Fire is a primary disturbance agent in the Andean cordillera, and strongly influences the Araucaria-Nothofagus forests on the landscape. In Tolhuaca National Park, forests have been shaped by a mixed-severity fire regime over the last several centuries. Changes in land-use practices have influenced the fire regime during that time, including the last half-century of fire exclusion. During the fire season of 2001-2002, over 60% of the park was impacted by wildfire. The main goal of this research is to evaluate the performance of remotely sensed burn severity mapping in this system and assess post-fire Araucaria-Nothofagus regeneration across a burn severity gradient. We retrospectively characterized burn severity using a relative version of the difference in Normalized Burn Ratio (RdNBR) derived from pre and post-fire Landsat imagery. We evaluated the post-fire vegetation response a decade after the fire and established plots along a burn severity gradient to characterize forest regeneration and the response of herbaceous and shrub species. At each of the same plots, we collected data on pre-fire stand structure to evaluate the performance of RdNBR to measure burn severity in the Araucaria-Nothofagus ecosystem and here we report on initial results.
Post-fire canopy mortality ranged from 0 to 100% which underscored the steep burn severity gradient observed on the landscape. The percentage of tree mortality in each plot had the strongest relationship to RdNBR and explained 85% of the variability between the satellite and field data. Shrub cover was significantly lower in moderate and high burn severity plots compared to unburned plots. However, the rhizomatous bamboo Chusquea culueo exhibited dense cover across all ranges of severity, given its ability to quickly resprout post-fire. Although Araucaria and several Nothofagus species are capable of resprouting, we found a substantially higher proportion of regeneration occurred from seed. Araucaria exhibited decreasing regeneration with increasing burn severity. Regeneration density was highest in low burn severity plots and the lowest in high burn severity plots. Nothofagus species did not exhibit the same pattern as the highest density of regeneration was observed in moderate severity plots. Our results suggest RdNBR is an appropriate metric of burn severity in this landscape and Araucaria forest has a great opportunity to regenerate following low severity fire. However, a human-ignited fire in 2015 burned approximately 60% of the same area, highlighting an increased susceptibility of the post-fire vegetation to subsequent fires.