Alternative Future Vegetation Pathways Reveal Potential Transformations of Western <scp>US</scp> Ecosystems

Managing ecosystems in an era of rapid change is inherently challenging not only because of uncertainty in future climate but also due to diverse responses of ecosystems to climate. Projections of ecological transformation alongside information about plausible vegetation trajectories can help land managers explore divergent scenarios and consider how modeled outcomes match their observations. Climate-analog impact models (AIMs) compare environmental information (e.g., vegetation types) between sets of climatically similar locations to infer change and can be used to identify multiple outcomes. We used AIMs to project changes in vegetation across the western United States under a mid-21st century climate scenario, characterize ecological transformation vulnerability based on projection divergence, and demonstrate how AIMs can inform decision-making. We projected high or very high vulnerability to ecological transformation across 29% of the western US, nearly 1 M km². Vulnerability varied among vegetation groups; 75% of alpine vegetation had high or very high vulnerability vs. 6% of desert scrub. We estimate that 9% of the study area faces a high likelihood of transformation based on combined measures of vulnerability and projection agreement. Transformation at the vegetation type (n = 50) level is projected for 40% (1.4 M km²) of the study area, based on primary projections. As vegetation shifts towards types supported by a more arid climate, forested area is expected to contract by 9% and subalpine forests specifically by 54%. Elsewhere, vulnerability is low or trajectories are uncertain, implying opportunities for managers to intervene. Dry forests, for example, could be stabilized through vegetation management and intentional fire use. Our findings suggest likely ecological transformations with significant downstream consequences for ecosystem services and natural resources. They are best used within decision-making frameworks that draw on multiple lines of evidence including local expertise and complementary knowledge systems.