Using dark diversity to disentangle the effects of protection and habitat quality on species diversity

The effectiveness of protected areas in preserving diversity is typically assessed by comparing them to control sites. However, studies often report inconsistent effects of protection on diversity. This inconsistency may partly result from hard-to-quantify gradients in habitat quality (i.e., the site's potential to harbor diversity), which can mask the impact of protection. We hypothesize that analyzing patterns of dark diversity, the assemblage of species that fit the site conditions but are locally absent, along with species pool size (the sum of observed and dark diversity), can help distinguish protection effects where protection and habitat quality are confounded. Specifically, protection should reduce dark diversity, as fewer species are absent, but it should not affect species pool size, which is primarily determined by habitat quality. Using marine protected areas and control sites across the Mediterranean Sea, we show that, as predicted, fish dark diversity decreases with protection but is independent of habitat quality, while species pool size was related to habitat quality but not protection. At the same time, we did not find an increase in species richness with protection. These results suggest that, by being less influenced by habitat quality, dark diversity provides a refined understanding of conservation effectiveness compared to richness alone. Since dark diversity can be estimated using species co-occurrence patterns, it remains useful even when habitat quality data are unavailable. Therefore, explicitly considering species pool size and dark diversity, alongside observed diversity, is critical for assessing control site adequacy and disentangling the effects of habitat quality and protection.