Incorporating large-scale disturbances into demographic range models: The importance of fire as a factor limiting the distribution of Pinus edulis

Abstract

Background/Question/Methods

Making accurate predictions about species’ future distributions requires a holistic understanding of the factors that determine species range limits. Previous studies of range limits have focused primarily on the direct effects of climate and competition as drivers of range limits. However, climate may also play an indirect role in shaping species distributions via its effect on disturbance regimes. Indeed, previous research on P. edulis demonstrated that climate and competition together were not enough to explain the geographic distribution: projected population growth rates were high at high-elevation locations where climate is was favorable but P. edulis is absent. Here, we used a stochastic demographic range model to evaluate the role of disturbance, particularly fire, as a factor influencing the geographic distribution of Pinus edulis, a tree species at the arid edge of the forest biome. We parameterized the model with census data on 23,426 trees in 1,941 forest inventory plots and incorporated fire as a stochastic disturbance based on fire occurrence data. We then projected the expected P. edulis distribution based on its long-term demographic performance.

Results/Conclusions

Fire damage was observed in more frequently in plots where P. edulis was absent than in plots where P. edulis was present (4.7% vs 1.7%). In plots that experienced fire, mortality rates of P. edulis were 54.9% in plots experiencing only ground fire damage, 78.6% in plots experiencing ground fire and crown fire damage, and 96.5% in plots experiencing crown fire damage, suggesting that fire can be a major driver of mortality in P. edulis. Applying stochastic fire return intervals (FRI) characteristic of three forest types of the Colorado Plateau region – Ponderosa pine, mixed conifer, and spruce-fir forests – we find that the long-term stochastic growth rate of P. edulis in the presence of fire is below the replacement level of 1.0 in the two forest types with lower FRI. While the direct effects of climate are an important driver of geographic distributions, the fundamental Earth system process of fire is likely necessary to explain P. edulis’ distribution. Cross-scale interactions between climate and disturbance can lead to complex dynamics that are not predictable from models that include only climate and competition.

Date
Aug 3, 2020 12:00 AM — Aug 6, 2020 12:00 AM
Location
Online only