Remotely sensed multispectral data collected from satellites provide a systematic, synoptic ability to assess conditions over large areas on a regular basis. Early use of this satellite data for land cover mapping was based on spectral differences of cover types, with little integration of ancillary data such as soils or topographic information (Iverson et al. 1990). In recent years, concurrent with trends toward integrating remotely sensed and ancillary data for improved classification accuracy (Cibula and Nyquist 1987; Frank 1988), there has been increasing interest in utilizing remotely sensed data for extracting biophysically important variables, relating observed spectral reflectance to leaf area index, biomass, net primary productivity, and vegetation moisture content (Waring et al. 1986; Hobbs and Mooney 1990). The concept of using remotely sensed spectral data to map and monitor the progress of succession within forests and other environments has not been extensively explored. However, the capability to map and predict successional stages of forest habitat types on a landscape to regional scale has important implications for animal habitat management, assessment of insect infestation susceptibility, prediction of fire behavior, and evaluation of plant and animal species diversity. Ecological models based on established successional change rates and trends permit the prediction of future environmental conditions, landscape patterns, and the propagation and effects of disturbances across these landscapes (Hall et al. 1988; Romme 1982). Despain (1990) provides two examples where information on habitat and cover types is important for park management purposes: the cumulative effects model for grizzly bears; and the prediction, assessment, and management of mountain pine beetle outbreaks in conifer forests. Accurate mapping of habitat and cover types can provide information on the distribution and pattern of specific plant communities important to animal species for food, cover, and breeding ground (Knight and Wallace 1989). The ability to map and predict successional stages of forest habitat types has implications for prediction of fire behavior and spread. Previous studies (Despain 1990; Romme and Despain 1989; Romme 1982; Taylor 1969) have noted the relationship between forest age and fire susceptibility. Older stands are comparatively more flammable than younger stands due to fuel accumulations on the ground and in the canopy, and have a higher propensity to propagate and sustain extensive crown fires. Spatial patterns of cover types may also be important, with a highly fragmented landscape mosaic providing natural firebreaks under typical weather conditions. Consequently, as Despain (1990) has noted, the ability to map forest habitat and cover types is of importance for estimation of fire intensity and spread. The use of a single habitat type provides a logical unit for environmental stratification of the study site. Since a habitat type integrates vegetation, climate, topography, and soils (Pfister and Amo 1980), using a single habitat type forces a restriction to selective ranges in climate, topography, and soils types. These constrictions will minimize the effects of abiotic variation on the recorded spectral reflectance, allowing analysis of spectral variation to be concentrated on the changes in biotic factors associated with succession.
Using remotely sensed spectral reflectance to indicate leaf photosynthetic efficiency derived from active fluorescence measurements
