PB8B-13J559
EPA/600/3-87/038
December 1987
FGETS (FOOD AND GILL EXCHANCE OP TOXIC SUBSTANCES):
A Simulation Model for Predicting Bioaccumulation
of Nonpolar Organic Pollutants by Fish
by
M.C. Barber, L.A. Suarez, and R.R. Lassicer
Biology Branch
Environmental Research Laboratory
Athens, CA 30613
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GA 30613
-------
DISCLAIMER
information in this document has been funded wholly or in part by the
United State* Environmental Protection Agency. It has been subject to the
Agency's peer and administrative review, and it has been approved for publica-
tion as an EPA document* Mention of trade names or commercial products does
not constitute endorsement or recommendation for use by the U.S. Environmental
Protection Agency.
ii
-------
FOREWORD
Ab environmental controls become more expensive and penalties for judgment
errors become more severe/ environmental management requires more precise
asHessment tools based on greater knowledge of relevant phenomena. As part of
this Laboratory's research on the occurrence, movement, transformation, impact,
and control of environmental contaminants, tho Biology Branch conducts research
to predict the rate, extent, and products of biological processes that control
pollutant fate in soil and water and develops methods for forecasting ecosystem
level effects suitable for exposure and risk assessment.
The Athens Environmental Research Laboratory, along with Office of Research
and Development laboratories in Corvallis, OR, Duluth, MN, and Gulf Breeze, FL,
is developing a system to assess ecological risks ftom exposure to environmental
toxicants. This system will provide the capability to assess risk associated
with different uses of chemicals resulting from various options for regulating
pesticides and toxic chemicals to protect organisms in their natural environment.
This report describes a component of that system, the Food and Gill Exchange of
Toxic Substances (FGETS) model.
The authors wish to acknowledge Drs. Gilman Veith and James McKim of Duluth
ERL whose personal communication!! and research regarding fish toxicity and toxi-
cokinetics motivated this research. The authors also acknowledge the invaluable
assistance of Dr. Samuel Karickhot'f of Athens ERL in the development of the model
presented herein.
Rosemarie C. Ruaso
Director
Environmental Research Laboratory
Athens, GA
iii
-------
ABSTRACT
A model for the bioaccumulatlon of nonpolar, nonmetabolized organic
chemicals by fish is described. This model, FCETS, simulates thermodynam-
lcally driven chemical exchange by fish assuming either aqueous exposure
only or Joint aqueous and food chain exposure. Parameterization of the
model incorporates allometrlc relationships between the fish's body weight
and Its gill and intestinal surface areas* lipid content of the fish, and
physico-chemical properties of the chemical (i.e., molecular weight,
melting point, and n-octanol/water partition coefficient). The model is
validated by comparing predicted and observed depuration rates of organic
chemicals by rainbow trout (Salmo galrdnerl). An application of the
model describing the bloaccumulation of polychlorlnated blphenyls by Lake
Michigan lake trout (SalvellnuB namaycush) also is presented.
This report covers a period from October 1, 1986, to September 30,
1987, and work was completed as of September 30, 1987.
lv
-------
I. INTRODUCTION
When Aquatic ecosystem are polluted with organic chemicals, fish in those
system will bioaccumulate the xenoblotics directly both fro* the water and
from prey that have become contaminated with the chemicals. For benthic species,
chemicals also htay be accumulated by dermal contact with contaminated sediments.
If these chemicals are not metabolized, their ultimate concentrations in fish
should be predictable based on principles of thermodynamic partitioning, The
purpose of this work is to present a dynamic model, FGETS (Food and Gill Exchange
of Toxic Substances), that describes thermodynamically driven bioaccumulation
of nonmetaholized organic toxicants by fish. This work is an extension of a
previously published model (Barber et al. 1987, Suarez et al. 1987} which
describes the uptake and depuration of organic toxicants across fish gills.
FGETS is a FORTRAN simulation model that predicts temporal dynamics of a
fish's whole body concentration* Ii.e., ppm - microgram ot chemical por gram
(live weight fish)] of a nonmetabolized, organic chemical that is bioaccumulated
either from water only, which is probably the predominant exchange route for
acute exposures, or from water and food jointly, which is more characteristic
of chronic exposure. These dynamics are calculated algebraically from the
fish's predicted total body burden and live weight. Additionally, FGETS cal-
culates time to death assuminq that the chemical's mode of action is simple
narcosis.
II. MODEL FORMULATION
The following discussion is a brief overview of the formulations used in
FGETS to predict a fish's ~¦otal body burden, Bf (Bf - micrograms of chemical
per fish), live body weight, W (W - grams of live weight), and time to death.
1
-------
II.1. Modeling Bf As suiting Only Water Exposure
Because the exchange of nonpolar organic chealcals across the gills of
fish apparently occurs by simple diffutlon (Opperhulzen et al. 1985, Oppezhuizen
and Scharp 1987), Pick's first law of diffusion can be used to Model this
process. Several models of diffusive exchange in fish have been proposed
(e.g., Yalkowsky et al. 1973, Hackay 1982, Mackay and Hughes 1984, Gobas et al.
1986, Gobas and Hackay 1987). Barber et al. 1987, however, nodeled the total
body burden of nonmetabolized, organic chemicals in fish starting with the
differential equation,
dBf/dt ¦ Sg Jg ¦ Sg ky (Cy " Cg) (1}
where Sg is the fish's total gill area (cm^), Jg is the net diffusive flux
(microgram/cm /day) acrosB the fish's gills, kw is the chemical's mass conduc-
tance (cm/day) through the inter lame liar water of the gills, and cw and Ca are
the chemical's concentration (ppm) in the environmental vater and the fish's
aqueous blocd, respectively. Kote that while the total tesistar.ee (i.e., the
reciprocal for conductance) to mass exchange across gills actually depends on
individual resistances in both the interlamellar water and the gill epithelium,
Equation 1 assumes that all resistance to exchange of a nonpolar organic is in
the water. For chemicals that have n-octanol/water partition coefficients,
K^, on the order of log Kqw>2' however, this assumption is reasonable (see
Barber et al. 1987).
To apply Equation 1, however, a functional relationship between the fish's
total body burden, Bf, and aqueous blood concentration, Ca, must be specified.
To this end, Barber et al. (1987) noted that a fish's whole body concentration,
Cf - Bf / w, could be expressed as
Cf « Bf / W - Pa Ca + PL Ci + Ps CB, (2)
where Pa, P\, and PB are the fractions of the whole fish that are water, lipid,
2
-------
and structural organic material, respectively, and and CB are the chaaical'a
concentrations (ppm) in the fish's lipid and structural organics, respectively.
Since the depuration rates of organic chemicals from different tissues within a
fish apparently do not differ significantly (Grzenda et al. 1970, van Veld et
al. 19B4, Branson et al. 1965, Norheim and Roald 1985), internal equilibration
between the aqueous, lipid, and structural organic phases can be assumed to be
rapid in comparison with exchange across the gills. Consequently, Equation 2
can be rewritten as
cf " Bf / W « (Pa + PX + PB Kg) Ca (3)
where Kj and Xs are thermodynamic partition coefficients between lipid and water
and between organic carbon and water, respectively. Because Cw equals Ca at
equilibrium (see Equation 1), it follows from Equation 3 that a fish's ultimate
biomagnification factor, BCF ¦ Cf/Cv, assuming only gill exchange, is simply
BCF » (~'r»1 Kj + PH Xh) (4)
Therefore, using Equations 3 and 4, Equation 1 can be rewritten as
dBf/dt - Sg kw (Cw - Cf/BCF) - Sg kw (Cw - Bf/(W BCF)) (5)
11.2. Modeling Bf Assuming Food and Water Exposure
To model the dynamics of a fish's total body burden that is accumulated
from both water and food, Equation 5 simply needs to be modified as
dBf/dt - Sg kw (Cw - Cf/BCF) ~ (6)
where is the net mass exchange (micrograms/day) across the fish's intestine
from food. Although this modification is straightforward, the selection of a
formulation to model is not. Consequently, FGETS has three different
optional formulations for this flux.
3
-------
The crudest formulation of iiiumi that fish can assimilate a constant
fraction of the chemical it ingests, i.e.,
^i -H Cp P (7)
where p is an assimilation efficiency (dimensionless) for the chemical, Cp is
the chemical's concentration (microgram/gram prey) in the ingested prey, and
F is the fish's daily feeding flux (gram prey/day). Since the exchange of
nonpolar organic chemicals across the intestines of fish also is driven by
diffusive gradients (Vetter et al. 1965), the assimilation efficiency, H ,
should be a decreasing function of the fish** total body concentration, Cf.
Consequently, Equation 7 generally will overestimate a fish's body burden.
FGETS allows for this nonthermodynamic formulation, however, because 1) food
exposure has been previously modeled in this way (Norstrom et al. 1976, Jensen
et al. 1982, Thomann and Connolly 1984), and 2) when U - 1.0, i.e., the fish
is assumed to Aosimilate all the ingested chemical, an absolute upper bound to
bioaccumulation can be estimated. Importantly, however, this upper bound will
exceed the thermodynamic limit to bioaccumulation.
A second candidate formulation for is
Ji - Cp P - Ce E (8)
where Ce is the chemical's concentration (microgram/gram) in the fish's daily
egestive/fecal flux, E (gram/day)* If the transit time through the gastrointes~
tinal tract is relatively slow, it might be reasonable to assume that the con-
centration of the chemical in the fish'B aqueous blood, intestinal fluids, and
fecal matter have equilibrated wi*Ji one another. In this case,
Cg ¦ c ^oc La
where Poc is the fraction of the fish's feces that is organic carbon and
is a thermodynamic partition coefficient between water and refractory organic
4
-------
matter (¦•• for example Karickhoff 1981, Bri99s 1981, and Chiou et al. 1986).
