Abstract
Satellite remotely sensed multispectral data provides a systematic, synoptic means for broad-area spatially-explicit estimation of biophysically important variables. By using ground measurements of biotic properties to calibrate remotely sensed multispectral data, vegetation properties measured at sample points can be extrapolated across a large geographic region (Graetz, 1990). Biophysical variables derived by this empirical method may include the successional state of the vegetation, or an intrinsic property of the vegetation, such as biomass, leaf area index, cover, or moisture content (Jensen, 1983; Waring et al., 1986; Spanner et al., 1984; Graetz, 1990). Spatially-explicit estimation of forest biophysical factors at landscape to regional scales has applications in forest management and ecology, including insect infestation susceptability, forest fire behavior, and estimating plant and animal species habitat and diveristy. Our previous research examined relationships -- between forest structure, successional state, and spectral reflectance characteristics. Results indicated that decreases in visible and middle-infrared spectral reflectance are related to the age and development of a coniferous forest stand. Spectral reflectance changes are rapid during the initial stages of stand regeneration, but the rate of change slows as the stand progresses into later successional stages (Blodgett and Jakubauskas, 1996; Jakubauskas, 1996). Our ongoing objectives are to develop methods for estimation of forest biophysical parameters from satellite remotely sensed data and to compare Yellowstone and Teton coniferous forests in terms of forest structure and successional pattern. Forest stands sampled in 1995 in Grand Teton National Park are 500 - 1000 meters lower in elevation than the Yellowstone sites (sampled 1992-1994), and subject to different temperature and precipitation regimes. Sampling in 1995 was directed at increasing our database of lodgepole-dominated forest stands in the Greater Yellowstone Area.