Agricultural intensification in tropical landscapes poses a new threat to the ability of biological corridors to maintain functional connectivity for native species. We used a landscape genetics approach to evaluate impacts of expanding pineapple plantations on two widespread and abundant frugivorous bats (Artibeus jamaicensis, Carollia castanea) and an understory tree species (Symphonia globulifera) in a biological corridor in Costa Rica. The bats are critically important for resilience of forest biodiversity. We hypothesized that increased agricultural intensification would lead to a loss of genetic diversity and an increase in genetic structure in the landscape. We sampled 735 bats in 26 remnant mature forest patches surrounded by different proportions of forest, pasture, crops, and pineapple. For the tree species, we sampled three life stages (adults, saplings, seedlings) using 324 samples per cohort across 18 remnant mature forest patches. We collected genetic data using 9-16 nuclear DNA microsatellite loci. We estimated genetic diversity using the metrics of allelic richness and heterozygosity, and gene flow using G’st. Genetic diversity and genetic structure was compared to relevant landscape metrics using a variety of landscape genetic analysis methods. We quantified landscape variables using four land cover maps representing the landscape in 1986, 1996, 2001 and 2011.
Results/Conclusions
Analyses of allelic richness and heterozygosity indicate that land cover type surrounding patches had no impact on genetic diversity of A. jamaicensis but declines in genetic diversity metrics were detected for C. castanea and S. globulifera in response to habitat loss and fragmentation. Least-cost transect analyses (LCTA) suggest that for A. jamaicensis, pineapple is the most permeable landcover type; this species flies long distances to reach patchily distributed food resources, and may use pineapple plantations as flyways. As hypothesized, forest is the most permeable land cover for gene flow of C. castanea. For both species, LCTA indicates that development plays a role in inhibiting gene flow. Genetic differentiation among adult trees was low while differentiation among seedlings was significantly higher indicating more restricted gene flow in the current landscape. For S. globulifera, we documented higher connectivity and genetic diversity inside the corridor compared to outside the corridor. The current study answers the call for landscape genetic research focused on tropical and agricultural landscapes, highlights the value of comparative landscape genetics in biological corridor design and management, and is one of the few studies of biological corridors in any ecosystem to implement a genetic approach to test corridor efficacy.