PS 19-186 - Natural disasters generate variability in individual life histories

Background/Question/Methods

Extreme climatic events may influence individual-level variability in phenotypes, survival, and reproduction, and thereby drive the pace of evolution. Individual variability can be quantified as the rate at which reproductive trajectories diversify across age through dynamic heterogeneity (entropy H). Major hurricanes may lead to strong intraspecific competition and force individuals to explore new niches increasing the rate of diversification in individual life courses (high H) or increasing the variance in life course trajectories (increased stochasticity). Alternatively, environmental pressure from major hurricanes may drive individuals to a certain optimal life course (low H, reduced variance). Here, we analyze if experiencing major hurricanes influences entropy, mean life expectancy (MLE) and mean reproductive output using data of individual life histories from Cayo Santiago rhesus macaque females over the course of 46 years. During such period, the population experienced three major hurricanes: Hugo (1989; cat. 3; sustained wind speed of 201 km h^-1), Georges (1998; cat. 3; 185 km h^-1), and Maria (2017; cat. 4; 220 km h^-1). We construct annual matrix population models and estimate the damping ratio, population entropy, variance in MLE and variance in mean reproductive output for 5,847 females.

Results/Conclusions

In the nonhurricane environment, λ = 1.121 (95% CI: 1.116, 1.126), while hurricane years together had a λ of 1.09 (95% CI: 1.085, 1.116). Years with major hurricanes showed an increased damping ratio (range: 1.50-1.54), suggesting that the population takes more time to recover to stable stage distribution following the perturbation of a natural disaster, relative to years of no hurricanes (1.48). We found that H increased with hurricane intensity. That is, hurricane Georges had the lowest relative entropy (H=0.67), followed by Hugo (0.76) and Maria (0.78) suggesting that strong competition during unfavorable years may force individuals into new niches. When looking at individual stochasticity, and thus disregarding the sequence of reproductive stages through time (stage dynamics), we found that hurricane years had a reduced variance and mean life expectancy of 17 years, relative to nonhurricane years (18 years) and a reduced variance in reproductive output of offspring (range: 5-5.7 offspring), relative to nonhurricane years (5.9). These results suggest a life history strategy that maintains survival at the expense of reproduction during hurricane years. Our work suggests that populations exposed to extreme climatic events may experience increased heterogeneity in stage dynamics but an overall decrease in MLE and reproductive output.