Non-Radiative Effect of Land Use/Land Cover Differences on Land Surface Temperature Simulated by Noah-MP Land Surface Model Over East Asia

Biophysical effects of land cover changes on land surface temperature (T-s), including two effects, that is, radiative effects (caused by surface albedo changes) versus non-radiative effects (caused by changes in the energy flux distribution), have attracted the attention of many researchers. However, the uncertainties of land use and land cover (LULC) and vegetation schemes on non-radiative effects, and the roles of non-radiative effects on T-s, remain uncertain. In this study, we investigate the local impact of LULC differences by comparing Noah with multiparameterization (Noah-MP) land model simulations with different LULC data sets and those with vegetation schemes whether they include interactive canopies. We attribute T-s changes induced by LULC differences obtained from Noah-MP simulations to radiative and non-radiative effects over East Asia based on the decomposed temperature metric theory. The results indicate that vegetation cools T-s over urban regions by 4.0-6.0?. Grassland and cropland decrease the T-s over barren regions. Forests are warmer (cooler) than grassland and cropland in high latitudes (in mid and low latitudes). Grassland is cooler than cropland. However, the magnitude of T-s changes induced by these differences is lower than that induced by differences between vegetation and urban. The non-radiative effect generally explains more than 50% of the local changes in T-s. The interactive canopies enlarge T-s changes for afforestation or replacements of urban region by vegetation. The effect of uncertainties in LULC data sets is negligible. This study highlights the importance of non-radiative biophysical effects and interactive canopies on local changes in T-s.