response Slope r^^^^^j^jj
(Hill 1996;Goidsborough & Robinson 1996
ioriSiardf,
DeNicola, 1996
Optimum Temperature
Maximum Temperature
Min Adaptation Temp
Max Photosynthetic Rate
Respiration Coefficient
14 °c
2 "c
OJ 1/d
O.OS 1/d
Home and Goldman, 1394, p. 140 - 0.0?
(greens)
adapted to cold conditions
Asaeda and Son, 2000 (tl.6)
Mortality Coefficient
Exponential Mort Coeff
P : Photosynthate
N : Photosynthate
Light Extinction 1.2 1/rn
0.001 ftac i d jnrof. judgment
0.1 maif/cf 110%/dif phow^nthesis-0
0.018 ratio
0.079 ratio
jRadfleld et al., S3 stoichiometry
based on max. Womass of ZOO
Phytoplankton Only:
C&W83
Carrying Capacity
Reduction in Still Water
Periphyton and Macrophytes Only:
' M B/m2 |ESF; Colby &Mclntire 78 = 80
1 fraction iColby « Bcintire 78
33
-------
MODEL VALIDATION REPORTS
ADDENDUM
Periphyton, Stigeoclonium tenue
oad From Lib. :::: Save to Library l
Plant Stigeoclonium, peri.
Plant Type. Periphyton
Tox city Record: Greens
Saturating Light
P Half-saturation
N Ha If- saturation
inorg. C Half-saturation
Temp. Response Slope
Optimum Temperature
Maximum Temperature
Min Adaptation Temp.
Max. Photo synthetic Rate
Respiration Coefficient
Mortality Coefficient
Exponential Mort Coeff
P Photo&yntha e
W : Photosynthate
Light Extinction
Plant Data:
133 UtH
fl.0093 mg/L
8.05
0.05* mfl/L
2
25 °c
|Asaeda 4 Son 20flO,HiIM996, 139; G » F
|" , p. 39 - 0.054
defaiiit
15 "c
0.3 1 / d
4.03 1/d
OJI01 frac / d
0.01 masr/ d
63ll ratio
b°3fl ratio
jBorchardt 1996, p. 211 0.0)
[C* WB3
iprof. judgment
iprof. judgment, 1%/d if photosyn = 0
[RedflairteTaiT^
1.2 1 im
based on max. biomass ef 280
Phytoplankton Only:
Periphyton and Macrophytes Only:
Carrying Capacity
Reduction in Still Water
5 aim2
1 fraction
Colby ft Mcintire 78 (80)
Colby & Mchitire 78
34
-------
MODEL VALIDATION REPORTS
ADDENDUM
Gastropod
Load from Lib. d Save to Libraiy ~|[ fiK_ ]|:::: Cancel h~ Print
Animal
Gastropod
Animal Type: JBenthic Invert, jj Toxicity Record: JOstraeod
Animal Data:
References:
Half Saturation Feeding • 0.1
Maximum Consumption j 0.85
Min Prey for Feeding j 0.4
Temp. Response Slope } 1,4
Optimum Temperature I 20
Maximum Temperature
Min Adaptation Temp.
Respiration Rate
Specific Dynamic Action
Excretion: Respiralion
Gametes : Ekomass
Gamete Mortality
Mortality Coefficient !
Carrying Capacity |
mg/L
g/g-d
mg/L
ifirof. judgement
= 0.025 g/m2 (ftldntira et ai. 1996 = 0.7)
38
5
O.OB2
0.2:5
oTiF
0.1
0.25
O.OHI1B9
200
°c
°c
I/a
(unitle£
ratio
ratio
I/O
I/a
mg/L
Mclnlire & Colby p. 172 (D.05)
McM Jhon In Thorp & Covich 1991, p. 321
prof, judyment
no, Leidy & Pioskey 1^0
Iprof. judgment
,___^_™^^
|about35% in marine snail (wvyw.szn.it)
jdefauit (Scavla and Park)
jprof. judgment
jHill, 1992
lobs. Walker Branch 00 g/m2)
Trophic Interactions:
Preference: Egestion: References:
(mlio) (fraction)
Sad. Refraciory Deiritus
Serf. Labile Detritus
Participate Refrac. Detritus
Particulats Labile Detrity$
Diaioms
Slue-Greens
Greens
Macrophytes
Detritivorous Invertebrates
Herbivorous Invertebrates
Predatory Invertebrates
Forage Fish
Bottom Fish
Small Game Fish
j O.OS
! OJ5
j 0
! o
j OJ5
i B.1
j OS
I OJ5
j 0
j 0
j 0
I 0
j 0
I 0
1
0.62
0
0
0.56
0.8
8.8
0.99
0
0
0
0
0
0
assim: L&P CS (0.56)
Bioaccumulation Data:
Mean age or lifetime
Initial fraction that is Lipid
(Wet Wt.)
Mean weight
720 days
2 years
default
0.33
calc. from Thorp and Cavich 1991
35
------- |