Forests in the western United States are facing a significant threat from climate change, leading to increased drought intensity, fire frequency, and burn severity. These factors are intricately linked and primarily driven by anthropogenic greenhouse gas emissions. The historical fire deficit has resulted in fuel accumulation, exacerbating the severity of wildfires. While historical climate changes have caused vegetation shifts, as evidenced by pollen analyses, the current rate of warming is unprecedented. The rapid pace of change may not allow forest communities sufficient time to adapt, potentially resulting in a climate-vegetation mismatch. Changes in dominant PFTs are expected to be driven by increased fires, persistent droughts, and warming. The response of different PFTs will depend on their physiological tolerances, phenology, fire adaptation, and competitive abilities. Although historical fires have mediated forest responses to climate change through species turnover or selection of fire-adapted species, the current high-severity, stand-replacing fires lack historical precedent. These fires clear forests and remove seed sources, potentially allowing grasses and shrubs to expand into previously forested areas. Future climate projections indicate continued warming and increased droughts, which will further stress younger seedlings and promote PFT alteration. Some models project a complete transition from forest to shrubland or grassland, while others suggest only shifts in dominant forest species. By the end of the century, a substantial portion of the western US landscape is projected to have climates incompatible with existing vegetation communities. This study addresses this critical issue by integrating a biogeography module into a biogeochemical model to assess the impact of climate change on dominant PFTs in the forested regions of the western US, focusing on the potential shifts under RCP 4.5 and 8.5 scenarios.

Previous research has highlighted the potential for significant vegetation shifts in the western United States due to climate change. Studies utilizing pollen analysis have demonstrated historical vegetation changes in response to climatic shifts, while others have shown individualistic species responses to gradual warming during the post-glacial period. However, the current rapid rate of warming, coupled with increased wildfire frequency and severity, is unlike past warming trends. This rapid change poses a challenge for forest adaptation, potentially leading to climate-vegetation mismatches. Some studies project a complete transition from forest to shrubland or grassland, while others suggest more nuanced changes in forest species composition. The consensus across studies emphasizes the significant role of increased fire frequency and severity, along with drought, in driving these changes. These studies provide the context for the current research which aims to quantitatively model the extent and consequences of these potential vegetation shifts.

This study utilized a coupled biogeographic and biogeochemical model to assess the impact of climate change on dominant plant functional types (PFTs) in the western United States. The biogeographic module, based on BIOME4, simulates the equilibrium distribution of dominant PFTs, while the biogeochemical model (TEM) tracks carbon, nitrogen, and water flow within the ecosystem. The model incorporates various factors, including climate data (temperature, precipitation, solar radiation, etc.), soil properties, fire dynamics (fuel availability, ignition, combustibility, spread), and PFT competition (based on Net Primary Productivity (NPP)). The model was run at a 0.5° x 0.5° spatial resolution. For the historical period (1984-2014), the model was calibrated using observed carbon fluxes and stocks. Future simulations (2015-2100) were conducted under RCP 4.5 and RCP 8.5 climate scenarios, utilizing downscaled climate projections. Twenty model runs were performed for each RCP scenario, and results were aggregated to account for model uncertainty. Key model outputs included changes in dominant PFTs, Net Primary Productivity (NPP), Net Ecosystem Productivity (NEP), Net Carbon Exchange (NCE), vegetation carbon, and soil organic carbon. Principal Component Analysis was used to assess the relative contributions of temperature, moisture stress, and fire to the shifts in dominant PFTs. The model considered various PFTs, including different forest types (e.g., boreal, mixed temperate, temperate coniferous, temperate deciduous), short grasslands, arid shrublands, xeromorphic forests and woodlands, and temperate broadleaved evergreen forests. The historical fire regime was incorporated into the model using data from LANDFIRE, accounting for fire suppression periods and fire return intervals. The model assumes sufficient seed sources of all PFTs are present after fires, though acknowledging the limitation of this assumption in reality.

