HOME     SCHEDULE     AUTHOR INDEX     SUBJECT INDEX         

PARENT SESSION
Oral Session #16: Landscape Ecology: Fragmentation, ecotones, and models.
Presiding: X. Wu
Monday, August 5. 1:00 PM to 3:45 PM. Gila Meeting Room, TCC.


Uncertainty analysis of a spatially explicit landscape model applied in the Missouri Ozark highlands .

SHANG, BO*,1, HE, HONG1, CROW, THOMAS2, SHIFLEY, STEPHEN3, DIJARK, WILLIAM3, 1 203 ANHEUSER-BUSCH NATURAL RESOURCES BUILDING, COLUMBIA, MISSOURI2 USDA FOREST SERVICE, NORTH CENTRAL RESEARCH STATION, GRAND RAPIDS, MINNESOTA3 USDA FOREST SERVICE, NORTH CENTRAL RESEARCH STATION, COLUMBIA, MISSOURI

ABSTRACT- Uncertainty analysis measures the effects of uncertainty in model inputs on the resulting model predictions. Two common uncertainties in spatial modeling are imprecise spatial location and inaccurate or incomplete site-level information. We applied a spatially explicit, stochastic simulation model, LANDIS, to a real landscape in the Ozark Highlands, Missouri, to evaluate the effects of uncertainties in model input. We explored three scenarios that modified the known data to introduce uncertainties in site level information such as species age structures, spatial location, and both, using 500 year simulations with 10 replicates. Uncertainties in the initial species age structure influenced both fire disturbance sizes and fire return intervals. Initiating the model with a uniform species age structure rather than the known species age structure produced greater variation in simulated age classes and introduced significant differences in the relative area and dominance for all species. However, introduction of a uniform age structure did not greatly affect the average patch perimeter/area ratio for most species. Uncertainties in spatial locations did not significantly affect fire disturbance or species relative area and dominance. But such uncertainties had immediate effects that caused predictions in the spatial pattern of age classes to differ from those for the real landscape for the first 200 years of simulation. These uncertainties introduced greater variation in the simulated spatial patterns, but the differences diminished with increasing simulation years.

KEY WORDS: Uncertainty analysis, LANDIS, Spatially explicit landscape model