Therefore, using Equations 3, 4, and 9, Equation 8 can be rewritten as
- Cp P - (Poc ^ Cf/BCP) E (10)
If on the other hand the feces do not equilibrate with the fish's aqueous
blood, then should be modeled kinetically and, in particular,
Ji - Si ki (Cia - Ca) (11)
where Si (cm2) is the surface area of the intestine, k^ is the mass conductance
(cm/day) of the chemical through the intestinal fluids, and is the concen-
tration of the chemical in the intestinal fluids or the aqueous portion of the
food resident in the intestine. To use Equation 11, C^a must be expressed as
a function of the prey's total body concentration, Cp. Stierefore, in a manner
similar to Equations 2 and 3, the total concentration, C^, of chemical in the
intestinal food is formulated as
Ci - »i / 1 - Pla cia ~ rlo cio
where is the total burden (ttlcroqraa) ot chemical in the intestinal fcod, 2
is the wet mass of the intestinal food, P^a and P^0 are the fractions of the
intestinal food that arc water and dry organic matter, respectively, and C^a
and Cio are the chemical's concentration in these two phases. Because enzymatic
and mechanical processes transform food in a fish's intestine into a more or
less mixed suspension, it seems reasonable to assume that C^a and C±0 maybe
equilibrated and hence
Ci - Bi / X -
-------
assimilation, and egestion, respectively, the system of Equation* 6 and
11-14 constitute a kinetic nodal for joint gill and food axchan9« of
organic toxicants by fish.
II.3. Modeling the Growth of Pish
A fish's growth can be modeled straightforwardly by the mass balance
equation,
dW/dt - F - E - R - SDA (J 2)
where F, E, R and SDA are the fish's mass fluxes (grams/day) of feeding,
egestion, routine respiration, and specific dynamic action, respectively.
Traditionally, a fish's maximal feeding flux, P, has been described
empirically by the allometric function,
t2
F - f, W (13)
where the coefficient, t\, is generally a function of water temperature (see
Palohelmo and Dickie 1965). An alternative expression for F, however, is
F - * (Smax - S), (14)
where 9 is the fish's ad libitum feeding rate, SMX is the fish's maximum
stomach/gut capacity, and S is the amount of food resident in the
gut. S, in turn, is modeled by
dS/dt - F - C (15)
92
where G ¦ gj S represents gastric evacuation (Holling 1966, Ware 1972, Jobling
1986). FGET8 allows the user to specify either the allometric type (i.e. Equa-
tion 13) or the Holling type (i.e., Equations 14 and 15) model to simulate fish
ad libitum feeding. Simulating a fish's growth for long time intervals assuming
continuous ad libitum consumption, however, is ecologically unrealistic because
proy availability usually limits feeding to submaximal levels. In general,
therefore, a fish's feeding, f, predicted by either the allometric type or the
6
-------
Holling type models should be adjusted as
r - f() r (16)
"here t() describes the fish's functional response to available prey. Although
f() eventually will ba modeled, presently FCETS assumes f() ¦ 0.5.
The proportion of P that a fish egests or defecates as nonassioilated
matter, E, depends on ration size and quality. Presently, however, FGEl'S
assumes that E is simply iom constant fraction of F, i.e.,
E • (1 -o ) F (17)
where u is an assinilation efficiency.
Like the allometric expression for F, the respiratory expenditure, R, of
fish associated with routine maintenance and activity traditionally has been
described by a power function of the form,
r2
R • ri M (18)
where the coefficient, rj, is y*nerally a function of wat»jr temperature, T.
In particular,
*1 " *"l(T0pt> «*P (T-T0pt)) (19)
where ri(T0pt) is the fish's respiratory coefficient at temperature Topt
(Paloheimo and Dickie 196S, Ursin 1967). On the other hand, the specific
dynamic action of fish, which is the additional metabolic e^ienditure in
excess of R required to aasimilate food (Jobling 1981), often has been
described simply as a constant fraction, <», of the fish's assimilated food,
i .e.,
5DA -oaf (20)
FCETS pressntly assumes that 0 equals 0.2 for all fish (see Stewart et al.
1983, Stewart and Binkowski 1986, and Yarzhombek et al. 1984).
7
-------
II.4. Predicting Narcotic Toxicity in Fish
Although &j«ntitative Structure Activity Response (QSAR) Models that
correlate lethal water concentrations for acute, predetermined periods of
exposure with chemical properties (e.g., Konemann 1981, Veith et al. 1983) are
ultimately required to predict fish mortality due to environmental exposures,
theso regression models have tended to neglect the fact that fish are killed
not by the water concentration of chemicals per se but by the accumulated
internal body concentrations of chemicals that result from such exposures.
Neely (1984), Friant and Henry (1985), McCarty et al. (1985), and McCarty
(1966) have discussed the importance of this distinction.
For chemicals whose mode of action is narcosis, i.o., a nonspecific and
reversible physiological intoxication, the relationship between internal body
concentrations and observed effects is apparently quite simple. In particular,
if two chemicals are narcotics And their thermodynamic chemical activities in
an animal's blood (or any phase when the assumption of internal equilibrium,
i.e., Equation 3, is valid) are equal, then they will produce the same level of
physiological effect (Ferguson 1939, Mullins 1954). Therefore, if LA denotes
chemical activity of a narcotic in a fish's aqueous blood that causes death,
the question we now want to address is what is the relationship between LA and
LC50(t)'s determined by acute toxicity tests of duration t days.
Let LC50(t)'s denote the molar LC50 of a narcotic chemical for an acute
aqueous exposure that lasts for t days and assume that during this exposure
period the biological characteristic of the fish, i.e., its body weight, gill
surface area, lipid content, etc., and the physico-chemical properties of the
chemical, i.e. its log Xw, (see Equations 3-5) are such that the fish's
aqueous blood concentration equilibrates to within at least 99% of the water
8
-------
concentration. If there are now two consecutive water concentrations, NC and
LC, such that at NC no fish are killed whereas at LC all fish are killed/ then
clearly the lethal internal blood concentration and hence LA can be estimated
from some measure of central tendency of these two concentrations• Moreover,
since Stephen (1977) argues that the geometric mean of NC and LC is a valid
estimator of LC50(t), in general LA can be estimated from any calculated LC50(t)
provided the fish essentially can equilibrate with the water and in particular
LA - aw vw LC50(t) (21 )
where av is the activity coefficient of the chemical in water and vw is the
molar volume of water (0.018 1/ mole).
Using data reported by Chiou (1985), we can estimate the aqueous activity
coefficient of an organic chemical, which is the reciprocal of its super cooled
liquid solubility, by the functional regression (see for details Jensen 1986 or
Rayner 198f>),
log(aw) - 1.131 logCKow) + 1.053 (n-37; r-0.98) (22)
This result is consistent with analogous linear regressions between aqueous
solubilities and Kw reported elsewhere, e.g., Chiou and Schmedding (1982)
and Miller et al. (1985), but is significantly different from the regression
log(aw) - 0.944 logtlC^) + 1.422 (n-??» r-0.98) (23)
reported by Yalkowsky et al. (1983). This difference, however, is due to the
fact that Equation 22 uses measured logP's whereas Equation 23 was obtained
using calculated logP's.
Since Equations 21 - 23 predict that
log(LC50(t)) ¦ m logd^) + b, (24)
where -1.131
-------
111. MODEL FAftAKETERXZATXON
Since reliable application of a model depends ultimately on its assumptions
and parameterization, Important assumptions used to parameterize the above
equations follow.
II1.1. Equation 1
Parameterization of Equation 1 depends on the fish's gill morphometry not
only via the total gill surface area, Sg, but also by way of the conductance,
Xw, which depends on both the spacing, d (cm), between lamellae and the mean
length, 1 (cm), of individual lamellae. In general, each of these gill
dimensions is dependent on the fish's body size according to the allometric
functions, S2
Sg - 8! W (25)
d2
d - dj W and (26)
h
1 M, W (27)
Although allometric coefficients and exponents for total gill surface areas
are readily available in the literature and have been tabulated by Hughes
(1972, 1964), de Jager and Dekkers (1976) and in Table I herein, allometric
parameters for interlamellar distances and lamellar lengths seldom are reported.
Fortunately, however, estimates of dj and d2 often can be made using the
relationship,
d - 0.102 p-1.142 (28)
where P is the density of secondary lamellae on one side of the fish's gill
filament (i.e., number of lamellae/mm of gill filament), since many gill
morphometric studies do report parameter values for the allometric function,
P
M - P, M 2 (29)
Equation 28 was exercised for 18 species of fish reported by Saunders (1962)
10
-------
and Hughes (1966) and tabulated values for P1 and p2 are reported by
Hughes (1972, 1984) and in Table 1» Finally, since allometric regressions for
lamellar lengths virtually are unreported in the literature (for exceptions
see Hughes 1984 and Stevens and Lightfoot 1986), a generalized allometric
relationship for this gill dimension can be calculated again using data from
Saunders (1962) and Hughes (1966) as
1 - 0.0187 w0,208 (30)
Since characterization of chemical uptake and excretion across a fish's
gill can be formalized as a problem dealing with mass exchange from laminar
water flow between adjacent secondary lamellae that form 'lamellar channels',
methods that have been used traditionally to analyze convective mass transfer
within arbitrarily shaped channels should provide the means to estimate the
interlamellar conductances, kw. In general, both analytical and numerical
solutions to problems concerning*such transport phenomena have been exprestted
as functions of dimensionless variables known as the Sherwood number, which
can be defined as
NSh ¦ D / (h k) (31)
where D is diffusivity (cm2/sec) of the chemical being transported, k is
the chemical's conductance (cm/sec) to and from the walls of the channel, and
h is the channel's hydraulic radius (cm) (see Kays 1966). Clearly, given any
three quantities in this equation, it is simple to calculate the fourth and, in
fact, this is precisely how the interlamellar conductance can be estimated. To
make such calculations, however, a fixed geometry and set of relevant boundary
conditions for the lamellar channels must be specified. Although lamellar
channels have been considered analogous to rectangular ducts (see Hughes 1966),
because they generally have very high aspect ratios (i.e., mean lamellar height/
interlamellar distance), they can also be considered without loss of generality
11
-------
to be essentially parallel plates. This assumed geometry means not only that
the hydraulic radius of the lamellar channels is simply the interlajnellar dis-
tance, d(cm), but also that Sherwood numbers for lamellar chemical exchange
can be estimated using previously published results for convective mass trans-
fer between permeable flat plates (Colton et al. 1971, Ingham 1984J. Since
chemicals are exchanged across fish gills by a counter current mechanism, a
constant diffusion gradient is probably maintained between the inter lamellar
water and the capillary blood along the length of the secondary lamellae (see
Layton 1987). Consequently the relevant boundary condition for estimating
lamellar Sherwood numbers would be that of constant flux.
