青藏高原冰川表面1km月均气温数据（1961-2020）

近地表空气温度（Ta）作为衡量全球变暖的关键指标，在冰川融化过程中起着关键作用。青藏高原地区分布着大量的冰川，但由于高寒缺氧导致的观测困难，青藏高原冰川地区的气温资料仍然稀缺，且已有资料存在精度较低、时序较短等问题。本资料整合气温观测资料及多种MODIS遥感产品，同时计算站点日出日落时间、太阳高度余弦以及气压作为辅助估算数据，利用随机森林模型估算了2002-2020年青藏高原月均气温，并根据估算结果基于贝叶斯岭回归原理对青藏高原冰川地区1961-2020年月均气温进行了重建。最终资料在冰川地区的RMSE为1.31℃。经比较，该产品在高原冰川地区的精度远高于其他同类型产品。基于本产品，可以分析近几十年来青藏高原冰川地区的气温变化趋势，并进一步讨论气候变化对青藏高原冰川的影响。

Near-surface air temperature (Ta), a key indicator for measuring global warming, plays a critical role in glacier melting. The Qinghai-Tibet Plateau hosts numerous glaciers, but air temperature data over its glacierized regions remain scarce due to observation difficulties caused by high altitude, harsh cold and oxygen deficiency. Existing datasets also suffer from issues such as low accuracy and short time series. This dataset integrates in-situ air temperature observations and multiple MODIS remote sensing products, and calculates sunrise and sunset times, the cosine of solar altitude, and atmospheric pressure at meteorological stations as auxiliary estimation data. The Random Forest model was used to estimate monthly mean air temperatures over the Qinghai-Tibet Plateau from 2002 to 2020. Based on the estimation results, monthly mean air temperatures from 1961 to 2020 over the plateau's glacierized regions were reconstructed using the Bayesian ridge regression method. The final dataset has a root mean square error (RMSE) of 1.31℃ in glacierized areas. Comparisons demonstrate that this product achieves far higher accuracy than other similar products in the plateau's glacierized regions. Based on this dataset, one can analyze the trends of air temperature change in the Qinghai-Tibet Plateau's glacierized regions over recent decades, and further discuss the impacts of climate change on glaciers in this area.