This research project will measure, describe, and model change in ecological patterns and processes over large geographical areas in ways that will test the utility of the landscape approach for analyzing and predicting changes in ecological processes (i.e. productivity, succession), for optimizing resource management, and for evaluating human impacts. The work will help establish guidelines for the development and testing of spatial ecosystem models by testing the relative performance of different spatial simulation techniques and by combining new computer technologies with data-rich coastal study sites. Approaches that can predict the way landscape patterns change are crucial to the development of landscape ecology, yet no coherent predictive tools have emerged. The investigators will synthesize current ecological data and gather new data for incorporation into spatial models that predict change in landscape patterns. The theories embodied within a spectrum of spatial models, ranging from process-based simulation to transition probability models, will be tested using a system of descriptive statistics and pattern matching techniques which will be developed and evaluated as part of this research. Models will be developed for three different coastal landscapes, for which much historical data and ongoing long term research already exists. Results will provide: (1) increased understanding of the processes controlling changes in landscapes; (2) principles for adjusting spatial and temporal scales to optimize predictability in models; (3) comparison of landscape data sets; (4) principles for determining the optimal type of model for a particular set of objectives, scale and resolution; and (5) new methods for examining the goodness-of-fit between predictions and data that re appropriate for spatial ecological data and which require a degree of spatial pattern recognition. Finally, to the extent possible, the project will attempt to synthesize the principles and methods into a generalized theory for predicting landscape dynamics. The investigators are innovative in their approach. Institutional support and facilities are excellent. Results from the work should be of significance to the establishment of new basic ecological research directions as well as to the areas of land use planning and resource management.

Agency
National Science Foundation (NSF)
Institute
Division of Environmental Biology (DEB)
Application #
8906269
Program Officer
James R. Gosz
Project Start
Project End
Budget Start
1989-07-15
Budget End
1992-12-31
Support Year
Fiscal Year
1989
Total Cost
$263,706
Indirect Cost
Name
University of Maryland Center for Environmental Sciences
Department
Type
DUNS #
City
Cambridge
State
MD
Country
United States
Zip Code
21613