Design and Estimation for Investigating the Dynamics of Natural Resources
Nusser SM, Breidt FJ, Fuller WA, Department of Statistics, Iowa State University. 1998. Design and Estimation for Investigating the Dynamics of Natural Resources. Ecological Applications. 8(2):234-245
Federal agencies, policy makers, and scientists have long been interested in monitoring natural resources and environmental conditions on a national and regional scale. One of the main objectives of these studies is to estimate temporal changes in the extent and condition of natural resources. In its simplest form, temporal change can be defined as the difference between population parameter values at two time points for a given population. A more complex investigation of change in an ecological system involves studying the underlying dynamics that produce an observed net change. We discuss the general problem of sample design and statistical estimation to support investigations of the dynamics of change in ecological systems, particularly when a limited number of temporal observations are available. We focus on large-scale natural resource monitoring surveys through the example provided by the National Resources Inventory (NRI), a longitudinal survey conducted by the U.S. Department of Agriculture (USDA). Sample design, data collection, and statistical methods for constructing an accessible database are outlined, with emphasis on features that support investigations concerned with temporal dynamics. An example from the 1992 NRI is presented to illustrate methods for investigating temporal changes in land use in relation to observed changes in erosion rates over time. Finally, we discuss how statistical methods developed for the NRI program can be applied more broadly to environmental monitoring studies.
- Type
- Journal Article
- Authors
- Nusser, S; Breidt, F; Fuller, W; Department of Statistics, Iowa State University
- Units
- NRIM
- Keywords
- Database/information systems, Environmental Monitoring, environmental statistics, multi-stage design, Survey sampling, two-phase estimation