Local Sherwood numbers at any distance, z, trom the entrance of a lamellar
channel can be expressed as functions of the dimensionless length,
x(z) - z D / (d2 v), (32)
where v is the mean flow velocity (cm/Bec) between the lamellae, and in
particular
1.1829 x(z)-1/3, if x(z)<0.1
NSh "1/3» <35>
whereas if x(l) > 0.1, then
Nsh - 3.7704 + 0.005232/x(l). (36)
12
-------
In cither case, if the hydrodynaaics of lamellar flow is then Modeled as
Poisuellian slit flow and in particular if
v - (0.5 d)2 A p/(3 M 1) (37)
where A P is the pressure drop across the gill (i.e., approximately 500 dynes/cm^),
and p is the dynamic viscosity of water (i.e., 0.01 poise) (Hughes 1966, Lauder
1984, Stevens and Lightfoot 1986, Barber et al. 1987), then the lamellar
Sherwood number given by Equations 35 and 36 can be expressed an a function of
only chemical diffusivity, interlamellar distance, and lamellar length.
The actual interlamellar conductance for Equation 1 can now be estimated
by substituting either Equation 35 or 36 into Equation 31, setting d equal to
h, and solving for k ¦ kw. Since the units of this conductance will be cm/sec,
it must be time-scaled appropriately for use in Equation 1.
Til.2. Equation 3
Reasonable choices for the partition coefficients, Xj and are assumed
to be the triolein/w&ter partition coefficient, Ktw, and a organic carbon/water
partition coefficient, Koc, respectively. Since both Ktw and Koc are collinear
with the n-octanol/water partition coefficient, Kow, however, Equation 3 is
effectively parameterized by according to the relationships,
K1 " Ktw * 1,44 Kow and (38)
*s - Koc " °'40 Kow <39)
(Karickhoff 1981, Patton et al. 1984, Chiou 1985). If Koc is equated to Ks,
Equation 3 must be modified slightly* i»e.r
Cf - Bf / W - (Pa + PX Kj. + P8 Poc Koc> ca' <40)
where P - is the fraction of the structural organic matter that is organic
OC
carbon. FGETS assumes that Poc equals 0.55.
13
-------
Because a fish's lipid fraction, P^, is generally a user input to rGETS
(see Section IV.1 regarding "pifish" bel ow), it is important for the parameter-
ization of Equations 3 and 4 to know empirical or functional relationships
between P^ and the fish's aqueous and structural fractions, Pa and Pa, since P}
Is being treated essentially as an independent variable. In general, the empir-
ical relation between Pj and Pa can be described adequately by
^a ¦ a0 " a1 P1 ^'
where ag and a^ are positive constants (Eschmeyer and Philips 1965, Love 1970,
Elliott 1976, Craig 1977, Shubina and Rychagova 1982, Beamish and Legrow 1983,
Gill and Weatherley 1983, Weatherley and Gill 1983). Since of the sua of Pa,
Pj, and P„ must be unity, this relationship consequently demands that
P. - (1 - a0) - <1 - a,) Plt (42)
For salnonid fishes, means for ag and aj a»e approximately 0.85 and 1.5,
respectively, and presently the«0 values are used to parameterize FGETS.
Consequently, using these assumed values of ag and aj and Equations 38 and
39 FXSETS estimates a fish's bioconcentration factor (approximately) as
BCF - (1.33 PI ~ 0.033) (43)
IIZ.3. Rquations 11 and 12
A fish's total intestinal surface area, SA, like its gill surface area
should depend allometrically on its body weight. Unlike gill aorphometric
studies, however, only a handful of studies have reported such morphometric
relationships for fish intestine, tttese are summarised by Pauly (1981) and
Kapoor et al. (1975). The parameterization of is further complicated by the
uncertainty of what particular anatomical surface area most closely corresponds
to the intestine's effective surface area. For example, results of Wilson and
Dietschy (1974) Indicate that the intestine's effective surface is greater than
14
-------
the outside surface area of the intestinal serosa but much less than the inter-
nal surface area of the intestinal mucosa. Allometric regressions for these
two surface areas are summarized in Table 2. Because it is unclear whether the
differences between the coefficients in Table 2 are due to type of surface area
measured or to the food habits of the fish, FGETS arbitrarily uses the surface
area of the outer intestine to parameterize S^.
Because the movement of food through the gastrointestinal tract is so
slow, an expression for intestinal conductance, k^, does not require a term
like a Sherwood number to account for forced convection of the intestinal
contents. Consequently, this conductance can be defined simply as
ki « D/b (44)
where D is again the chemical's aqueous diffusivity (cm2/sec) and b is the
barrier thickness (cm) associated with- intestinal transport. This conductance
must be scaled to cm-2/day for use in Equation 11. Presently, FGETS assumes a
constant barrier thickness for all fish and chemicals equal to SO micrometers
(see Wilson and Dietschy 1974).
III.4. Equation 21
Recall that to estimate accurately the lethal chemical activity, LA, by
equation 20, it is essential that the fish essentially be equilibrated with the
water. Using the formalism of the gill exchange model summarized in section
II.1, this requires that
Ca/Cw - 1 - k ¦ (1 - exp(-(kt t)/BCF) (45)
where kj - (Sg kw)/W is the fish's first order uptake rate for the chemical.
Since the only physio-chemical property that these uptake rates depend upon
is chemical diffusivity, which varies only slightly between chemicals, k^ is
essentially a constant for a given size of fish (see for example Opperhuizen -
15
-------
1986) and consequently Equation 45 essentially depends only on the fish's lipid
fraction and the chemical's Kw via BCF. If e equals 0.01 (i.e., the fish is
99% equilibrated), then for 30-day-old fathead minnows like those used by Veith
et al. (1983), LA should be estimated using only LCSO (96 hr)'s for chemical's
with log Kow<4.• In particular, because 30-day-old fathead minnows typically have
lipid fractions (pj) and daily uptake rates (kj) approximately equal to 0.04 and
1000, respectively (Call et al. 1980, Eaton et al. 1983), for a 96-hour exposure
Equation 45 becomes
Kow " ~ 4000/(0.11 ln(c) (46)
(see Equation 43). Therefore, if e equals 0.01, chemical concentrations in
fish and water will be equilibrated only for chemicals whose logP is less
than 3.89.
IV. MODEL VALIDATION
Aqueous uptake and elimination of organic chemicals by fish generally have
been described using the linear, first order model
dCf/dt - k1 Cw - k2 Cf, (47)
where k^ and k2 are rate constants with dimensions of reciprocal time. Since
chemical concentrations in fish may decrease due to 1) growth of the fish
(i.e., biodilution); 2) branchial, fecal, or urinary excretion} and 3) metabolic
transformation, k2 is, in general, the sum of at least five individual transfer
rates, i.e.,
*2 " T + Eg + Cf + eu + p (48)
where y is the fish's specific growth rate
-------
of nonpolar, nonmetabolized organic chemicals, then thia equation can be
simplified and rewritten as
eg - k2 " Y
Since the gill exchange component of FGETS predicts that tg equals (Sg ky)/
(W BCF)t FGETS could be validated initially by comparing these predicted rates
to (k2- y ) that can be calculated from the literature. Using published data
for rainbow trout, Salmo gairdnerl, (Branson et al. 1975; Neely 1979; Niimi and
Cho 1981; Niimi and Oliver 1983; Oliver and Niimi 1983, 1984, 1985; Niimi and
Palazzo 1985; Branson et al. 1975, 1985; Niimi 1986) the following functional
regression was calculated
log(k2 -y) - 0.919 log(Sg kw)/(W BCF) - 0.429 (n-68; r-0.75) (50)
Because the 95% confidence interval for this regression's slope is (1.15, 0.74),
the predicted gill excretion rates are directly proportional to (k2 - Y) and
in particular
k2 - Y " (0.37 Sg kw)/(W BCF) (51)
This result is quite remarkable because, under nominal conditions, the functional
surface area of fish gills has been estimated to be approximately 36% of their
anatomical area which Sg measures (Booth 1978, Duthie and Hughes 1987, Gehrke
1987).
Clearly, if the majority of a fish's excretion is not branchial, the
foregoing analysis would be fortuitous. There are, however, theoretical con-
siderations to support the assumption that the gills are indeed a fish's prin-
ciple excretory organ for such chemicals. The relative contribution of branch-
ial and fecal excretion can be estimated by the ratio of the mass flux (Eg - mass
chemical/day) across the gills to the fecal excretory flux (Ef • mass chemical/
day) when contaminated trout are placed in clean water and fed clean food
17
-------
ad libitum, if fecal excretion la modeled kinetically (i.e., Equation II),
the following inequality can be calculated
Eg/Ef ¦ (Sg ky Ca)/(S^ ki C#) (52)
¦ (Sg Nsh b)/(Si d)
> (Sg 3.77 b)/(Si d)
For trout this inequality becomes
Eg/Et > 19.23 w0,339 (53)
If this relationship is adjusted for the trout's functional gill area, i.e.,
0.36 Sg, then, for 1 kg trout like those studied by Niimi and Oliver (1983),
branchial excretion would be estimated to be at least 70 times greater than
the trout's fecal excretion. Although this calculation supports the assumed
importance of gill excretion, it is at odds with results reported by McKim
and Heath (1983) and McKim et al. (1966).
Qualitative validation of FGETS was established by simulating the bioac-
cumulation of Aroclor 1254 by Lajco Michigan lake trout (Salvellnufc namaycush)
assuming first gill exchange only and then joint gill and food exchange.
Appendix C presents the FGETS user file used for this model application.
Inspection of the simulation results presented in Appendix D si jws that FGETS
accurately reproduced the bioaccumulation of Aroclor 1254 by lake trout for
fish up to 6 years old when joint kinetic exchange is assumed. The observed
whole body concentrations of PCB's indicated in these FGETS plots are data for
fish collected in 1971 (see Thomann and Connolly 1984). The failure of FGETS
to predict whole body concentration of PCB's in older lake trout is probably
due to the fact that these older 1971 trout were exposed to water concentrations
that were historically higher than 8.5 ng/1.
