Colorado Landcarbon: Accounting for Wildfire

Disturbance disrupts the balance between gross primary productivity and respiration, resulting in a net C loss for some time after a stand-replacing fire. However, our understanding of this process is based on a limited number of studies. Ecosystem C recovery post-fire must be explicitly and carefully examined in order to generate accurate predictions of C cycle impacts of future wildfires and change in fire regimes. Montane ponderosa and lodgepole pine forests, either single-species stands or mixed, dominate surface area in the Southern Rockies. These species have drastically different relationships with wildfire; the current narrative portrays ponderosa pine as accustomed to low-severity surface fires with low regeneration rates following high-severity wildfire, whereas lodgepole pine forests readily regenerate after a high-severity stand-replacing wildfire. Forests at the transition between lower montane and upper montane may be more sensitive to future climate change than their lower counterparts; e.g., a stand-replacing disturbance could cause montane ponderosa pine forests to yield to lodgepole pine. It is important to understand how wildfire impacts ecosystem C fluxes in these ecosystems and how landscape dynamics, including topographical changes in climate and distance from forest seed source, can be used to predict C cycle responses to future wildfire patterns. To date, no single study has collected data at an adequate temporal resolution to fully characterize the short-term, intermediate-term, and long-term response and recovery of forest soil respiration to pre-burn conditions. The aim of this work is to predict soil respiration and net primary productivity in pine forests of the southern Rocky Mountains based on time since fire, fire severity, forest type, and forest and soil properties, such as tree basal area, leaf area index and soil carbon pools. We sampled 5 wildfires and 1 high-severity prescribed fire as well as nearby unburned reference forests. The following time-since-fire intervals were sampled along a 30-yr chronosequence: 1-5 years (n=1), 5-10 years (n=1), 10-20 years (n=3), and 20-25 years (n=1).