Landscape genomics aims to identify genomic regions having adaptive significance by combining genomic and environmental data using regression methods. As regards its genetic component, next-generation high throughput sequencing technologies became available at the beginning of the 2000s. Compared with the traditional DNA sequencing technologies developed in the late 1970s, they provide information characterizing much larger parts of the genome, are faster, and contribute in major discoveries and applications in medical research and evolution. But what about the environmental component of landscape genomics? Are the spatial and temporal resolutions of environmental variables improving compared with their genetic counterpart? Yes to a certain extent, but the scale at which adaptation of species to their local environment is investigated determines the richness and cost of available variables. On a large geographic scale, there is a plethora of data sets of very good quality and low cost. On a local scale, there is a plethora of cutting-edge sensors to be distributed in the field, and of multi-spectral satellite images whose use also results in high costs. In this context, multi-scale variables derived from high resolution digital elevation models provide reliable surrogates for topo-climatic variables, offering suitable alternatives for inclusion in landscape genomic models.