Thomann and Connolly (1984) also constructed a bioaccumulation model of
PCB's by Lake Michigan lake trout. Although their model also fits observed
18
-------
data, their model indicate! that gill uptake la insignificant when compared
with PCB accumulation via the food chain. In particular, Thornann and Connolly
concluded that 99% of the total body burden of PCB was accumulated through
food. This conclusion conflicts with FGETS's prediction that the ratio of gill
uptake to food aptake is approximately 1:3. Acknowledgment of this difference
is important if heterogeneous, time varying exposures, instead of constant PCB
exposure scenarios, are simulated.
V. CONCLUSION AND PROSPECTUS
Initial analysis indicates that FGETS can simulate observed patterns of
bioaccumulation and depuration of a single, nonpolar, slowly or nonmetabolized
organic pollutant by individual fish quite well. Extensions planned for FGETS
includes
A. Simulation of the bioaccumulation of mixtures of two or more
nonpolar, nonmetabolized chemicals simultaneously.
B. Integration of FGETS to realistic predator-prey models to more
accurately describe food chain exposure.
C. Incorporation of models for the metabolic transformation of organic
pollutants.
0. Modification of FGETS to model kinetic exchange of polar or charged
organics.
E. Consideration of nonkinetic mechanisms that affect whole body
concentrations of organic chemical — such as the rapid catabolism or
depuration the lipid reserves associated with reproduction.
19
-------
REFERENCES
Al-Hussaini, A.H. 1949. On the functional Morphology of the alimentary tract
of some fish in relation to differences in their feeding habitat anatomy
and hiatology. Quart.J.Micr.Sci. 90s 109-139.
Barber, M.C., L.A. Su^rez, R.P.. Lassiter. 1987. Kinetic exchange of nonpolar
organic pollutants by fish. Environ.Toxicol.Chen. Submitted.
Beamish, F.W.H. and M. Legrow. 1983. Bioenergetics of the southern brook
lamprey, Ichthyomyzon gagei. J.Anim.Ecol. 52:575-590.
Booth, J.H. 1978. The distribution of blood flow in the gills of fishi
application of a new technique to rainbow trout (Salmo gairdneri).
J.Exp.Biol. 73:119-129.
Branson, O.R., G.E. Blau, H.C. Alexander, and W.B. Neely. 1975. Bioconcentra-
tion of 2,2',4,4'-tetrachlorobiphenyl in rainbow trout as measured by an
accelerated test. Trans.AnuFish.Soc. 104t785-792,
Branson, D.R., I.T. Takahashi, W.M. Parker, and G.E. Blau 1985. Bioconcen-
tr&tion kinetics of 2,3,7,8 tetrachlorodibenzo-p-dioxin in rainbow trout.
Environ.Toxicol.Chem. 4i779-788.
BriggB, G.C. 1981. Theoretical and exporimental relationships between soil
adsorption, octanol-water partition coefficients, water solubilities,
bioconcentration factors, and the parachor. J.Agric.Food Chem. 29:1050-
1059.
Burnstock, G. 1959. The morphology of the gut of the brown trout (Salmo
trutta). Quart.J.Micr.Sci. 100: 183-198.
Call, D.J., L.T. Brooke, and P.-Y. Lu. 1980. Uptake, elimination and
metabolism of three phenols by fathead minnows. Arch.Environ.Contain.
Toxicol. 9:699-714.
20
-------
Chiou, C.T. 1985. Partition coefficients of organic compounds in lipid-water
systems and correlation* with fish bioconcentration factors. Environ.Sci.
Technol. 19:57-62.
Chiou, C.T., R.L. Malcolm, T.I. Brinton, and D.E. Kile. 1986. Water solubility
enhancement of some organic pollutants and pesticides by dissolved humic
and fulvic acids. Environ.Sci.Technol. 20i502-508.
Chiou, C.T. and D.W. Schmedding. 1982. Partitioning of organic compounds in
octanol-water systems. Environ.Sci.Technol. 16:4-10.
Colton, C.K., K.A. Smith# P.Stroeve, and E.W. Merrill. 1971. Laminar flow
mass transfer in a flat duct with permeable walls. AIChE 17:773-780.
Craig, J.F. 1977. The body composition of the adult perch, Perca fluviatilia
in Windermere, with reference to seasonal changes and reproduction.
J.Anim.Ecol. 46:617-632.
de Jager, S. and W.J. Dekkors. *1976. Relations between gill structure and
fish activity. Nether.J.Zool. 25:276-308.
Duthie, G. and G.M.Hughes. 1987. The effects of reduced gill area and hyperopia
on the oxygen consumption and swimming speed of rainbow trout. J.Exp.Biol.
127:349-354.
Eaton, J.G., V.R. Mattson, L.H. Mueller, and O.K. Tanner. 1983. Effects of
suspended clay on bioconcentration of Kelthane in fathead minnows.
Arch.Environ.Contam.Toxicol. 12:439-445.
Elliott, J.M. 1976. Body composition of the brown trout (Salmo trutta L.) in
relation to temperature and ration size. J.Anim.Ecol. 45:273-289.
Eschmeyer, P.H. and A.M. Philips. 1965. Pat content of the flesh of siscowets
and lake trout from Lake Superior. Trans.Amer.Pish.Soc. 94:62-74.
Ferguson, J. 1939. The use of chemical potentials as indices of toxicity.
Proc.R.Boc. London Ser. B 127:387-404.
21
-------
Friant, 8.L. and L. Henry. 1985. Relationship batwin toxicity of certain
organic compounds and thair concentrations in tissues of aquatic organisms!
a perspective. Chemosphere 14(1897-1907.
Gehrke, P.C. 1987. Cardio-respiratory morphometries of the spangled perch
Leioptherapon unicolor (Cunter, 1B59), (Porcoidei, Teraponidae). J.Pish
Biol. (accepted 4 March 1987).
Gill, H.8. and A.H. WeatherLey. 1983. Protein, lipid and caloric contents
of bluntnose minnow, Pimephales notatus Rafinesque, during growth at
different temperatures. J.Pish Biol. 25(491-500.
Gobas, P.A.P.C., A. Opperhuizen, and 0. Hutzinger. 1986. Bioconcentration
of hydrophobic chemicals in fishi relationship with membrane permeation.
Environ.Toxicol.Chem. 5i637-646.
Gobas/ P.A.P.C. and D. Mackay. 1987. Dynamics of hydrophobic organic
chemical bioconcentration in fish. Environ.Toxicol.Chem. 6t495-504.
Cohar, h.A.P. and A.F.A. Latiff. 1959. Morphological studies on the gut of
some scarid and labrid fishes. Public.Mar.Biol.Stn. Ghardaqa (Red Sea)
No. 10< 145-189.
Grzenda, A.R., D.F. Paris, and W.J. Taylor. 1970. The uptake, metabolism, and
elimination of chlorinated residues by goldfish (Carassius auratus) fed a
14C-DDT contaminated diet. Trans.Am.Pish.Soc. 99i385-396.
Hakim, A., J.S.0. Hunshi, and G.M. Hughes. 1978. Morphometries of the
respiratory organs of the Indian green snake-headed fish, Channa punctata.
J.Zool. 184(519-543.
Holling, C.8. 1966. Functional response of invertebrate predators to prey
density. Mem.Entomol. Soc.Can. 48t1-B6.
Hughes, G.M. 1966, The dimensions of fish gills In relation to their function.
J.Exp.Biol. 45i177-195.
22
-------
Hughe*, G.M. 1972. Morphometries of fish gills. Respir.Physiol. 14x1-25.
Hughes, G.M. 1978. On the respiration of Torpedo marmorata. J.Exp.Biol.
73t85-105.
Hughes, G.M. 1980. Ultrastructure and morphometry of the gills of Latimeria
chalumnae, and a comparison with the gills of associated fishes.
Proc.R.Soc.Lond.B 208:308-328.
Hughes, G.M. 1984. General anatomy of the gills. Ins W.S. Hoar and D.J.
Randall (eds). Fish Physiology, vol XA. Academic Press, pp 1-63.
Hughes, G.M. and N.K. Al-Kadhomiy. 1986. Gill morphometry of the mudskipper
Boleophthalmus boddarti. J.Mar.Ass.U.K. 66:671-682.
Hughes, G.M. and I.E. Gray. 1972. Dimension and ultrastructure of toadfish
gills. Biol.Bull. 143*150-161.
Hughes, G.M., S.F. Perry, and J. Piiper. 1986. Morphometry of the gills of
elaamobranch Scylorhinus stellar in in relation to body size. J.Exp.Biol.
121:27-42.
Ingham, D.B. 1984. A note on the mass transfer in the entrance to a flat duct
with permeable walls. Int.J.Heat Mass Transfer 27:2421-2422.
Jensen, A.L., S.A. Sprigarelli, and M.M. Thoirunes. 1982. PCB uptake by five
species of fish in Lake Michigan, Green Bay of Lake Michigan, and Cayuga
Lake, New York. Can.J.Fish.Aquat.Sci. 39:700-709.
Jensen, A.L. 1986. Functional regression and correlation analysis.
Can.J.Fish.Aquat.Sci. 43:1742-1745.
Jobling, M. 1981. The influence of feeding on the metabolic rate of fishes:
a short review. J.Fish Biol. 18:385-400.
Jobling, M. 1986. Mythical models of gastric emptying and implications for
food consumption studies. Environ.Biol.Fishes 16:35-50.
23
-------
Kapoor, B.G., H. Smit, and X.A. Verighina. 1975. The alimentary canal and
digestion in teleost. Im F.R. Russell and M. Yonge (eds). Advances
in marine biology, vol.13. Academic press, pp 109-239.
Karickhoff, S.W. 1981. Semi-ejnp^rical estimation of sorption of hydrophobic
pollutants on natural sediments and soil. Chemosphere. 10:833-646.
Kays, W.M. 1966. Convective heat and mass transfer. McGraw-Hill, pp 387.
Konemann, H. 1981. Quantitative structure-activity relationships in fish
toxicity studies. Part 1: relationship for 50 industrial pollutantB.
Toxicology. 19:209-221.
Landolt, J.C. and L.G. Hill. 1975. Observations of the gross structures and
dimensions of the gills of three species of gars (Lepisosteidae).
Copeia. 1975(3)«470-475.
