The influence of matrix type in the relationship between patch size and amphibia richness: A global Meta-Analysis

Abstract The species-area relationship (SAR) is one of the main topics debated in ecology. Amphibians are highly sensitive to habitat loss and fragmentation, due to their traits and reproductive requirements. In the present study, we developed a meta-analysis to address a mean effect size to SAR regarding forest amphibians in man-made fragmented landscapes, and tested the influence of seven in-study and landscape variables that may affect this relationship globally. We conducted a search using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) protocol and calculated an effect size for each one of the 28 studies elected by our criteria. There is a positive relationship between amphibia richness and patch area, with a mean effect size of 0.68 and high heterogeneity among studies. The meta-regressions showed that only the matrix type surrounding the patches affected the variance between studies, explaining 46%. Clear cut was the matrix with the highest effect size, followed by silviculture, but both had a low representation of studies. Agriculture, grazing and urban matrices had a significant influence on patch area and the species richness relationship, but with distinct mean effects sizes and variances between-studies. The matrix type around the remnants may substantially affect results and interpretation of the SAR, possibly due to the varying capacity of amphibia species to cross and inhabit each matrix type. If possible, SAR should be compared within a context of multiple landscapes, and not only with a patch-focused and local approach.