COS 218-3 - Incorporating topographical information improves simulated patterns of post-fire conifer regeneration in the Southwest U.S.

Abstract: The western U.S. is expected to experience hotter and drier climate, leading to increasing frequency and intensity of wildfire in the future. Such wildfire favorable conditions are resulting in tree mortality and failed post-fire seedling regeneration in forest ecosystems. Despite strong relationships between topography and plant regeneration, vegetation processes in ecosystem models have not fully addressed topography-mediated effects on the probability of plant regeneration or only considering climate-related factors such as radiations and moisture conditions. Our objectives for this study were to modify the regeneration algorithm in the PnET succession extension in the LANDIS-II model and examining how the modification alters patterns of plant regeneration under future climate conditions. We used a seedling survival data based on a planting experiment in the footprint of the 2011 Las Conchas Fire in northern New Mexico to modify the function of the probability of plant regeneration and simulated the Las Conchas Fire landscape for 2012-2099 with observed and projected climate data (e.g., RCP 4.5 and 8.5). By adding topographical and additional regional climatic variables such as heat load index (HLI), ground slope and spring precipitation, plant regenerations of three common southwestern conifer tree species (piñon, ponderosa pine and Douglas-fir) significantly decreased, leading to significantly lower aboveground biomass regardless of climate scenario. The new algorithm altered spatial patterns, including decreases in regeneration at higher elevation and increases in regeneration at lower elevation relative to the original algorithm. In addition, we found the new algorithm decreased regeneration on eastern aspects. Our findings indicate that ecosystem models can better capture tree regeneration following wildfire by incorporating topographical information into processes determining regeneration probability. This information will improve model utility for projecting forest ecosystems response to the combined effects of climate change and wildfire.