LINKING CONTEMPORARY VEGETATION MODELS WITH SPATIALLY EXPLICIT ANIMALPOPULATION-MODELS

Spatially explicit models for animal populations (SEPMs) necessarily e mbody assumptions about plant community structure and dynamics. This p aper explores the advantages and limitations of directly linking anima l SEPMs with models for vegetation dynamics. Such linkages may often b e unnecessary. For instance, in research focussed on questions with sh ort time horizons, the spatial patterning of vegetation can be reasona bly approximated as a tired landscape templet for animal population dy namics. But if one needs to consider longer time scales (e.g., decades to centuries), landscapes will be dynamic. Models of vegetation dynam ics provide useful tools for predicting landscape dynamics. We outline the sorts of output from vegetation models that might be useful in an imal SEPMs. We discuss as a concrete example recent forest simulators, which predict with reasonable accuracy some variables (e.g., tree spe cies composition), but which, to date, are quite poor for others (e.g. , seed production). Moreover, because vegetation models target a restr icted range of temporal and spatial scales, they may be more useful fo r certain consumer groups than for others. Despite these cautionary ob servations, we believe that the time is ripe for fruitful linkages bet ween models of vegetation dynamics and animal SEPMs.