祁连山国家公园多年冻土分布（2010）

祁连山国家公园多年冻土分布数据源是青藏高原多年冻土地温与热稳定型分布图，该数据是综合遥感冻融度日、积雪覆盖日、叶面积指数、土壤容重、高精度土壤含水量等数据，采用集成学习方法，运用基于距离块重采样训练数据的支持向量回归模型SVR (Support Vector Regression），对青藏高原（TP）2010年代（2005-2015年）的237个钻孔进行了现场年平均地温（MAGT）测量。根据交叉验证结果，选择最优模型模拟年平均地温。最后，根据年平均地温模拟结果划分多年冻土稳定型和范围。数据格式为TIF格式。

The permafrost distribution data source of Qilian Mountain National Park is the distribution map of permafrost ground temperature and thermal stability over the Qinghai-Tibet Plateau. This dataset integrates multiple data including remote sensing-derived freeze-thaw degree days, snow cover days, Leaf Area Index (LAI), soil bulk density, and high-precision soil moisture content. An ensemble learning method was adopted, specifically the Support Vector Regression (SVR) model based on distance block resampling of training data, to conduct field measurements of mean annual ground temperature (MAGT) on 237 boreholes across the Qinghai-Tibet Plateau (TP) during the 2010s (2005–2015). The optimal model for simulating mean annual ground temperature was selected based on cross-validation results. Finally, the stability and spatial distribution range of permafrost were classified according to the simulated mean annual ground temperature. The data format is TIF.

搜集汇总

背景与挑战

背景概述

该数据集提供了2010年代祁连山国家公园的多年冻土分布情况，基于2005-2015年的遥感与现场测量数据，采用集成学习方法模拟年平均地温并划分冻土稳定型。数据具有高空间分辨率（100m - 1km），以TIF格式存储，适用于冰冻圈和冻土研究，并遵循开放获取许可。