Understory vegetation contributes to microclimatic buffering of near-surface temperatures in temperate deciduous forests

Abstract

Accurately predicting microclimate is considered a high priority for understanding organismal responses to climate change at biologically relevant scales. However, approaches to developing robust microclimate datasets and understanding of the biophysical processes altering microclimatic regimes are limited. We developed and evaluated an approach for predicting microclimatic temperatures in montane forests that incorporates the influence of complex vegetation structure and landscape physiography. Additionally, we determined spatiotemporal mismatches between free-air and microclimatic temperatures to highlight the location, phenology, and magnitude of differences in predicted temperature. We combined temperature datalogger measurements with LiDAR-derived vegetation and GIS-derived landscape physiographic characteristics to downscale free-air temperatures to microclimatic (3 m2 spatial resolution) temperatures in the Great Smoky Mountains. We assessed the contribution of forest vegetation layers in altering microclimatic temperatures and model accuracy, and compared coarse-grain temperature maps with microclimatic temperature maps. Understory vegetation structure contributes to microclimatic buffering of near-surface, forest temperatures and enhances the accuracy of maximum temperature predictions during the growing season by altering the effects of solar insolation and topographic convergence index on microclimatic temperatures. Elevation and solar insolation covaried with spatiotemporal mismatches between free-air and microclimatic temperatures, suggesting that these landscape physiographic characteristics may contribute to deviations between macro- and micro-scale temperature. Our findings demonstrate the importance of including complex vegetation characteristics and biophysical interactions as climate forcing factors in microclimate modeling. We also demonstrate the plausibility of accurately predicting microclimatic temperatures over broad extents, an important step in predicting potential organismal responses to climate change.