Lauder, G.V. 1984. Pressure and water flow patterns in the respiratory tract
of the bans (Micropterus salmoldes). J.Exp.Biol. 113i151-164.
Layton, H.E. 1987. Enecgy advantage of counter-current oxygen transfer in
fish gills. J.lheor.Biol. 125i307-316.
Love, R.M. 1970. The chemical biology of fishes. Academic Press. 547p.
Mackay, D. 1962. Correlation of bioconcentration factors. Environ.Sci.
Technol. 16i274-278.
Mackay, D. and A.l. Hughes. 1984. Three-parameter equation describing the
uptake of organic compounds by fish. Environ.Sci.Technol. 18(430-444.
McCarty, L.S. 1986. The relationship between aquatic toxicity QSARs and
bioconcentration for some organic chemicals. Environ.Toxicol.Chen.
5:1071-1080.
24
-------
McCarty, L.S., P.V. Hodson, G.C. Craig, and K.L.E. Kaiser. 1985. Th* use of
quantitative structure-activity relationships to predict the acute and
chronic toxicities of organic chemicals to fish. Environ.Toxicol.Cheo.
4:595-606.
McKim, J.M. and E.M. Heath. 1983. Dose determination for waterborne 2,2',5,5'
-[14CJ tetrachlorobiphenyl and related pharmacokinetics in two species of
trout (Salmo gairdneri and Salvelinus fontinalis): A mass-balance approach.
Toxicol.Appl.Pharm. 68s177-187.
McKim, J.M., P.K. Schmieder, and R.J. Erickson. 1986. Toxicokinetic modeling
of [14C]pentachlorophenol in rainbow trout (Salmo gairdneri). Aquat.
Toxicol. 9i59—80.
Miller, M.M., S.P. Waslk, G. Huang, W.Shlu, and D. Mackay. 1985. Relationships
between octanol-water partition coefficient and aqueous solubility.
Environ.Set.Technol. 19«522-529i
Milton, P. 1971. Oxygen consumption and osmoregulation in the shanny, BlennJus
pholis. J.Mar.Biol.Ass.U.K. 51<247-265.
Montgomery, W.L. 1977. Diet and gut morphology in fishes, with special
reference to the monkeyface prickleback, Cebidichthys violaceus
(Stichaeidaes Blennioidei). Copeia. 1977(1)i178-182.
Muir, B.S. and G.M. Hughes. 1969. Gill dimensions for three species of tunny.
J.Exp.Biol. 51(271-285.
Mullins, L.J. 1954. Some physical mechanisms in narcosis. Chem.Rev.
54(289-323.
Murphy, P.G. and J.V. Murphy. 1971. Correlations between respiration and
direct uptake of DDT in the mosquito fish Gambusia affinls. Bull.
Environ.Contam.Toxicol. 6«581-588.
25
-------
Neely, W.B, 1977. A material balance study of polychlorinated biphenyls in
Lake Michigan. Itie Science of the Total Environment. 7:117-129.
Neely, W.B. 1979. Estimating rate constants for the uptake and clearance of
chemicals by fish. Environ.Sci.Technol. 13:1506-1510.
Neely, W.B, 1984. An analysis of aquatic toxicity data: water solubility and
acute LC50 fish data. Chemosphere. 13:813-819.
Niimi, A.J. 1986. Biological half-lives of chlorinated diphenyl ethers in
rainbow trout (Salmo gairdneri). Aquat.Toxicol. 9:105-116
Niimi, A.J. and C.Y. Cho. 1981. Elimination of hexachlorobenzene (HCB) by
rainbow trout (Salmo gairdneri), and an examination of its kinetics in
Lake Ontario salmonids. Can.J.Fish.Aquat.Sci. 38:1350-1356
Niimi, A.J. and S.L. Morgan. 1980. Morphonetric examination of the gills of
walleye, Stizostedion vitreum vitreum (Mitchill) and rainbow trout, Salmo
gairdneri (Richardson). J.Fish Biol. 16:685-692.
Niimi, A.J. and b.G. Oliver. 1983. Biological half-lives of polychlorinated
biphenyls (PCB) congeners in whole fish and muscle of rainbow trout (Salno
gairdneri). Can.J.Fish.Aquat.Sci. 40:1388-1394.
Niimi, A.J. and V. Palazzo. 1985. Temperature effect on the elimination of
pentachlorophenol, hexachlorobenzene and mirex by rainbow trout (Salmo
gairdneri). Water Res. 19:205-207.
Norheim, G. and S.0. Roald. 1985. Distribution and elimination of hexachloro-
benzene, octachlorostyrene, and decachlorobiphenyl in rainbow trout, Salmo
gairdneri. Aquat.Toxicol. 6:13-24.
Norstrom, R.J., A.E. McKinnon, and A.S.w. deFreitas. 1976. A bioenergetics-
based model for pollutant accumulation by fish. Simulation of PCB and
me thy Inter cury residue levels in Ottawa River yellow perch (Perca flavescens).
J.Pish.Res.Bd.Can. 33:248-267.
26
-------
Oikawa, S. and Y. Itazawa. 1985. Gill and body surface areas of the carp in
relation to body mass, with special reference to the metabolic-size
relationship. J.Exp.Biol. 117:1-14.
Oliver, B.G. and A.J. Niimi, 1983. Bioconcentration of chlorobenzenes from
water by rainbow trout: correlations with partition coefficients and
environmental residues. Environ.Sci.Techno1. 17:287-291,
Oliver, B.G. and A.J. Niimi. 1984. Rainbow trout bioconcentration of some
halogenated aromatics from water at environmental concentrations. Environ.
Toxicol.Chem. 31271—277•
Oliver, B.G. and A.J. Niimi. 1985. Bioconcentration factors of some halogenated
organics for rainbow trout: limitations in their use for prediction of
environmental residues. Environ.Sci.Technol. 19:842-849.
Opperhuizen, A, 1986. Bioconcentration of hydrophobic chemicals in fish. In:
T.M. Poston and R. Purdy(edn) Aquatic Toxicology and Environmental Pate:
ninth volume. ASTM STF 921. American Society for Testing and Materials,
pp 304-315.
Opperhuizen, A. and S.M. Scharp. 1987. Relationship between aqueous oxygen
concentration and uptake and elimination rates during bioconcentration of
hydrophobic chemicals in fish. Environ.Toxicol.Chem. 6:335-342.
Opperhuizen, A., E.W.v.d. Velde, F.A.P.C. Gobas, D.A.K. Liem, and J.M.D.v.d.
Steen. 1985. Relationship between bioconcentration in fish and steric
factors of hydrophobic chemicals. Chentosphere. 14:1871-1896.
Paloheimo, J.K. and L.M. Dickie. 1965, Food and growth of fishes. 1. A growth
curve derived from experiment data. J.Fish.Res.Bd. Canada. 22:521-542.
Patton, J.S., B. Stone, C. Papa, R. Abramowitz, and S.H. Yalkowsky. 1984.
Solubility of fatty acids and other hydrophobic molecules in liquid
trioleoglycerol. J.Lipid Res. 25:189-197.
27
-------
Pauly, D. 198)* The relationahipe between gill surface area and growth
performance In fiah: a generalisation of the von Bertalanffy'a theory
of growth. Meereatorach. 28:251-282.
Piiper, J., P. Scheid, S.F. Perry, and G.M. Hughe a. 1986. Effective and
morphological oxygen-diffuaing capacity of the gilla of the elaamobranch
Scyliorhinue atellaria. J.Exp.Biol* 123(27-41.
Price, J.W. 1931. Growth and gill development in the email mouthed black
baaa, Micropterua dolomien Lacepede. Studiea, Ohio State Univ. 4. 46pp.
Rayner. J.M.V. 1985. Linear relatione in biomechanics! the statistics of
acaling functions. J.Zool., Lond.(A) 206(415-439.
Robotham, P.W.J. 1978. The dimenaiona of the gilla of two species of loach,
Noemacheilua barbatulua and Cobitia taenia. J.Exp.Biol. 76:181-184.
Saundera, R.L. 1962. The irrigation of the gilla in fiahea. 11. Efficiency
of oxygen uptake in rolatipn to respiratory flow Activity and concentra-
tions of oxygen and carbon dioxide. Can.J.Zool. 40:817-862.
Sharma, S.N., G. Cuhu and B.R. Singh. 1982. Gill dimensions of a hillstream
fiah, Botia lohachata (Piscea, Cobitidae). Proc.Ind.Natn.Sci.Acad. B48>
81-91.
Shubina, L.l. and T.L. Rychagova. 1982. Dynaaica of fat and water metabolism
in Caapian ahad, Alosa caspia caapia, in relation to ita biological
peculiaritiea. J.Ichthyology. 22(123-127
Singh, B.R., A.N. Yadav, J. Ojha, and J.8.D. Hunahi. 1981. Croaa structure
and dimenaiona of the gilla of an inteatinal air-breathing fiah
(Lepidocephalichthya guntea). Copeia. 1981(1): 224-229.
Stephan, C.E. 1977* Methoda for calculating an LC50* in: F.L. Mayer and
J.L. Hame link {eda> Aquatic Toxicology and Hazard Evaluation A5TM STP 634.
American Society for Teating and Materials, pp 65-84.
28
-------
Stevens, E.D, and E.N. Lightfoot. 1986* Hydrodynamics of water flow in front
of and through the gills of skipjack tuna. Comp.Biochem.Physiol.
83A:255-259.
Stewart, D.J., D. Vteininger, D.V. Rottiers, ana T.A. Edsall. 1983. An
energetics model for lake trout, Salvelinus namaycush: application to the
Lake Michigan population. Can.J.Fish.Aquat.Sci. 40:681-698.
Stewart, D.J. and F.P. Binkowski. 1986. Dynamics of consumption and food
conversion by Lake Michigan alewives: an energetics-modeling synthesis.
Trans.Amer.Fish.Soc. 115:643-659.
Suarez, L.A., M.C. Barber, and R.R. Lassiter. 1987. GETS, A simulation model
for dynamic bioaccumulation of nonpolar organics by gill exchange: A
user's guide. EPA/600/S3-86-057.
Thomann, R.V. and J.P. Connolly. 1984. Model of PCb in the Lake Michigan lake
trout food chain. Environ.Set.Techno1. 18:65-71.
