Evaluating different approaches to integrate genome-wide genetic diversity in spatial conservation prioritization

Locations for new protected areas should be identified with consideration of intraspecific genetic diversity because genetic variation enhances the adaptability of species to novel environmental conditions. The process of siting new protected areas is called spatial conservation prioritization and is often formulated as a mathematical problem to be solved with the aid of computer software. Integrating multidimensional continuous metrics of genome-wide genetic diversity in spatial conservation prioritization involves significant challenges in terms of computation speed and memory requirements. In this study, we tested and compared several approaches to approximate the multidimensional continuous metrics of genetic diversity using multidimensional discrete genetic clusters or unidimensional discrete conservation features. To this end, we applied two tools for spatial conservation prioritization, RAPTR and PRIORITIZR, to genomic datasets of two fish species (the white seabream Diplodus sargus and the red mullet Mullus surmuletus) sampled throughout most of their distribution range in the Mediterranean Sea. We found that most of the approximations tested either did not achieve all genetic conservation targets or resulted in excessively high conservation costs. Nonetheless, approaches based on multidimensional genetic clusters allowed generating prioritizations with relatively small gaps in genetic objectives at very high computation speed. This might be an interesting option for exploratory or interactive analyses such as those performed in workshop with stakeholders.