The model predicted substantial shifts in dominant PFTs in the western United States by the end of the century under both RCP 4.5 and RCP 8.5 scenarios. Under RCP 4.5, 40% of areas originally dominated by trees transitioned to shrubland (7%) or grassland (32%), while under RCP 8.5, 58% transitioned to shrubland (18%) or grassland (40%). These transitions were most pronounced in areas projected to experience the greatest temperature increases and decreased soil moisture, often associated with severe, stand-replacing fires. Principal Component Analysis indicated that moisture stress was the primary driver of these shifts (55% of variance under RCP 4.5 and 53% under RCP 8.5), followed by fire (27% and 26% respectively) and temperature. Geographically, the most significant shifts to grasslands were observed in the Northern Rockies and Puget Trough, while transitions to shrublands were more prevalent in the Southwest. The model also showed an upslope shift in the mean elevation of boreal and temperate coniferous forests, reflecting an adaptive response to warmer and drier conditions. The shift in dominant PFTs resulted in a decrease in total NPP, vegetation carbon, and soil organic carbon by the end of the century, primarily attributed to increased fires and unfavorable climate conditions. While NEP showed an increase under both scenarios, primarily due to the expansion of grass and shrub PFTs, this increase did not offset the loss in carbon storage potential due to the decline in tree PFTs. The cumulative net carbon exchange (NCE) indicated a net carbon loss for the western United States of -60 GgC under RCP 4.5 and -82 GgC under RCP 8.5. Temperate coniferous forests contributed most to this carbon loss. Figures 1-5 in the original paper provide visual representations of these changes in PFT distribution, carbon stocks, and fluxes.

The findings of this study project a substantial shift in dominant PFTs in the western United States by the end of the century, driven by climate change and increased fire severity. This transition from tree-dominated to shrub and grass-dominated ecosystems signifies a critical alteration in the region's carbon storage capacity. While an increase in NEP was observed, primarily due to the increased productivity of grass and shrublands, this does not compensate for the overall decline in carbon storage associated with the loss of forests. The upslope migration of tree PFTs suggests an adaptive response but also points towards a potential reduction in suitable habitats for tree species, which are crucial for high carbon storage. The observed trends align with other modeling studies and observational data from networks such as LTER and Carbon Flux Tower, although some differences exist in the extent of PFT transitions. For instance, this model shows forest persistence in some areas, while other studies predict complete transitions. The discrepancies may be attributed to differences in model parameters, spatial resolution, and the specific regions considered. The significant role of fire in accelerating PFT shifts is highlighted, with fires acting as a catalyst that facilitates the establishment of grass and shrublands. This study emphasizes the urgent need for proactive conservation strategies to mitigate the negative impacts of climate change on western US ecosystems.

This study demonstrates the significant potential for climate change to drive widespread shifts in dominant plant functional types across the western United States, resulting in substantial losses in carbon storage. The model projections highlight the importance of moisture stress and fire severity as key drivers of these changes. While increases in NEP from shrub and grassland expansion partially offset some carbon losses, the overall reduction in carbon storage from the loss of forests is considerable. This underscores the need for urgent action to mitigate climate change and implement effective forest and grassland management strategies to maintain the ecosystem services provided by these biomes. Future research could focus on refining model parameters, such as improving the representation of seed dispersal and post-fire regeneration processes, to enhance the accuracy of future projections and inform more targeted conservation and adaptation measures.

The model makes several simplifying assumptions, including the assumption of readily available seed sources for all PFTs after fires, which might not hold true in reality. The model operates at the PFT level rather than the species level, which limits the ability to fully capture the complexities of species interactions and ecological dynamics. The model does not explicitly account for human land-use changes, which could further influence vegetation shifts. Uncertainty associated with climate projections and model parameters introduces variability in the results. Future studies should refine the model to account for these limitations and enhance its predictive accuracy.