Forest fires and climate-induced tree range shifts in the western US

Climate change significantly impacts plant biogeography, causing shifts in plant distributions. While range shifts have been documented globally, the rates often lag behind the velocity of climate change. Several factors influence the rate of range shifts, including species characteristics, landscape features, and biotic interactions. Competition with pre-existing vegetation is hypothesized to slow range expansion, while disturbance, such as wildfire, could potentially accelerate it by reducing competition and opening niche space. Wildfires have been observed to drive range contractions, but evidence of range expansion facilitation is less clear. This study investigates the hypothesis that forest fires facilitate range expansions in tree species responding to contemporary climate change using observational data from the USDA Forest Inventory and Analysis (FIA). Understanding these dynamics is crucial for resource management, conservation, and sustainable future planning in the face of both climate change and increasingly frequent and intense wildfires.

Extensive research supports the expectation that plant distributions will change in response to climate change, with documented range shifts globally, especially in mountainous areas. However, observed range shift rates are often too slow to keep pace with climate change, highlighting the need for further investigation into factors affecting these rates. While species characteristics and landscape features play a role, the effects of biotic interactions, particularly competition, remain poorly understood. While some studies suggest competition might slow leading-edge range shifts, empirical evidence is limited. Conversely, theory and modeling suggest that disturbances, like wildfires, could facilitate range expansion by reducing competition. Existing research reveals that fire can drive range contractions, but the evidence for range expansion facilitation is less established. Furthermore, the complexities of post-fire succession and fire-regime adaptations add another layer of challenge to understanding the impact of fire on range shifts.

This study utilized plot-level tree and seedling species lists from the USDA Forest Inventory and Analysis (FIA) program, encompassing 74,069 plots within the Northwestern Forested Mountains and Marine West Coast Forest ecoregions of the continental U.S. The FIA data included tree age, diameter at breast height (dbh), and fire disturbance history (fires within 5 years prior to plot establishment and causing >25% damage/mortality). Climate data (1981-2010 averages) were obtained from the AdaptWest Project. After filtering for sufficient sample size and consistency in range shift direction across life stages and fire history (to minimize confounding by fire adaptations), eight tree species were selected for analysis. Seedling Only Range Displacement (SORD) was calculated using two metrics: Schoener's D (climatic niche equivalency) and Euclidean centroid distance (magnitude and direction of niche displacement). Schoener's D assesses the difference in climatic niches between seedling-only and tree-plus-seedling plots, while Euclidean centroid distance approximates range shift rate by comparing the climatic niche centroids. The difference between burned and unburned centroid distances (CDBCD) served as the primary comparison metric. Further analyses examined the contribution of individual climate variables (Mean Temperature of the Warmest Month (MTWM), Mean Temperature of the Coldest Month (MTCM), Mean Summer Precipitation (MSP), and Mean Winter Precipitation (MWP)) to SORD. Robustness checks were conducted using alternative sets of climate variables and minimum sample size thresholds.

Of the eight tree species analyzed, two (*Pseudotsuga menziesii* and *Quercus chrysolepis*) exhibited significantly greater SORDs in burned areas compared to unburned areas (p<0.05). This difference was evident in both Schoener's D and Euclidean centroid distance metrics. The average trend across all eight species showed that the centroid distance in burned plots was 89% greater than in unburned plots. Five species (*Chrysolepis chrysophylla*, *Pinus ponderosa*, *Pseudotsuga menziesii*, *Quercus chrysolepis*, and *Quercus kelloggii*) showed seedling-only plots with significantly greater mean summer precipitation and significantly lower mean temperature of the warmest month than their tree-plus-seedling plots in unburned areas. This difference was amplified in burned plots (p<0.01), indicating that wildfire exacerbated these climate-related differences between life stages. The observed SORDs, when roughly converted to spatial distance, were consistent with altitudinal range shifts observed in western Europe (up to 214 m in mean elevation). The study area experienced statistically significant increases in temperatures and decreases in precipitation between 1961-1990 and 1981-2010. The three subalpine species did not exhibit the same pattern of lower MTWM and higher MSP in seedling-only plots, potentially due to limited suitable establishment sites at higher elevations.

The findings provide empirical evidence that wildfire can significantly increase the potential range shift rate for certain tree species responding to climate change. The greater SORDs in burned areas support the hypothesis that competition constrains range shift rates. This suggests that the removal of competitors via wildfire can accelerate range expansion, highlighting a previously underappreciated mechanism of interaction between climate change and wildfire. The results reinforce the idea that multiple barriers, including competition, dispersal limitations, and geographic barriers, impede plant species' ability to track optimal climatic conditions. The study's limitations, including potential for non-maturing seedling populations, fire-regime adaptations, and the imperfect relationship between climatic and spatial distance, warrant consideration when interpreting the findings. Future research should investigate the relative contributions of wildfire's direct (abiotic) and indirect (biotic) impacts on range shifts, considering diverse fire severities and species traits.

This study demonstrates that wildfire can accelerate climate-induced tree range shifts by reducing competition. The findings highlight the importance of considering biotic interactions and disturbance regimes when predicting vegetation responses to climate change. Future research should focus on understanding the mechanisms underlying these interactions, incorporating detailed fire characteristics and species-specific traits, to improve predictions of vegetation distribution in a changing climate.

Three main sources of uncertainty limit the study's conclusions: (1) The possibility of non-maturing seedling populations, which are not truly leading-edge populations; (2) The confounding effects of fire succession and fire-regime adaptations on species regeneration; and (3) The imperfect proxy nature of using climatic distance to represent geographic distance, especially across topographically diverse landscapes. The FIA data also limit the ability to explore the nuanced mechanistic interactions between wildfire and range shifts because fire characteristics are not thoroughly described in the data.