高分辨率北半球多年冻土数据集（2000-2016）

整编了目前北半球数量最多的年平均地温（1002个）和活动层厚度（452个）地面观测数据，利用四种统计学习模型融合这些地面观测与多源遥感等数据产品，集合模拟得到了代表2000-2016年北半球多年冻土区年平均地温、活动层厚度、多年冻土发生概率和多年冻土水热分带数据集，空间分辨率为1公里，验证表明具有更高的精度。可为北半球多年冻土区的工程规划、设计、环境模拟与评价等提供数据支持，也可作为北半球多年冻土现状的数据基准，评估未来多年冻土变化及其影响。

This dataset is compiled from the largest available ground observation records of mean annual ground temperature (MAGT, 1002 sites) and active layer thickness (ALT, 452 sites) across the Northern Hemisphere to date. Four statistical learning models were employed to fuse these ground observations with multi-source remote sensing and other ancillary data products, and ensemble simulations were conducted to produce a gridded dataset depicting mean annual ground temperature, active layer thickness, permafrost occurrence probability, and permafrost hydrothermal zonation in Northern Hemisphere permafrost regions during 2000–2016. The dataset has a spatial resolution of 1 km, and validation experiments have confirmed its high accuracy. It can provide data support for engineering planning, design, environmental simulation and evaluation in Northern Hemisphere permafrost regions, serve as a data benchmark for the current state of Northern Hemisphere permafrost, and support the evaluation of future permafrost changes and their associated impacts.

搜集汇总

背景与挑战

背景概述

该数据集是一个高分辨率北半球多年冻土数据集，覆盖2000年至2016年，空间分辨率为1公里，整合了大量地面观测数据并利用统计学习模型融合多源遥感数据，生成了年平均地温、活动层厚度、多年冻土发生概率和水热分带等关键变量。数据集精度较高，旨在为北半球多年冻土区的工程规划、环境模拟和现状评估提供数据支持，并可作为未来变化研究的基准。