Temporal patterns of ecosystem processes on simulated landscapes in Glacier National Park,Montana, USA

The mechanistic, spatially-explicit fire succession model, Fire-BGC (a Fire BioGeoChemical succession model) was used to investigate long-term trends in landscape pattern under historical and future fire regimes and present and future climate regimes for two 46000 ha landscapes in Glacier National Park, Montana, USA. Fire-BGC has two spatial and temporal resolutions in the simulation architecture where ecological processes that act at a landscape level, such as fire, are simulated annually from information contained in spatial data layers, while stand-level processes such as photosynthesis, transpiration, and decomposition are simulated both daily and annually. Fire is spread across the landscape using the FARSITE fire growth model and subsequent fire effects are simulated at the stand-level. Fire-BGC was used to simulate changes in landscape pattern over 250 years under four scenarios: (1) complete fire exclusion under current climate, (2) historical wildfire occurrence and current climate, (3) complete fire exclusion under a possible future climate, (4) future wildfire occurrence and future climate. Simulated maps of dominant tree species, aboveground standing crop, leaf area index, and net primary productivity (NPP) were contrasted across scenarios using the metrics of patch density, edge density, evenness, contagion, and interspersion. Simulation results indicate that fire influences landscape pattern metrics more that climate alone by creating more diverse, fragmented, and disconnected landscapes. Fires were more frequent, larger, and more intense under a future climate regime. Landscape metrics showed different trends for the process-based NPP map when compared to the cover type map. It may be important to augment landscape analyses with process-based layers as well as structural and compositional layers.