OVERVIEW OF THE ATLSS LANDSCAPE FISH MODEL (ALFISH)

Holly Gaff, Louis J. Gross, and Rene" Salinas
The Institute for Environmental Modeling
Departments of Ecology and Evolutionary Biology
and Mathematics
University of Tennessee, Knoxville 37996-1610


ASSESSMENT OBJECTIVES

The ATLSS Landscape Fish model (ALFISH) has as its main objective the
capability to compare in a spatially-explicit manner the relative effects
of alternative hydrologic scenarios on fresh-water fish densities
across South Florida. Another objective is to provide a measure of the
dynamic, spatially-explicit food resources available to wading birds.
By providing a model for the key resource base for wading birds, ALFISH 
allows the linkage of the hydrologic effects on fish densities with
models for wading bird foraging. ALFISH has been developed in regular
consultation with several field biologists and has made use of a 
variety of data sources on fish distributions to estimate parameters.


METHODS

ALFISH operates by splitting the landscape into spatial grid cells, of
size 500 m by 500 m, characterizing the within-cell variability in
water depth in a statistical manner. Within each cell, there is a
distribution of elevations based upon an average hypsograph obtained by
averaging data from a number of locations. Cells also contain
permanently wet areas of small size, called ponds, with the remaining
cell areas which may be subject to periodic dry down and reflooding
called the marsh areas. Fish densities change within cells between
areas of various depths as a cell drys down or rewets, since differing
fractions of area within cells will be at differing water depths. The
spatial cells are coupled in the model by movements of fish density
between cells due to differences in relative fish densities or
differences in relative water depth between cells. 

The main inputs to ALFISH are hydrology data at a 500 m resolution from
the ATLSS High Resolution Hydrology Model, time of year, and food 
resources available from lower trophic levels. For the application of 
ALFISH to the Restudy, all lower trophic level resources are assumed 
constant, independent of hydrology. The time step of the model is 5-days
and the study area includes all of the region for which hydrology data 
are being produced for each scenario, with some outputs being computed 
only for particular sub-regions within the study area.

ALFISH considers two fish functional groups: Small Fish which includes 
all fish species with a maximum possible length of 7 cm and 
Large Fish which includes all fish species with maximum lengths of 
greater than 7 cm. ALFISH treats the fish functional groups as an age-size
structured model. The fish in each functional group grow in size 
every 5-day time step. The age of maturity and the fecundity are 
unique to each functional group. Four causes of mortality are
included: (i) Background mortality, or the natural mortality of an 
uncrowded population, which is dependent on fish age class, but is 
independent of population size; (ii) Density-dependent
mortality from starvation; (iii) Loss due to predation from other
functional groups; and (iv) Death due to dry down in which some fraction
of fish density do not successfully reach deeper water as a cell drys.

Movement in ALFISH has two phases. First, within cell movement takes
place allowing fish density to move between the pond and the marsh
areas of various depths within a particular cell. Secondly, fish
density can shift between the marsh areas of adjoining cells, based on
differences between water depth and fish densities in these cells. 

ALFISH has been applied to all the major Restudy scenarios, with 
comparisons made to F2050 as the base scenario. The ALFISH outputs 
include a wide variety of maps of both average and single time
period total fish densities across the region, maps of fish densities
available to wading birds (e.g. fish in appropriate size ranges and at 
appropriate waters depths to be available to wading birds), various
time series of fish densities averaged across various subregions, time 
series of histograms of the size distribution in particular
subregions, and tables summarizing the total fish number densities 
broken down by basin and by year. 


FUTURE PLANS

Several major modifications to ALFISH are either ongoing or planned.
The ongoing efforts include incorporation of maps of alligator hole 
distributions, use of these to appropriately estimate the area
in ponds across the model region, performing uncertainty analysis 
to ascertain whether the data are consistent with fish refuges 
during dry-down in addition to ponds, and incorporation of ALFISH
with an individual-based wading bird model. Planned extensions of 
ALFISH include linkage to a dynamic model of lower trophic levels 
that is hydrologically driven and production of a model appropriate 
for estuarine regions.


REFERENCES

DeAngelis, D. L., W. F. Loftus, J. C. Trexler, and R. E. Ulanowicz. 
1997. Modeling fish dynamics and effects of stress in a hydrologically 
pulsed system. J. of Aquatic Ecosystem Stress and Recovery 6:1-13.

Gaff, H., D. L. DeAngelis, L. J. Gross, R. Salinas, and M. Shorrosh. 
1999. A dynamic landscape model for fish in the Everglades and its 
application to restoration. Manuscript submitted to Ecological 
Applications.