Can Artificial Retreat Sites Help Frogs Recover After Severe Habitat Devastation? Insights On The Use Of "Coqui Houses" After Hurricane Maria In Puerto Rico

On September 2017, Hurricane Maria swept over Puerto Rico as a Category 4 storm. Severe canopy loss, augmentation of forest floor debris, and a significant increase in temperature and light reaching the understory were among the most evident changes at El Yunque National Forest, where a population of Eleutherodactylus coqui frogs has been monitored over the past 30 years. When sampling was re-established, the frogs could be heard calling, but it was very difficult to find them among the complexity of vegetation in the forest floor. We inferred that canopy disturbance had left frogs without optimal arboreal habitats for retreat, nocturnal perching, feeding, and reproductive activities, and wondered whether they would use artificial habitats placed in the forest understory. To test this, two types of artificial habitats (i.e., "coqui houses") were introduced in the forest understory, consisting of either open PVC pipes or single-entrance natural bamboo shoots. Surveys were conducted twice a month for 15 months in an experimental transect with coqui houses, and a control transect without them. Data were collected on the occupancy rate of the artificial sites, type of usage, time of day occupied, and the number of E. coqui observed. The effects of time since the hurricane, microhabitat temperature, type of coqui house, and seasonality on the occupancy rate were also evaluated. Results showed that coquis used bamboo houses mostly during daytime as retreat and nesting sites, whereas the PVC houses were used mostly at night as calling sites. Daytime occupancy of coqui houses showed a significant bell-shaped pattern over time since the hurricane. This may be explained by a steady increase in usage after severe forest damage, a peak during the stressful cool-dry season, and a decline afterwards as the forest began to recover. No differences were found in frog counts between experimental and control transects, probably because the coquis could also hide among the fallen vegetation, but either disparities in forest conditions or inappropriateness of the methods for estimating population numbers may have overshadowed this effect. Coquis used artificial houses more often during the most stressful environmental conditions, suggesting that these shelters may serve to enhance habitat quality for amphibians after extreme weather events.