Abstract. The ground temperature at a fixed depth is a crucial boundary condition for understanding the properties of deep permafrost. However, the commonly used mean annual ground temperature at the depth of the zero annual amplitude (MAGT_dzaa) has application limitations due to large spatial heterogeneity in observed depths. In this study, we utilized 231 borehole records of mean annual ground temperature at a depth of 15 meters (MAGT_15m) from 2010 to 2019 and employed support vector regression (SVR) to predict gridded MAGT_15m data at a spatial resolution of nearly 1 km across the Qinghai-Tibet Plateau (QTP). SVR predictions demonstrated a R² value of 0.48 with a negligible negative overestimation (-0.01 °C). The average MAGT_15m of the QTP permafrost was -1.85 °C (±1.58 °C), with 90% of values ranging from -5.1 °C to -0.1 °C and 51.2% exceeding -1.5 °C. The freezing degree days (FDD) was the most significant predictor (p<0.001) of MAGT_15m, followed by thawing degree days (TDD), mean annual precipitation (MAP), and soil bulk density (BD) (p<0.01). Overall, the MAGT_15m increased from northwest to southeast and decreased with elevation. Lower MAGT_15m values are prevail in high mountainous areas with steep slopes. The MAGT_15m was the lowest in the basins of the Amu Darya, Indus, and Tarim rivers (-2.7 to -2.9 °C) and the highest in the Yangtze and Yellow River basins (-0.8 to -0.9 °C). The baseline dataset of MAGT_15m during 2010–2019 for the QTP permafrost will facilitates simulations of deep permafrost characteristics and provides fundamental data for permafrost model validation and improvement.

中文翻译：

---

青藏高原15米深度多年冻土温度基线（2010-2019年）


摘要。固定深度的地面温度是了解深层永久冻土特性的关键边界条件。然而，常用的零年振幅深度的年平均地温（MAGT _dzaa ）由于观测深度的空间异质性较大，存在应用局限性。在本研究中，我们利用 2010 年至 2019 年 231 个 15 米深度的年平均地温 (MAGT _15m ) 钻孔记录，并采用支持向量回归 (SVR) 来预测网格 MAGT _15m 值为 0.48，负高估可忽略不计 (-0.01 °C)。青藏高原多年冻土平均MAGT _15m 为-1.85℃（±1.58℃），其中90%的值在-5.1℃至-0.1℃之间，51.2%的值超过-1.5℃ 。冰冻度日数 (FDD) 是 MAGT _15m 最显着的预测因子 (p<0.001)，其次是解冻度日数 (TDD)、年平均降水量 (MAP) 和土壤容重 (BD) （p<0.01）。总体而言，MAGT _15m 由西北向东南增大，随海拔升高而减小。较低的 MAGT _15m 值普遍存在于陡坡的高山区。 MAGT _15m 在阿姆河、印度河和塔里木河流域最低(-2.7～-2.9℃)，在长江、黄河流域最高(-0.8～-0.9℃)。 ℃）。 2010-2019年青藏高原多年冻土MAGT _15m 基线数据集将有助于模拟深层多年冻土特征，为多年冻土模型验证和改进提供基础数据。