The frequency and severity of extreme climate events are expected to increase with climate change. However, modeling the ecological impact of these events is challenging; most models use a large spatial scale. One example of these climate extremes is hurricanes. Hurricane disturbance regularly occurs in tropical forests, creating gaps in the forest canopy. These gaps increase light availability in areas where the most limiting resource is light, leading to shifts in population dynamics and species composition. Hurricanes are predicted to increase in intensity (strength) due to climate change, although their frequency is projected to remain the same. Many ecological studies have looked at the potential influence of increasing hurricane frequency on tropical ecosystems, but few of these have looked at the potential impacts of changing intensity. In this study, we propose a method to model the ecological impacts of increasing hurricane intensity using a tropical shrub in Southeastern Florida, Ardisia escallonioides, as a case study.
To model hurricane wind speed, we used the generalized extreme value (GEV) distribution parameterized by historical hurricane and tropical storm data. We created different future intensity scenarios under climate change and incorporated these hurricane disturbances into a demographic model of A. escallonioides.
Results/Conclusions
Using the GEV to model hurricane activity around our specific study area, we found a higher occurrence of stronger storms compared to previous studies that modeled larger geographic areas (e.g. the entire state of Florida). We altered the mean and variance parameters of the GEV according to future hurricane intensity projections under climate change. We combined our GEV model with the A. escallonioides population model by (1) discretizing the probability density function into a vector of the five Saffir-Simpson hurricane categories and (2) increasing the amount of damage caused by each category of hurricane to reflect increased water damage (the Saffir-Simpson category is only based on storm wind speed and does not include water damage). We found that the stochastic growth rate was most influenced by changes in the disturbance matrix.
This study provides a broadly applicable approach to modeling extreme events under climate change on an ecologically-relevant scale. For instance, this GEV technique could be used to model the impacts of increased flooding or drought on different populations. By incorporating our quantitative model of climate events, we can systematically explore how the frequency and intensity of these events may affect population dynamics (e.g. stochastic growth rate), potentially informing conservation and management decisions.