To study the distribution of genetic and phenotypic variation in different environments and at different spatial scales is important in order to understand the process of local adaptation and how populations will respond to future climate change. In my thesis I study populations of moor frogs (Rana arvalis) at different spatial scales, first along a 1700 km latitudinal gradient (Paper I, II, IV) and, second, in a system of inter-connected wetlands (III, IV). In Paper I, I present evidence for a major latitudinal break-point in larval life-history traits which is linked to a post glacial contact zone between two lineages that colonized Scandinavia after the last ice age. Using Q_ST-F_ST comparisons I found divergent selection acting on life-history traits, where a major source of differentiation comes from the two colonization routes. In Paper II I focus on genomic variation, demographic history and selection along the gradient. Using demographic modeling I confirm the proposed demographic history and show historical signatures of gene flow between regions and over the contact zone. In terms of genetic variation showing extreme differentiation as well as associations with growing season length I identify numerous variants under putative divergent selection, some of which have functions relating to immunity and development. I further show that differentiation outlier variation is higher in the north, as compared to neutral variation and variation associated with growing season length, which both decrease with latitude. These patterns are shaped by gene flow over the contact zone and the increased strength of drift at higher latitudes. I reduce the spatial scale in Paper III and characterize larval environments, landscape and geographical distance, to partition their influence on genetic variation. I show that environment explained more of the genetic variation than landscape and geographic distance, indicating that adaptive divergence can persist under high gene flow. Using the environmental variables, I identify genetic variants under putative divergent selection with functions associated with development and immunity. Using data from both scales, Q_ST-F_ST comparisons and gene-phenotype associations I show in Paper IV that selection on both larval traits aligns across scales, whereas selection on plasticity only aligns in size at metamorphosis. This further connects to the influence of temperature and seasonal time constraints in colder environments. Finally, I find several genetic variants associated with the traits and plasticity at both spatial scales with functions relating to immunity and metamorphosis.