Abstract

1. Cape Sable Seaside Sparrow (Ammodramus maritimus mirabilis) Predicted Population Size Environmental Modification Fire Succession Population Object SIMSPAR Flow Diagram Flooding Succession Sparrow Object Landscape Water File Output Vegetation Elevation Input Files Status of Transition Sparrow Distribution Map Demographical Analysis of the Spatially-Explicit Individual-Based Cape Sable Seaside Sparrow Model The basic coding has been completed and the process of validation is underway. This has lead to the examination of results of the model at a population level and an individual level to insure parity of results between the the two versions. The model based on the new code has shown a significant performance improvement over the old code, thereby allowing calculations to be completed in hours that were requiring days. ERIC CARR and LOUIS J. GROSS The Institute for Environmental Modeling, Department of Ecology and Evolutionary Biology University of Tennessee, Knoxville, TN 37996-1610 Current Topographic Range This poster is based in part on: Nott P, Effects of Abiotic Factors on Population Dynamics of the Cape Sable Seaside Sparrow and Continental Patterns of Herpetological Species Richness: An Appropriately Scaled Landscape Approach [dissertation] .Knoxville(TN) : University of Tennessee; 1998) Abstract SIMSPAR (see P. Nott, 1998) is a spatially-explicit individual-based model for the Cape Sable Seaside Sparrow (Ammodramus maritimus mirabilis), an endangered species present in the Florida Everglades. It was developed as an evaluation and management tool for comparing the relative impacts of hydrological policies on the Cape Sable Seaside Sparrow, and parameterized for the western sub-population. As part of the model development, SIMSPAR is being migrated to an object oriented program to increase model run time performance, enhance output flexibility, allow for possible expansion to other subpopulations, and more fully integrate with other ATLSS component models. The model changes will allow for direct analysis of the life history of each individual model sparrow, from which one can draw implications about the relative impacts of various environmental changes on population-level responses. For example, complete family trees can be extracted for demographical analysis and for possible extension to analysis of pedigrees for inclusion of genetics. This allows the examination of the derived life history parameters of the model for comparison to data and for use in sensitivity and uncertainty analysis. Notably, net reproductive rates, survivorship schedules, and causes of fledgling death are all statistics that can be explored. A comparison of these demographic parameters can then be made across various hydrological scenarios and models. Output Flexibility Major Object Structure of the Model Individual sparrow information contained within the sparrow object can be readily outputted to a file from within the code. Each sparrow is given a unique ID which allows the storage of who its parents and children. From this data, the genealogical structure of the entire population over the thirty year period can be reconstructed during post simulation analysis. This will allow for detailed examination between various scenarios for hydrological comparison or sensitivity analysis. Since the population is always current, snapshots can be taken at any time to capture the state of the population for comparison. Overall,this approach will facilitate examining the data during individual iteration while maintaining the averaging over the entire monte carlo set. The performance increase will also allow for an increase in the number of iterations that can be implemented by the monte carlo routine. The new software architecture is based on two primary classes or types, a population class and a sparrow class. In the model, each individual sparrow is represented by a single sparrow object, which is a specific instance of the sparrow class. The sparrow class defines all behavior and data associated with the sparrow object and the allowed actions that can change the object. By packaging the data in a single unit, the information is able to be kept continuously updated in such a way that the state of a sparrow is never undefined. The class structure also facilitates the outputting of data due to the centralization and structure associated with the object. The population class organizes the individual sparrow objects within a single defined population. The population object provides a command and control function for the sparrow objects it contains while performing the necessary interaction with the underlying landscape. Goals of Code Migration of the Sparrow Models Basic Objects and Structure • Increase the run time performance of the model. • Increase flexibility in retrieval of data for output. • Provide uniformity of ATLSS code to allow future integration with other ATLSS Models. • Compatibility with the existing landscape structures to allow future expansion of range and dynamics of the model. • Expansion of model to larger topographic range.