Population responses to climate variability: the importance of temporal scale

Abstract

New title: Population responses to observed climate variability across multiple taxa

Global climate change is likely to lead to concomitant changes in climate means, variability, and extremes, resulting in more variable and unpredictable environments to animal and plant populations. Increased variation in climatic conditions, such as in temperature and precipitation, could influence population dynamics by changing the mean and variation of vital rates and ultimately the population growth rate. However, empirical evidence for the effect of climate variability on wild populations is so far mixed.

In this study, we explored population responses to climate variability based on long-term empirical data (≥20 consecutive years of annual abundance estimates) of 56 wild populations across multiple taxa (mammals, amphibians, reptiles, plants and insects) from the Northern Hemisphere. We sought to understand the influences of climate variabilities (temperature and precipitation in active and inactive seasons) on the temporal variation of annual population growth rates and the geometric mean of annual population growth rates (i.e. long-term population persistence) in order to identify organisms that are particularly sensitive to fluctuating climatic conditions. Population and climate variabilities were quantified from the coefficients of variation (CV) at two temporal scales (using a four-year moving time-window and across ≥ 20 years) to assess the effect of temporal scale on conclusions.

At the short-time scale (4 years; when high local variations are more pronounced), we observed a slight positive relationship between CV of population growth rate and CV of climatic conditions (precipitation and temperature) in the active season, suggesting that populations fluctuated more under variable than under stable climatic conditions. In contrast, at the long-time scale (≥ 20 years), population variability was not associated with climate variability in the active season, and a negative relationship between CV of growth rate and CV of precipitation was observed in the inactive season. At both temporal scales, we found a negative relationship between life span and CV of population growth rate, indicating that the populations of short-lived species generally fluctuate more than those of long-lived species, but are similarly affected by climate variability. No statistically significant relationship was found between climate variability and the geometric mean of annual population growth rates.

Our results suggest that temporal fluctuations in population size across taxa are partially associated with climate variability, but that these fluctuations do not seem to translate into overall shifts in the long-term persistence of populations. This finding thus indicates that other factors than observed climate variability are probably more important to the dynamics of wild populations.