Do remnant forest patches provide microclimate buffering? A case study from Sri Lanka

Tropical forests have declined in extent considerably owing to large scale deforestation for providing land for multiple uses, and this has reduced their collective capacity to provide many important ecological functions. In the present study, conducted in the tropical island of Sri Lanka, we focus on remnant patches of two common and co-occurring vegetation types in the dry zone, Dry Mixed Evergreen Forest and Scrub Forest, to test (a) whether small remnant forest patches facilitate microclimate buffering and, if so, (b) whether the two forest types differ in terms of their potential to buffer the microclimate, and (c) whether the extent of canopy cover influences the microclimate buffering capacity. Co-occurring Grasslands were used as reference habitats. We used multiple temperature, relative humidity and vapour pressure deficit related metrics, and showed that Dry mixed Evergreen Forest and Scrub Forest had significant microclimate buffering potential, the latter to a lesser extent than the former (Dry mixed Evergreen Forest > Scrub Forest). These results are noteworthy, given that the patches of both forest types were relatively small (mean +/- standard deviation: Dry Mixed Evergreen Forest = 0.63 +/- 0.13 km(2) and Scrub Forest = 0.10 +/- 0.07 km(2)), intermixed and scattered across the same landscape. The study also showed that, after controlling for effects of patch size, the extent of canopy cover had a significant impact on most of the microclimate metrics. The impact of patch size on microclimate metrics was less pronounced. These findings highlight the need for conserving remnant forest patches, especially of the Dry Mixed Evergreen Forest, as refuges of biodiversity, particularly in the context of tropical forests in developing countries facing severe threats of deforestation.