Quantifying Spatiotemporal Post-Disturbance Recovery Using Field Inventory, Tree Growth, and Remote Sensing

Abstract

Forest recovery following a disturbance lasts decades to centuries, and the rate depends on pre- and post-disturbance condition and local environmental factors. Existing approaches of field observations, remote sensing, statistical chronosequence, and ecological modeling have one or more drawbacks, including short time frames, generalized details, indirect indicators, hard parameterization, and defective assumptions. Using aboveground live biomass (AGLB) as an example, we developed an approach called “Disturbance and Recovery Assessment across Space and Time (DRAST).” For a specific post-disturbance year, DRAST utilizes Field Inventory and Analysis data sets and the Forest Vegetation Simulator, as well as pre- and post-disturbance remote sensing to create two rasters: (1) what the AGLB would look like over the disturbed area had the disturbance not occurred and (2) what the AGLB would look like over the disturbed area in the actual presence of the disturbance. These two rasters are compared annually to examine the spatiotemporal recovery pattern. We demonstrated DRAST with the 2013 Rim fire in California, United States, by creating two sets of AGLB for 100 years. Our results showed that (1) the AGLB consumed by Rim fire was 3.52 Tg and (2) 45.9% of the burned area needs <5 years to recover, followed by 6.4% (5–10 years), 6.1% (>95 years), 5.9% (10–15 years), 5.4% (15–20 years), 4.8% (20–25 years), and 4.3% (25–30 years). In conclusion, DRAST can provide spatially explicit and highly detailed ecological indicators for decades under the two scenarios of “no disturbance” and “actual disturbance occurrence” for recovery analysis.