Unnithan, R.R. 1965. On the mucosal coefficient ot a few Indian fre&h water
fishes. Science and Culture. 31»387-388.
Ursin, E. 1967. A mathematical model of some aspects of fish growth, respira-
tion, and mortality. J.Fish.Res.Bd. Canada. 24:2355-2390.
van Veld, P.A., M.E. Bender, and M.H. Roberts, Jr. 1984. Uptake, distribution,
and clearance of chlordecone by channel catfish (Ictalurus punctatus).
Aquat.Toxicol. 5:33-49.
Veith, G.D., D.J. Call, and L.T. Brooke. 1983. Structure-toxicity relationship
for the fathead minnow, Pimephales promelas: narcotic industrial chemicals.
Can.J.Fish.Aquat.Sci. 40:743-748.
Vetter, R.D., M.C. Carey, and J.S. Patton. 1985. Coassimilation of dietary
fat and benzo(a)pyrene in the small intestine: an absorption model using
the killifish. J. Lipid Res. 26:428-434.
29
-------
Ware, D.M. 1972. Predation by rainbow trout (Salmo gairdneri): the influence
hunger, prey density, and prey size. J.Fish.Res,Bd.Can. 29:1193-1201.
Weatherley, A.H. and H.S. Gill. 1983. Protein, lipid, water, and caloric
contents of immature rainbow trout, Salmo gairdneri Richardson, growing
at different rates. J,Fish Biol. 23:653-673.
Wilson, F.A, and J.M. Dietschy. 1974. Hie intestinal unstirred layers its
surface area and effect on active transport kinetics. Biochim.Biophys.
Acta. 363:112-126.
Yalkowsky, S.H., O.S. Carpenter, G.L. Flynn, and T.G. Slunick. 1973. Drug
absorption kinetics in goldfish. J.Pharm.Sci. 62:1949-1954.
Yalkowsky, S.H., S,c. Valvani, and D. Mackay. 1983. Estimation of the aqueous
solubility of some aromatic compounds. Residue Reviews 85:43-55.
Yarzhombek, A.A., T.V. Shcherbina, N.F, Shmakov, and A.G, Gusseynov. 1984.
Specific dynamic effect of {ood on fish metabolism. J.Ichthyology.
24:111-117.
30
-------
Table 1. Summary of allometric parameters for gill surface area and
lamellar density.
S1
®2
p1
P2 i
Bource
Anabas testudineus
5.56
0.615
36.5
-0.152
Hughes (1972)
baracuda
0.274
1.281
-999*
-999
Hughes (1980)
Blennius pholis
7.63
0.849
28.3
-0.139
Milton (1971)
Boleophthalmus boddarti
0.93
1.050
53.1
-0.229
Hughes and Al-Kadhomiy
(1986)
Botia lohachata
9.13
0.700
78.0
-0.005
Sharma et al. (1982)
Catostoinus conunersonii
7.98
0.639
-999
-999
de Jager and Dekkers
(1976)
CatoBtomus commersonii
11.17
0.587
25.0
-0.107
Saunders (1962)
Channa punctata
4.70
0.592
-999
-999
Hakim et al, (1978)
Cobitis taenia
4.67
0.864
45.5
0.000
Robotham (1978)
Coryphaena hippurus
52.08
0.713
33.8
-0.036
Hughes (1972)
Cyprinus carpio
8.46
0.794
32.2
-0.079
Oikawa and Itazawa
(1985)
Gambupla affinis
2.33
0.873
-999
-999
Murphy and Murphy
(1971)
Ictalurus nebulosa
2.65
0.845
-999
-999
de Jager and Dekkers
(1976)
Ictaluru8 nebulosa
4.98
0.728
15.8
-0.091
Saunders (1962)
Katsuwonus pelamis
56.75
0.841
59.0
-0.076
Muir and Hughes (1969)
Leiopotherapon unicolor
4.68
1.040
41.2
-0.087
Gehrke (1987)
LepidocephalichthyB guntea
4.94
0.745
45.0
-0.221
Singh et al. (1981)
Lepisosteus sp.
3.94
0.738
38.8
-0.060
Landolt and Hill (1975)
Micropterus dolomieu
7.31
0.820
30.0
-0.062
Price (1931)
-------
Tabic 1. continued
Noemachceilus barbatulus
3.60
0.577
36.4
0.000
Opsanus tau
5.60
0.790
16.0
-0.075
Platichthys flesus
6.36
0.824
-999
-999
Raja clavata -999 0.970 -999 -0.154
Saccobranchus fossilis 1.86 0.746 31.6 -0.095
Salmo gairdneri 3.90 0.900 -999 -999
Salmo gairdneri
1.84
1.125
-999
-999
Salmo gairdneri
3.14
0.932
27.5
-0.064
Scomber scomber
4.42
0.997
27.1
0.023
Scyliorhinus canicula
2.62
0.961
17.2
-0.071
Scyliorhinus stellaris
-999
A
0.779
-999
-0.167
Stizostedion vitreum
0.80
1.129
-999
-999
Thunnus thynnus
24.43
0.901
60.9
-0.089
Tinea tinea
12.20
0.657
-999
-999
Tinea tinea
8.67
0.698
25.5
Torpedo marmorata
1.18
0.937
34.2
geometric mean
5.50
0.816
33.7
* "-999" indicates that the parameter was not available.
Robothan (1978)
Hughes and Gray (1972)
Hughes and Al-Kadhoniy
(1966)
Hughes et al. (1986)
Hughes et al. (1974)
de Jager and Dekkers
(1976)
Niimi and Morgan (1980)
Hughes (1984)
Hughes (1972)
Hughes (1972)
Hughes et al. (1986)
Niimi and Morgan (1980)
Muir and Hughes (1969)
de Jager and Dekkers
(1976)
Hughes (1972)
hughes (1978)
32
-------
Table 2. Summary of allometric pa ranter* intestinal surface areas of fish.
Parenthetic letters signify the parameters are for mucosal surface
areas, (m), or serosal surface area, (s).
6.339
6.223
Goblo gogio(in)
Rutilus rutilus (m)
Salmo trutta (s)
Solea solea (s)
various species (m)
1.198
2,12
1.83
•2
0.591
0.580
0.571
0.57
0.92
Pauly (1981)
Pauly (1981)
Burnstock (1959)
Ursin (1967)
M-Hussaini (1949)
Unnithan (1965)
Gohar and Latiff (1959)
Montgomery (1976))
33
-------
Appendix A. Definition the variables appearing the the text.
b - barrier thickness for diffusion across the intestine (cm);
BCF- (Pa + P^ + Pb K#) - bioconcentration factor in whole fish assuaing
gill exchange only (dlmenslonless);
¦ total burden of a chemical in food residing in the intestine (micro g);
Bf ¦ total body burden of a chemical (micro g / fish);
Ca - concentration of a chemical in the aqueous fraction of the fish (ppm);
C} - concentration of a chemical Jn the lipid fraction of the fish (ppm);
Cf ¦ whole body concentration of a chemical (ppm - micro g / g live fish);
C8 ¦ concentration of a chemical in the structural fraction of the fish (ppm);
Cw ¦ concentration of a chemical in water (ppm);
d ¦ mean lnterlamellar distance (cm);
D » aqueous diffusivity (cm^/sec);
E » egestive flux (g /day);
F -ad libitum feeding flux (g /day);
h - hydraulic radius of a channel (cm) • 2 (channel cross section area)/
(channel perimeter);
1 » mass of food resident in the Intestine (g);
Jg - diffusive flux of chemical across the gills (microgram/cm /day);
kw - chemical conductance through lnterlamellar water (cm /day or cm / sec);
K} - partition coefficient between lipid and water (dimensionlest);
Koc» partition coefficient between bulk organic carbon and water (dimensionles
*ow" partition coefficient between n-octanol and water matter (dlmenslonless)>
Ks ¦ partition coefficient between structural matter and water (dlmenslonless)
1 - mean length of lamellae (cm);
Pa * fraction of whole fish that is aqueous water (dlmenslonless);
Pi « fraction of whole fish that is lipid(dimensionless);
Ps « fraction of whole fish that is structural organic matter (dlmenslonless);
R ¦ routine respiratory flux (g /day)*
SDA* respiratory flux due to specific dynamic action (g /day);
- total intestinal surface area (cm2);
Sg * total gill surface area (cm2);
v - mean velocity of lamellar flow (cm/sec)
W - live body weight of fish (g);
a - assimilation efficiency of food (dlmenslonless);
0 ¦ assimilation efficiency of chemical from food (dlmenslonless);
V » dynamic viscosity (poises);
P * density of secondary lamellae on gill filaments (• lamellae/mm);
34
-------
App«ndix B. Abbreviated User Guide to FGET8.
B.I User Input
PGET8 require* a user input file which has the following general format*
c
c file
c
POETS#DAT
toxlab
molwt
logp
mp
spplab
famlab
liflab
act-gi11
wt
wtunits
mod$opt
plfish
ct ish
ctunita
cwater
cwunita
temp
cprey
plprey
bmf
tine
tuni ts
string
number
number
number
string
string
string
number
number
string
stringl string2 string3
string number1 number2
number
string
stringl string2 number1 number2 number3 number4
string
stringl string2 number1 number2 number3 number4
number1
number1
number1
number1 number2
string
/ end.
FGETS treats all records that begin with a "c" or "I" in column 1 as
comments. The exclamation symbol "I" can also be used anywhere in the record
field to start an end-of-line comment. Therefore users can document FGETS input
file in as much detail as desired. All records that begin with "/" are consid-
ered to define model input and may appear in any order. Blanks before or after
the input slash delimiter are not significant. Each input slash is followed by
a keyword or phrase, as indicated above, which identifies record's data. Key-
words must be spelled in full without any embedded blanks and must be separated
from the record's remaining information by at least one blank character.
Letter case is not significant since each record is transliterated to lower
case. A brief description of each of these input records follows.
RECORDt "/ toxlab string"
This record simply specifies the name of the chemical whose exchange
kinetics is being simulated and is used for output purposes only.
35
-------
RECORDi "/ logp number"
This record specifies the chemical's log Kow, where Kow is the chemical
n-octanol/water partition coefficient.
RECORDi "/ molwt number"
This record specifies the chemical's molecular weight(i.e., 9/mole).
