OOS 17-3 - Measuring and modeling post-fire forest trajectories in northwestern Cascadia

Predicting forest dynamics under changing climate and fire activity remains an important but challenging task. While fire effects in fuel-limited systems are relatively well-studied, critical insights are missing in forests with infrequent, climate-limited fire regimes. Moist coniferous forests west of the Cascade Mountain Crest in Washington and northern Oregon, USA (‘northwestern Cascadia’) are among the most productive in the world, providing important ecological, cultural, and economic services. These forests are characterized by infrequent (200+ year fire return intervals) stand-replacing fires limited by climate and weather conditions. Yet, future projections of vegetation and fire risk for northwestern Cascadia forests remain uncertain due to limited contemporary fire events and prolonged successional dynamics. Within the last decade, fires have burned approximately a quarter million hectares across the region, presenting an opportunity to study forest recovery following fire and plan climate adaptation strategies for guiding forest management into the future. Here, we leveraged a long-term field plot network and simulation modeling to analyze post-fire vegetation trajectories and reburn risk in recently burned (2017–2020) stands across northwestern Cascadia, asking: (1) How are tree regeneration, early seral pre-forest conditions, and fuel profiles following recent fires affected by pre-fire stand age and burn severity? (2) When and under what settings do post-fire vegetation trajectories support opportunities for management interventions to promote desired outcomes? Within each stand, we measured structure, regeneration, understory vegetation, and surface and canopy fuel loads in 1-ha plots established 2–5 years post-fire. Plots were stratified by pre-fire stand age (young, mid-seral, late-seral structural classes) and burn severity (unburned; low, <30% mortality; stand replacing, ≥90% mortality). For each stand, we used the individual-based forest landscape and disturbance model, iLand, to simulate post-fire vegetation development for ~75 years (ca. 2100). We compared mean and variability in model outputs across strata, and used the range of simulated conditions to identify the potential timing and location of opportunities for implementing different management scenarios defined by our agency and Tribal partners. Preliminary results suggest pre-fire stand structure is a dominant control on post-fire outcomes in northwestern Cascadia forests, with older stands recovering forest dominance more quickly and having greater fuel loads than younger stands. Conversely, more severely burned stands had longer tree regeneration timelines and more variable fuel load responses. These findings contribute to fundamental understanding of post-fire forest trajectories in northwestern Cascadia and will help support adaptive management decisions through improved regional forest projections.