Print PageThere is significant potential for widespread and rapid restructuring of habitats across Southwestern U.S. landscapes over the next several decades due to climate change. A key question is how the vegetation of ecological systems (after Comer et al. 2003) across the region will respond in terms of species composition as well as geographically. However, to understand and project likely responses at the scale of landscapes it is essential to first understand the extent to which certain climate variables influence the composition of particular ecological systems, each of which respond to different combinations of climate variables. In this study, a large number of vegetation field plots were used to sample ecological systems. Twenty different bioclimate variables, such as growing season vapor pressure deficit or mean temperature of the wettest quarter, were extracted from existing climate data at the field plot localities. With these data, within each ecological system, a principal components analysis was used to identify the minimum set of least correlated bioclimate variables (usually 1-4) that explain the largest proportion of variability among all bioclimate variables. Then a nonmetric multidimensional scaling (NMDS) procedure was applied to the field plot species composition data, yielding 2-3 indices of compositional similarity among the field plots. Finally, a multivariate analysis of variance was applied using the set of bioclimate factors as the independent variables and the NMDS indices as the response variables.
Results/Conclusions
The vegetation composition of individual ecological systems respond to significantly different sets of bioclimate variables. For example, both temperature seasonality and mean temperature of wettest quarter are the major bioclimate factors for the vegetation of the Sonora-Mojave creosotebush-white bursage desert scrub ecological system. Vegetation of the Rocky Mountain aspen woodland ecological system responds largely to the amount of precipitation in the warmest quarter. In climate terms, the minimum temperature of the coldest month and isothermality (the range in diurnal temperature normalized by the overall range in temperature) best explain the vegetation composition of the Colorado Plateau Mixed Bedrock Canyon and Tableland ecological system. To some extent all systems responded to combinations of bioclimate variables but often just one or two variables dominated. Knowing which climate variables best explain the vegetation composition of these systems provides land managers with specific climate/vegetation monitoring parameters for different ecological systems. Additional work is needed to better establish the compositional directions of change in time and space along identified bioclimate gradients.