RECORD1 "/ mp number"
This record specifies the chemical's melting point(i.e., Celsius).
Tttis data together with the chemical's logp is used by PGETS to
calculate the toxicant's chemical activity.
RECORD 1 "/ spplab string"
This record specifies the scientific nam^1 of the fish being modeled by
FGETS. Por example, rainbow trout must be specified as Salmo gairdneri.
This record and records, "/ famlab ..." and "/ liflab ...", are used
to extract appropriate gill morphometric parameters from the data base
file, MORPHO.DAT, for the simulation.
RECORDS "/ famlab string"
This record specifies the family of the fish being modeled by FGETS.
hECORDt "/ liflab string"
Thiu record specifies the life form of the fish being by FGETS. Presently,
the only recognized input for string is "freshwater" or "marine".
RECORDi "/ act-gill number"
This record specifies the fraction of the fish's anatomical gill that is
physiologically active. Typical values for the variable range betv«en
1/3 and 1/2 (Booth 1978, Piiper et al. 1986, Duthie and Hughes 1987).
This value is used to adjust the kinetic exchange rates predicted by FGETS
(see Barber et al. 1987).
RECORDt "/ wt number"
This record specifies the initial live weight of the fish being modeled
by FGETS. The units wt (i.e., g, kg, pounds, oz, etc.) are specified by
the record, "/ wtunits string".
RECORDi "/ wtunits string"
See RECORDi "/ wt number" above.
36
-------
RECORDt "/ mod$opt atringl string2 string3"
This record specifies various modeling options and maybe input in any
order.
String-Hgrcwth(arg)" specifies the function, g, that FGETS will use to
model the fish's growth, i.e., dW/dt ¦ g, where w is the fish's gram live
weight and t is time in days. Presently, there are three different options.
If arg-"linear, number", then linear growth, i.e.,
dVf/dt ¦ number * W (g.1)
is simulated. When arg«"allometric", then fish growth essentially is
modeled by
dW/dt - a *F - R (g.2)
where alpha is the fish's assimilation efficiency, and F and R are the
fish's daily feeding and respiratory fluxes, which are described by the
allometric functions, F"f1*W**f2 and R«r1*W**r2, respectively. If
arg-"holling", fish growth also is modeled by (g.2) but with the modifi-
cation that the fish's feeding is described by a Holling type formulation,
i.e.,
F ¦ - S) (g.3)
whet« Sroax is the fish's maximum atomach/gut capacity and S is the amount
o£ food presently resident j.n the gut, which is itself modeled by
dS/dt - F - G (g.4)
where G»g1*S**g2 represent gastric evacuation (Holling 1966, Ware 1972,
Jobling 1986). When the fish's growth is modeled by either the "allometric"
or "holling" option, FGETS attempts to retrieve all the required physiolog-
ical parameters form the data file, PHYSIO.DAT, based on either the fish's
family or species. If parameters do not exists in the data file, then
FGETS terminates with an appropriate error message.
String«"gill" specifies that FGETS will simulate bioaccumulation assuming
there is only gill exchange of the toxicant.
String*"joint(arg)" specifies that FGETS will simulate bioaccumulation
assuming there is joint gill and food exchange of the toxicant. Both
options, "joint" and "gill", can be specified concurrently.
RECORDt "/ plfish string numberl number2"
This record specifies the fraction of the fish's live weight that is
lipid. If string equals "database", FGETS estimates the fish's lipid
37
-------
fraction using an allometric function,
plfish - pj * w(g li ve) ••
1 2
that is retrieved from the PCET8 database and no number needs to be
specified. Alternatively, users can input their own allowetric function
by specifying "/ plfish allometric fj fj If this option is selected,
1 2
PCETS assumes that
plfish » f! * W ** f2
1 2
where Wt has units of wt$unlts below. There are two other valid options
for string. If "/ plfish constant is input, the fish'* 'ipid is held
constant during the simulation while if "/plfish exp fj " is input,
1 2
the fish's lipid fraction is generated dynamically as
plfish - fj_ * exp (f^ • t)
1 2
where the unit of t are specified by t$units below.
RECORD: "/ cfish number"
This record specifies the fish's initial whole body concentration on a
live weight basis of the chemical. The units of cfish are specified by
"/ cfunits .... "
RECORD* "/ cfunits string"
This record specifies the units of cfish. FGETS converts cfish from
cfunits to ppm for internal calculation and model output.
RECORDi "/ plprey number"
This record specifies the fractional lipid content of the fish's prey
(also see "RECORD: "/ plfish number" above). FGETS attempts to assign
a default value for this variable if it is not input or is specified as
a negative number. Presently, default values are available only for
salmonids, which are assumed to be piscivores. These defaults are assigned
by SUBROUTINE FODWEB.
RECORDt V cprey number"
This record specifies the toxicant's concentration in the fish's prey
(also see "RECORD» "/ cfish number" above). The units of this record
are assumed to be the same as those specified for record, "/cfish .
FGETS attempts to assign a default value for this variable if it is not
input or is specified as a negative number. This default is assigned by
SUBROUTINE FODWEB assuming that the prey is thermodynamicslly equilibrated
with the water. This default can be adjusted by record, "/bmf
See belew.
38
-------
RECORD; "/ bmf number"
This record specifies the bic-tgnification factor (bmf) toe the fish's
prey and is required only if the default option for cprey is used. The
default value for this record ia btftt-1. Xf the user want* the prey to
be biomagnified above thermodynamic equilibrium, then brnfM. On the other
hand, if the uaer assumes that the prey has not yet equilibrated with the
water, bmf<1. Based on preliminary analysis (Barber et al. 1987) typically
0
-------
where numberl and number3 haw unite of cwunita and number2 has unit*
of 1/tunits.
The unit* of cw(t) (a.g.» ppm, mg/1, ate.) and of t (e.g., daya, years,
ate.) ara specified by the record*, "/ cwunita string" and "/tunits string",
respectively. FGETS converts cw(t) fro* cwunita to ppm tor interna 1 calcu-
lationa and model output by appropriately converting either time series
data from file abc.dat or the parameters, number 1,..., number4. ttie values,
number 1,,.. ,'number 4, must be separated by one or more blanks.
RECORDi "/ temp stringl string2 number 1 number2 number3 number4"
ITiis record specifies the temporal dynamics of water temperature (Celsius).
Data specified on this record is processed like record "/ cwater . ..."
Por example, if temp(t) is the ambient water temperature at tljm t, and if
"/ temp file xyz.dat" is specified, then FGETS will read time series of
water temperatures from the file xyz.dat using a FORTRAN free formated
read statement that is equivalent to
READ (JTABLE, •, BND-1120) t, temp(t)
where file xyz.dat is attached to unit JTABLE. Again FGETS requires that
the records of this file be ordered according to ascending time, i.e.,
t(i) < t(i+1) for all i-th records, since FGETS linearly interpolates
water temperatures between all consecutive times, t(i) and t(i+1).
When a user has concomitantAdata for chemical concentration and vacer
temperature, tho files abc.dat and xyz.dat, specified by records,
"/ cwater file ..." and "/ tamp file ...", respectively, may be assigned
to be the same. Zn this case FGETS will read time series of chemical
concentrations and ambient water temperatures from the specified file
using a FORTRAN free formated read statement that is equivalent to
READ (JTAHLE, *, END-1120) t, cw(t), temp(t).
As before, if "/ temp function constant numberl " is specified, then
FGETS will generate ambient water temperatures as
temp(t) ¦ numbe r1
where numberl has units of Celsius. If "/temp ..." is specified as
"/ temp function sin numberl number2 number3 number4", then FGETS
will generate ambient water temperatures as
temp(t) ¦ numberl*sin(number2*t+number3) + number4
where number1 and number4 have units of Celsius and number2 has units
of 1/tunits. Finally, if "/ temp function exp numberl number2 number3 "
is input, then FGETS generates ambient water temperatures as
temp(t) ¦ number1«exp(number2*t) + numbers
where numberl and numher3 have units of Celsius and number2 has units of,
t/tuni t*.
40
-------
RECORDi "/ tin* number1 number2"
This record speciflea the beginning tine/ numberl, and ending time, nvmber2,
for the FGBTS '¦ simulation. The units of these tines are specified by the
record, "/ turits string." FGETS converts these times into days for
internal usage and subsequent model output. If the user does not Bpecify
number2, the the simulation's beginning time is assumed to be t*0 and
numberl is assigned as the simulation's ending time.
RECORD: "/ tunits string"
This record specifies the time units (e.g., hours, days, years, etc.)
associated with the user input records, "/ time "/ cwater
and "/ temp ...." The time units associate with these three data records
must be the samel The user must be careful to verify that the time units
associated with these three records and any asnociated exposure files (see
RECORDs "/ cwater ..." and RECORD: "/ temp ..." above) are indeed the
same.
RECORD: "/ end."
This record specifies the end of a user input sequence.
41
-------
B.2 Required Database File*
DATA FILEi MORPHO.DAT
This file contains morphometric data for the allometric functions:
s ¦ gill area (cm**2) ¦ sj • wt •• »2>
p ¦ # lamellae / mm gill filament « p 1 • wt *• p 2>
1 ¦ lamellar length (cm) ¦ lj • wt •• l2»
Data is organized in this file in seta of three records each. Each set
represents morphological data for one species. ltie XXX-th set of
MORPHO.DAT contains the following information:
XXX.1 species/family/lifeform
XXX.2 reference
XXX.3 si >2 f) | p 2 ^2
A value of -999 designates that the parameter was not reported.
DATA FILE: PHYSIO.DAT
This file contains physiological data tor the allometric functions:
cnax » maximum obsirved ingestion (gram/day) - cmaxl • wt *• cmax2j
gmax ¦ maximum capacity of stomach (gram) - gmaxl * wt •• gmax2;
fsat - size of satiation meal consumed during (0, tsat), gram
« fsatl * wt fsat2; *
tsat ¦ time (min) to satiation when feeding with an initially empty
stomach;
¦ tsatl • wt ** tsat2 + tsat3;
evac ¦ stomach evacuation (gram/day) - evacl * g ## evac2 ;
in general evac2 - 1/2, 2/3, or 1.0 (see jobling 1981);
imax ¦ maximum intestinal capacity (gram) - imaxl • wt ** imax2;
o2 • routine respiration (mg o2 consumed/ hr) at temperature tref
- o2$1 • wt ** o2$2;
pi ¦ fraction fat ¦ Pi * wt ** pj ;
1 2
as well as constant parameters:
alpha - assimilation efficiency;
rq ¦ respiratory quotient 1 co2 respired/ 1 o2 consumed;
tref « physiological reference temperature (c);
qtO ¦ q10 for temperature deviation from tref.
