A PROBABILISTIC MODEL OF ECOSYSTEM PREVALENCE

We have developed a probabilistic model of ecosystem response to clima te change. This model differs from previous efforts in that it is stat istically estimated using Olson ecosystem and climate data, yielding a joint multivariate probability of prevalence for each ecosystem, give n climatic conditions. We expect this approach to permit simulation of inertia and competition which have, so far, been absent in transfer m odels of continental scale ecosystem response to global change. We sim ulate inertia through estimating the probability of an ecosystem persi sting in situ beyond the prevailing climate conditions (and possibly b eyond the limits usually prescribed in deterministic schema such as Ho ldridge Zones). Conversely, we can observe today that the same environ mental conditions support different ecosystems in different locations. Thus, despite the probability that one ecotype will dominate others a t a given point, others would have the possibility of establishing an early foothold. This is similar to stand models showing changes in bio me composition. We believe these modifications to transfer models will play a significant role in the simulation of the transient response o f ecosystems to climate change.