Data from: Seed sourcing in the genomics era: Multispecies provenance delineation for current and future climates

<strong>Abstract</strong> Restoration interventions require knowledge on the suitability of seed sources. Provenance delineation for ecological restoration of degraded environments has begun to incorporate genome-wide information on adaptive variation, but this has only been completed on a small number of plant species. Rarely is provenance delineation using a genomics approach applied to species occurring across different habitats, and in the context of future climate scenarios, despite their potential importance for successful long-term restoration. Here, we use neutral genetic data to investigate patterns of genetic differentiation and a landscape genomics approach to model putatively adaptive genetic variation under multiple climate scenarios for two species co-occurring in a predominantly mesic environment, and two species co-occurring in a semi-arid environment. We then model the genomic similarity of seed sourcing locations to hypothetical restoration sites, both under current and future climate scenarios. We found that the geographical extent of provenances and the amount of genomic change required to track the projected climatic conditions over time differed within the pairs of co-occurring species and between habitats. Additionally, future climate scenarios had differing effects on provenance patterns between the two landscapes indicating a differential response to changing climate between species from mesic and arid habitats. This implies that provenance guidelines can be both species and habitat dependent. We discuss how these results can be utilised to design seed sourcing strategies for successful restoration, and how these methods could be more broadly applied to delineate provenances of other species and locations. <br> <strong>Usage Notes</strong> <strong>FST matrices</strong> Pairwise FST matrices of putatively adaptive genetic data for four species used in GDM analysis. Datasets have been previously published (see https://doi.org/10.5061/dryad.k3j9kd56n and https://doi.org/10.5061/dryad.0p2ngf1xn) and the versions utilised for this manuscript have been archived here. <strong>Climate and geographical information</strong> Climate and geographical information for four species used in GDM analysis. Climatic data was downloaded with 1 km cell resolution from the Worldclim 2.1 database (Fick &amp; Hijmans, 2017; Hijmans et al. 2005). Bio3 = isothermality, bio4 = temperature seasonality, bio8 = mean temperature of the wettest quarter, bio10 = mean temperature of the warmest quarter, bio12 = annual precipitation, bio15 = precipitation seasonality, bio16 = precipitation of the wettest quarter, bio17 = precipitation of the driest quarter. <strong>R code</strong> R code for extrapolating GDMs to future climates and calculating genomic similarity to hypothetical restoration sites.