Contrasting responses of near-surface temperature to historical land cover change in CESM

Previous modeling studies have made significant contributions to understanding the climatic effects of historical land use and land cover change (LULCC). However, the absence of transient land cover simulations may lead to uncertainties or inaccuracies in assessing their impacts. Further investigation of differences between fixed and transient LULCC simulations are needed. Here we employ the Community Earth System Model (CESM) to analyze contrasting responses of mean and extreme near-surface air temperature to historical land cover change. Our results show that forest cover in Europe generally follows a linear upward trend, while East Asia experiences deforestation processes during the historical period. It’s found that temperature changes do not exhibit similar seasonal variation and have regional dependence, with Europe showing more pronounced seasonal variability. It’s also demonstrated using fixed land cover simulations exaggerates the temperature responses, leading to an overestimation of temperatures. In Europe, the overestimation of mean and extreme near-surface air temperature is approximately 0.2°C and 0.3°C, respectively. However, the overestimation is about 0.1°C in East Asia. Besides, we further disentangle the local and non-local effects in the temperature changes and show that non-local atmospheric feedbacks dominate the temperature responses in Europe, while local and non-local effects exhibit similar temperature variations in East Asia. Further efforts to explore the non-local effects of realistic land cover change could help enhance our understanding of climatic effects of land cover change at mid-latitudes.