Data from: Riverscape genomics of a threatened fish across a hydroclimatically heterogeneous river basin

Understanding how natural selection generates and maintains adaptive
genetic diversity in heterogeneous environments is key to predicting the
evolutionary response of populations to rapid environmental change.
Detecting selection in complex spatial environments remains challenging,
especially for threatened species where the effects of strong genetic
drift may overwhelm signatures of selection. We carried out a basinwide
riverscape genomic analysis in the threatened southern pygmy perch
(Nannoperca australis), an ecological specialist with low dispersal
potential. High-resolution environmental data and 5162 high-quality
filtered SNPs were used to clarify spatial population structure and to
assess footprints of selection associated with a steep hydroclimatic
gradient and with human disturbance across the naturally and
anthropogenically fragmented Murray–Darling Basin (Australia). Our
approach included FST outlier tests to define neutral loci, and a
combination of spatially explicit genotype–environment association
analyses to identify candidate adaptive loci while controlling for the
effects of landscape structure and shared population history. We found low
levels of genetic diversity and strong neutral population structure
consistent with expectations based on spatial stream hierarchy and life
history. In contrast, variables related to precipitation and temperature
appeared as the most important environmental surrogates for putatively
adaptive genetic variation at both regional and local scales. Human
disturbance also influenced the variation in candidate loci for
adaptation, but only at a local scale. Our study contributes to
understanding of adaptive evolution along naturally and anthropogenically
fragmented ecosystems. It also offers a tangible example of the potential
contributions of landscape genomics for informing in situ and ex situ
conservation management of biodiversity.