Remote Sensing of Vegetation Responses During the First 20 Years Following the 1980 Eruption of Mount St. Helens: A Spatially and Temporally Stratified Analysis

Abstract

The variety of disturbance mechanisms involved in the 1980 eruption of Mount St. Helens (e.g., heat, burial, and impact force) and the resulting diversity of vegetation responses have provided abundant opportunities for disturbance-zone-specific research (Frenzen 1992; Frenzen et al. 1994). As evidenced by the research reported in this volume, tremendous amounts of knowledge can be acquired from studies that focus on vegetation responses within individual disturbance zones, such as the debris-avalanche deposit. As the responses to the eruption continue to develop, however, it becomes increasingly important to understand the larger context for specific study sites: What responses are common among disturbance zones? And what responses are distinctive to certain disturbance mechanisms and sites? Can the lessons from a local area be generalized throughout the disturbed area or even throughout a zone dominated by one disturbance type? These questions can be addressed in at least two ways: r Studies at different sites can be compared for similar or divergent processes and states. r Spatial-analysis tools can be used to evaluate response patterns across the disturbed area. The first approach can be achieved through a comparative analysis of the disturbance-zone-specific studies reported in this volume and elsewhere. In this chapter, vegetation-response patterns present across disturbance zones are evaluated for, among other things, consistency with the ground-based observations and analyses available from other studies. It is important in studying the first 20 years of vegetation response across the disturbed landscape at Mount St. Helens that the analysis be stratified, both spatially and temporally. Previous research indicated that early vegetation responses were largely dependent on the variable effects of disturbance mechanisms (Adams et al. 1987; Lawrence and Ripple 2000) as well as the substrates resulting from the volcanic events (del Moral and Clampitt 1985). Areas that had substantial biotic legacies (buried seeds, sprouting roots, or downed woody debris), for example, had much faster early vegetation development than areas dominated by primary successional processes , although posteruption management practices might be at least equally important (Figure 8.1). Within these broad categories , additional distinctions might be made based, for example , on the nature of the volcanic deposits (e.g., pumice deposits versus mudflow remnants). In addition, the area devastated by the 1980 eruption has experienced three distinctive posteruption management regimes (Franklin et al. 1988): (1) Mount St. Helens National Volcanic Monument (Monument), where natural processes of ecological recovery were allowed to dominate after its establishment in 1982 (although seeding with nonnative plants and salvage logging occurred in limited areas before that time); (2) Gifford Pinchot National Forest (GPNF) outside the Monument, which experienced salvage logging that was substantially completed by 1984 and then planted with commercial conifer species, mainly Douglas-fir [Pseudotsuga menziesii (Mirbel) Franco] seedlings; and (3) private forestlands, where portions were seeded with non-native grasses and legumes, salvage logging was completed by 1982, and commercial conifer species were planted, also mainly Douglas-fir seedlings, although limited areas were left unplanted. Finally, spatial stratification by disturbance type and management regime enables important comparisons between analyses across disturbance zones and many of the analyses that provide more detail regarding individual disturbance zones. Temporal stratification enables the analysis of changes in factors driving vegetation response over time. Early responses were likely most affected by factors relating to growth of survivors and success of colonizers, while later responses might have been affected more by factors relating to growth rates, both of individual plants and of established colonies through reproduction. Further, the timing of any such shifts in the relative influence of important factors could have depended on disturbance mechanisms or management practices. The purpose of this study was to examine patterns of vegetation responses relative to selected ecological driving factors within each disturbance zone resulting from the 1980 eruption 111 Distance to surviving vegetation was most important factor in predicting debris avalanche cover, in first and second seven growing season period after the eruption. Avalanche unit and slope were initially important predictors of cover but dropped off in later time periods as elevation became more important as a predictor. By the third period, elevation was the most important predictor and distance from surviving vegetation was the second predictor.. Overall predictive power of the model declined from 40% to 22 percent over the time period of the study. Distance from crater was an important factor in all time periods.