Evaluating warming trend over the tibetan plateau based on remotely sensed air temperature from 2001 to 2020

The Tibetan Plateau (TP), the Third Pole of the world, has experienced significant warming over the past several decades. Previous studies have mostly relied on station-observed air temperature (T_a), reanalysis data, and remotely sensed land surface temperature (LST) to analyze the warming trend over the TP. However, the uneven distribution of stations, the poor spatial resolution of reanalysis data, and the differences between LST and T_a may lead to biased warming rates. This paper first maps T_a over the TP from 2001 to 2020 based on multi-source remote sensing data, and then quantifies the spatio-temporal variations of remotely sensed T_a and elevation dependent warming (EDW) of this region. The monthly mean T_a is estimated using machine learning (ML) method year by year, and its accuracy is validated based on station-observed T_a. The coefficient of determination (R² ranges from 0.97 to 0.98 and the mean absolute error (MAE) ranges from 1.01 to1.04 °C. The remotely sensed T_a is used to analysis warming trend and EDW over the TP. The overall warming trend of the TP during 2001–2020 is 0.17 ℃/10a, and warming mainly distributed in the eastern TP, central TP and western Kunlun Mountains. Among the four seasons, autumn shows the most significant warming, tripling the annual warming rate. Winter shows a significant cooling trend, with the warming rate of -0.18 ℃/10a. The study also reveales the existence of EDW at both the annual and seasonal scales. This paper suggests the potential of remotely sensed T_a in global warming study, and also provides an improved understanding of climate warming over the TP.