Abstract
Global change results in an increase in temperature mean and variability and generates new environmental conditions for wild populations. In order to respond to this new treat, wild populations can move to another place (dispersion), change their behaviour or their phenology (phenotypic plasticity) or respond by evolutionary change (involving changes in alleles frequency). However, phenotypic plasticity or dispersion are unlikely to sustain population responses to climate change over the long term, contrary to the evolutionary response. Wild populations could evolutionary respond to climate change only if the increase in temperature represents a selective pressure on fitness-related traits. Yet, quantifying the selective pressures acting on wild populations remains challenging, especially in response to climate change.During this PhD, I focused on the selection pressures related to climate warming during a 26-year monitoring of blue tits (Cyanistes caeruleus) in southern France. First, I explored the bias inducted by spatial autocorrelation on estimation of natural selection in the wild when it is not taken into account. Second, I quantified the impact of climate warming on natural selection acting on laying date. Along with a rise in mean temperatures, climate change is also characterized by an increase in the frequency of extreme climatic events (ECE). Hence, I finally explored the impact of multiple ECEs such heavy rainfalls and extremes temperatures on the strength of natural selection acting on laying date.Our results showed that spatial autocorrelation is severely biasing our estimation of natural selection in the wild if it is not taken into account. In fact, natural selection is continuously overestimated by a non-explicit spatial model when the fitness is spatially autocorrelated. Hence, we developed 4 spatially-explicit models of selection, and we compared them in order to assess their reliability. Also, we detected a strong warming in our study site, which was correlated with an increase in natural selection strength. In fact, our results indicated an increase in the strength of selection by 46% every +1°C anomalies in maximum April temperature. Beyond this selective impact of the warming trend, we detected an impact of the extreme hot days occurrence during the nestlings stage on the fledglings success. More interesting, these ECEs significantly increased the strength of selection acting on laying date, independently form the mean temperatures.To conclude, my PhD helped to a better understanding of natural selection estimation in the wild, and the selective impact of climate warming. The first part allowed to develop specific models of selection in order to take into account spatial autocorrelation, a universal phenomenon present at any scale. The second part confirmed the selective impact of warming, pointing out a potential evolutionary response to climate change in the future. The last part demonstrated and quantified the selective impact of multiple ECEs, which has not been demonstrated before. We call now for more studies assessing the selective impact of climate warming, which could allow to predict the evolutionary responses in the future.