Résumé
Local adaptation is generally studied at a large geographic scale, i.e. among populations. However, within-population environmental heterogeneity can be large and there is increasing empirical evidence of adaptation at microgeographic scale (i.e. within dispersal neighborhood [1,2,3]). A recent review [4] advocates more studies on the interplay of evolutionary drivers at a wider range of scales.We used the NEMO simulation framework in a hierarchical island metapopulation model to investigate the evolutionary response to hierarchical patterns of selection both within and among populations. By varying strength of selection, demographic parameters and the genetic architecture of traits, we investigated polygenic adaptation patterns and processes at microgeographic and population scales by using generalized Q-statistics and Fq-statistics for hierarchical population structures.Our results show that microgeographic adaptation can be expected under a broad range of parameter values including relatively low selection intensity. Patterns of differentiation at QTLs (Fq-statistics) at all scales are primarily shaped by the pattern of environmental heterogeneity, the intensity of selection and the number of QTLs determining the trait under selection. We also show interactions across spatial scales: e.g. patterns of within-population heterogeneity have an impact on patch occupancy and differentiation between populations. Furthermore, the discrepancy between Q-statistics and Fqstatistics due to covariance among QTLs depends on the interaction between the differentiation patterns at different hierarchical levels of the population structure. These results will be discussed from the point of view of the expected genome-wide distribution of marks of selection, particularly those that can be detected as changes of QTL-allele frequencies.