| The foothills and mountains of southern California support fire-prone shrublands that are adapted to the summer drought and winter rains of the Mediterranean-type climate. Humans have altered the region's natural fire regime through increased ignitions and fire suppression, and the magnitude and direction of those effects varies across the landscape. In low-elevation coastal areas, ignitions at the wildland urban interface have increased fire frequency beyond its natural range of variability. Also, urban growth threatens the high biodiversity in the region through habitat loss and fragmentation. This research used an integrated simulation modeling strategy to evaluate the long-term consequences of altered fire regimes and urban growth on the composition and distribution of native shrublands in a study area administered by the National Park Service. First, an urban growth model (UGM) was calibrated to predict the effects of future development on habitat loss and spatial pattern from 2000--2050. Due to the steep terrain in the region, three scenarios were compared with development prohibited beyond 25%, 30%, and 60% slope. Next, a spatially explicit simulation model of landscape disturbance and succession (LANDIS) was used to predict the effects of high fire frequency on dominant plant functional types. Because human settlement is the primary driver of increased ignitions, the UGM was integrated with LANDIS to evaluate the combined effects of urban development and high fire frequency. The UGM predicted that urban area would increase from 11% of the landscape in 2000 to 26%, 35%, and 47% in 2050, respectively, for the three management scenarios. The spatial pattern of the vegetation became highly disconnected when development was allowed up to 60% slope. The LANDIS simulations predicted that shrubs dependent on fire-cued seed germination were most sensitive to high fire frequency and could potentially convert to other vegetation types under short fire return intervals. In the integrated model runs, fire frequency did not increase as expected because development patterns were aggregated and thus did not substantially increase the wildland urban interface where ignitions were more likely to occur. However, integrating the models contributed to more realism in the simulations than using the models separately. |