Data are organized in thiB file in sets of five records each. Each set
represents physiological data for one species. The XXX-th set of
MORPHO.DAT contains the following information:
XXX.1 species/family
XXX.2 reference
XXX.3 cmaxl cmax2 gmaxl gmax2 fsatl fBat2 tsatl tsat2 tsat3
XXX.4 imaxl imax2 evacl evac2 alpha o2$1 o2$2 rq q10 tref
XXX.5 pi pi
1 2
A value of -999 designates that the parameter was not reported.
42
-------
Appendix C. Example FGETS user input sequence.
c
c file i lake_trout.dat
c updatei 20-aug-l987 09i13t48
c
c notes concerning parametersi
c
c i. assume cw • (13+12.4+2.9+5.7)/4 - 8.5 ng/1 based ons
c cw » 13 ng/1 (veith (1972) see weininger (1978));
c » 12.4 ng/1 calculated by weininger (1978)) assuming 60% of total
c pcb reported by haile (1977) is particulate bound;
c - 2.9 ng/1 mean of 8 samples for aug J979 (range 1.1-11.2) reported
c by rice et al. (1982);
c ¦ 5.7 ng/1 mean of 4 samples for apr 1980 (range 4.7-7.1) reported
c by rice et al. (1982);
c
c note that neely (1977) and jensen et al. (1982) assumed cw = 10 ng/1
c while thomann and connolly (1984) assumed cw «• 5 ng/1?
c note also that doskey and andren (1981) report 2.6e-11 molar aroclor 1254
c in lake michigan which converts to 8.48 n/1 (see ii.1 below).
c
c ii. physico-chemical properties of aroclor 1254: aroclor 1254 is by weight
c 54% chlorine, since tetra-, penta-, and hexa-pcb's are 48.6%, 54.3%,
c and 59.0% chlorine, respectively, assume aroclor 1254 is essentially
c a penta-pcb. therefore:
c 1) molwt » 326.25 g/mole.
c 2) mp - 102.6 celsius, i.e., mean of 5 penta-pcb's from mackay
c et a 1. (1982)>
c 3) logp - 6.62 based on tha following considerations!
c tw/in solubility of aroclor 1254 1 .825e-2 mg/1 (schnoor «t al. 1987),
c which converts to 5,594e-8 molar (assuming 326 g/mole). consequently,
c
c log(5.594©-8)«-0.944*logp - 0.01*102.6 + 0.323 «> logp » 6.94
c
c or
c
c logp—0.862(log(5.5948e-8) + 0.01*002.6-25) ) + 0.710 - 6.29,
c
c see yalkowsky et al. (1983) and chiou and schmedding (1982),
c respectively, therefore assume logp-(6.94+6.29)/2"6.62. note
c a mean calculated logp for penta-pcb is 7.47 (n-5 see yalkowsky
c et al. 1983) while a mean measured logp for penta-pcb's is 6.11
c (n»3 see miller et al. (1985) and chiou (1985)).therefore logp
c equals (7.47+6.11)/2 ¦ 6.79.
c
c references!
c -chiou 1985. environ.sci.technol. l9i57-62.
-doskey and arden 1981. environ.sci.technol. 15s705-710.
-haile 1977.
-jonsen et al. 1982. can.j.fish.aquat.sci. 39:700-709.
-mackay et al. 1980. chetnosphere 9:257-264.
-miller et al. 1985. environ.sci.technol. 19(522-529.
-noely 1977. sci. total environ. 7H17-129
-rice et al. 1982. j. great lake res. 8:265-270.
-schnoor et al. 1987. epa/600/3-87/015.
-Stewart et al. 1983. can.j.fish.aquat.sci. 40:681-698.
-thomann and connolly 1984. environ.sci.technol. 18:65-71,
-veith 1972.
-weininger 1978.
-yalkowsky et al. 1983. residue reviews 85:43-55.
-------
/
toxlab
aroclor 1254 (penta-pcb'»)
/
molwt
326.25
/
logp
6.62
/
mp
100.0
/
spplab
salvelinua namaycush
/
f atnlab
ealmonidae
/
lif lab
freshwater
/
wt
100.0
/
wtunits
9
/
mod$opt
growth(hoiling, 0.5) gill joi
/
plfish
database
/
cf ish
0.0
/
cfuni ts
ppm
/
ovater
function constant 8.5
/
cwunits
ng / 1
/
tine
8
/
tunita
year
/
temp
function sin 4.0 6.283185 0.
/
act-gi11
0.37
/
cprey
5.0
/
plprey
0.07
/
bmf
1.0
ta
as reported
by thoman and connoly (1984) .
I frequency in year*"-1
(1984) and Stewart et al.(1983)
yr
cf(ppm)
wt(g live^
4
4.0 +- 3.0
1710 +- 0.0
5
fi.O +- 5.5
2800 +- 0.0
6
8.5 +- 3.5
3590 +- 0.0
7
13.5 +- 5.0
4310 +- 0.0
8
18.5 +- 4.0
5180 +- 0.0
/ end.
44
-------
Appendix D. Example FGETS output for user input presented in Appendix C.
***summary of user input specifications***
input toxicant! aroclor 1254 (penta-pcb's)
input molwt: 326.
input logp: 6.62
input mps 100.
input cwater: cv(ppm) ¦ 3.500E-06
input temps Celsius - 4.00 * sin( 1.720E-02 * t(days) + 0.000E+00) + 8.
input tstart: O.OOOE+OO days
input tend: 2.922E+03 days
input fishid: salvelinus namaycush
input wtO: 100. live weight, g
input growth: hollingj c/cmax= 0.500
input plfish: database function, pi = 2.158E-03 • w(g) •• 0.497
input cfish: O.OOOE+OO ppm
input active$gill: 0.370
input joint: kinetic
input plprey: 7.000E-02
input bmf: 1.00
input cprey: 5.00
***parameters for narcotic toxicity***
model lc50: 8.544E-04 ppm » 2.619E-09 molar ¦ 1.508K-02 activity
model water solubility: 1.708E-02 ppm ¦ 5.236E-00 molar
•••summary of parameters for fish growth***
model mean temperature:
model growth rate (g/ g/day):
model ingestion rate (g/ g/ day):
model respiration rate (g/ g/ day):
8.00
c
0.270
*
w(g)
* *
-0.787
( 1 . 250E-03)
0.263
•
w(g)
*«
-0.602
9.886E-03
•
w(g)
**
-0.319
•••summary of parameters for gill only exchange***
model gill surface area (cm**2):
model interlamellar distance (cm):
model lamellar length(cm):
model uptake rate, k1(1/ day):
model excretion rate, k2(1/day):
2.86 • w(g) ** 0.983
2.343E-03 * w(g) ** 7.306E-02
1.870E-02 * w(g) ** 0.208
512. * w(g) ** -0.104 ( 245. )
1.087E-02 * w(g) ** -0.445 ( 4.086E-O4)
•••summary of parameters for food exchange***
model uptake rate, k1(1/day): 3.605E-03 * w(g) •• -0.136 ( 1.396E-03)
model elimination rate, k2(1/day): 1.537E-03 • w(g) •* -0.785 ( 7.236E-06)
45
-------
growth of salvelinus namaycush
3851.
3701 .
3551.
3401 .
3251.
3101.
2951 .
2801.
2651 .
2500.
2350.
2200.
2050.
1900.
1750.
1600.
1450.
1300.
1150.
1000.
850.2
700. 1
550. 1
400.1
250.0
100.0
*•*
**
**
*•*
**
• •
* • *
*•*
• *
* **
#**
* *
**
**
***
***
* * *
* **
* • *
* * * *
* # #
* • * *
**** + +
C.OOtf+OC 5.B4E+02 J•17E+03
x-axisi days
y-axisj wt, g live
1.75E+G3 2.33E+03 2.92E+03
mode i
mean temperature:
8.00
c
mode 1
total ingestion:
1.451E+04
9
model
total evacuation:
1.42SE+04
9
model
total assimilation:
1.053E+04
9
model
total egestion:
3.699E+03
9
model
total respiration:
4.665E+03
9
46
-------
gill exchange of aroclor 1?54 (penta-pcb*s) by talvelinus naoaycush
22.50
21.60
20.70
19.80
18.90
18.00
17.10
16.20
15.30
14.40
13.50
12.60
11.70
10.80
9.900
9.000
8.100
7.200
6.300
5.400
4.500
3.600
2. 700
1 .flOO
0.9000
0.00O0E+00 •*»«•*»»»•«»*••*••-~ > —+ +
0.00E+00 5.84E+02 1.17E+03 1.75E+03 2.34E+03 2.92E+03
x-axifli days
y-axiBi whole body concentration (ppm)
model mean water conc«i
model mean temperature:
model total gill uptaket
model total gill excretion*
8.500E-06 ppm
8.00 c
1.048E+04 micro g
2.651E+03 micro g
47
-------
joint exchange of «roclor 1254 (penta-pcb's) by aalvelinus naaaycush
22.50
21.60
20,70
19.80
18.90
18.00
17.10
16.20
15.30
14.40
13.50
12.60
11.70
10.80
9.900
9.000
8.100
7.200
6.300
5.400
4.500
3.600
2.700
1 .800
0.9000
0.0000E+00
• * • • o
»••• o
0.00E+00 5.84E+02 1.17E+03 1.75E+03
2.34E+03 2.92E+03
x-axist days
y-axis: whole body
model mean water conc.t
model mean prey conc.:
model prey bmft
model mean temperaturet
model joint gill uptake:
model joint gill excretion:
model joint gut uptaket
model joint gut excretion*
concentration (ppm)
8.500E-06 ppm
5.00 ppm
1.00
8.00 c
1.048E+04 micro g
1.156E+04 micro g
3.369E+04 micro g
118. micro g
48
------- |