Research Abstract.--Quantitative relationships between ecological processes and landscape patterns will provide land managers insight into how certain wildlife species respond to landscape changes under various disturbance regimes. Understanding these relationships is an essential component of ecosystem management and is required by land managers to maintain and conserve dynamic ecosystems at various temporal and spatial scales. By describing the expected range of natural variation (RNV) in habitat conditions over an extended period of time, land managers are better positioned to identify land management practices that may cause deviations from a species’ RNV. The overall purpose of this study was to quantify the spatial and temporal dynamics of wildlife habitat under the effects of a natural fire disturbance regime in southwestern Colorado. We addressed three questions.

1. Wwhat is the spatial and temporal variation in suitable habitat under a natural fire regime?
2. Under the same disturbance regime, how do habitat dynamics differ among species with different life history strategies and habitat associations?
3. Wwithin a constantly changing landscape, are there relatively stable areas of suitable habitat?

For this study, we developed a stochastic disturbance-succession landscape simulation model (RMLANDS), and spatially-explicit, habitat suitability models for four species: pine marten (Martes americana), elk (Cervus elaphus nelsoni), three-toed woodpecker (Picoides tridactylus), and olive-sided flycatcher (Contopus borealis). We simulated changes in landscape structure resulting from natural fire disturbance and succession processes and determined habitat conditions at 30-year intervals over a 600-year period for a 228,500 ha mountainous area located in the San Juan National Forest (SJNF). We quantified the amount and configuration of each species’ suitable habitat using several landscape metrics and explored the independent and combined effects of habitat area and configuration on habitat dynamics. Based on these results, we described a RNV in habitat area and configuration for each species.

Overall, our results suggest that natural landscape processes induce changes in species habitat amount and configuration, thereby having a potentially strong influence on species’ populations within dynamic landscapes. First, for each species, habitat was dynamic over time and exhibited a shifting habitat mosaic. However, each species’ RNV was modest and occurred within a narrowly-bounded equilibrium state at the scale of our study area. Habitat configuration was more variable over time than habitat area for all species. Second, the same landscape changes resulted in significant differences in habitat dynamics among species. Habitat dynamics for generalist species were more variable over time than for habitat specialists. Lastly, the temporal stability of habitat patches was variable across the landscape depending on the position a habitat patch occupied on the landscape, fire disturbance behavior, and species habitat associations. However, despite habitat dynamics, there was a small percentage of habitat that remained relatively stable over time. Relatively stable habitats may provide conditions to support source populations within a source-sink metapopulation model.