EcoSim: A Proposal for Null Models Software (SGER)

点击翻译按钮获取中文摘要

This proposal will fund the initial development, testing, and distribution of ecological simulation software (EcoSim). In the study of ecology and evolution, simulations are an important analytical tool. They allow an investigator to ask "what if" questions, and to test patterns against stochastic models. "Null models" are simulations that deliberately exclude a mechanism of interest, and provide benchmark tests for patterns in nature. In spite of their popularity, null models require a good deal of computer expertise to construct. Many biologists without programming skills have been unable to use null models in their work.. For biologists who are also computer programmers, it is still very time-consuming to build, test, and de-bug null models. EcoSim is a set of stand-alone software modules that will satisfy the needs of biologists for null models. EcoSim will be developed specifically for personal computers, and will initially be written for the Windows 95 operating system. Data will be easily imported and exported, providing full compatibility with other graphics and statistics software. Pull-down menus, checkboxes, and a powerful data editor will allow the user to select a variety of simulation options, including power tests. Output will be fully documented with on-screen graphics. The modules and the interface will be programmed in Delphi, a powerful new programming language. The primary audience for EcoSim will be community ecologists and biogeographers. However, the modules will be widely used by many kinds of biologists. College laboratory instructors, demographers, review authors, paleontologists, conservation biologists, environmental consultants, fisheries biologists and many others will find useful tests in EcoSim. The modules will provide analytical tools applicable to both basic and applied ecology. EcoSim will be thoroughly doc umented and tested before its release to the scientific community. EcoSim will be distributed free-of-charge through Internet and a World Wide Web site. EcoSim will be a standalone program that includes on-line help, a user's manual, sample data sets and advice on using null models. Funding from a SGER will allow us to build the user interface and two modules: (1) species diversity and (2) niche overlap. These models are likely to used by the greatest number of ecologists and educators. Future implementations of EcoSim will include modules for (3) body size ratios, (4) co-occurrence, (5) incidence functions, (6) species-area relationships, (7) range overlap, (8) shared species, and (9) cluster formation. A tenth module will allow the user to construct customized null model tests with a minimum of programming. This tenth user-module will greatly expand the utility of EcoSim for problems in population genetics, demography, and behavior. EcoSim has the potential to affect a large number of ecologists by providing them with analytic tools. A similar problem in the history of cladistic analysis points to the importance of software availability on the development and infrastructure of science.