ES-03. Western US forest vegetation recovery under compound disturbances

Abstract
Increased frequency and severity of disturbances (e.g., fire, insect and pathogen infestations, drought, regional warming) and the interactions among them over past decades resulted in 6.3 billion dead trees in western US forests. However, it is unclear how these dead forest stands recover; especially, whether these forests undergo state changes or show resilience to these disturbances. Forest composition, structure, and functional types depend on post-disturbance forest recovery and are important indicators of forest state change and resilience. We focused on three ecoregions: Southern Rockies (SR), Northern Rockies (NR), and Pacific Northwest (PNW) and used a wealth of multi-scale observations from individual trees to ecoregions collected from field inventories, unmanned aerial systems (UAS) , the NEON airborne observation platform, and satellites (Landsat and GEDI) to answer; (1) How does postfire carbon recovery trajectory vary as a function of fire severity, regional climatic stress (drought occurrence) and site characteristics (elevation) in western U.S conifer forests? (2) What types, combinations, and sizes of disturbances lead to abrupt vegetation state change, and 3) What is the carbon potential in western forests under these frequent and severe disturbances? GEDI based canopy height analysis shows that all three ecoregions demonstrate an initial decline of vegetation during the first 9-25 years postfire followed by a gain of vegetation. Canopy height in the SR fully recovers to the unburned background state within the first 3 decades with a mean recovery rate of 0.39 my-1 while recovery in the NR and PNW may take up to 5 decades or longer if fire returns within this period. The PNW exhibits the slowest recovery rate (0.19 my-1). Our study further shows that after a fire, both Pacific Northwest and Northern Rockies show a loss of canopy volume of ~4.4 - 10% while the Southern Rockies gains more than 20% of canopy volume compared to the unburned background state. Time since fire, fire severity, and elevation were identified as the most significant drivers of postfire vegetation recovery, likely because it integrates the distance to seed source, vegetation composition, and the local climate (e.g., temperature). While these initial results advance our understanding of how forest recovery varies as a function of fire, inclusion of compound, large disturbances integrated with multi-scale data within and among ecoregions will answer essential macrosystems ecology questions in western U.S. forests.