中国湖泊表层温度及去除极端高温的湖泊表层温度日尺度数据集（1985-2022）

本数据包含中国2260个湖泊的湖泊表层水温以及去除极端高温后的湖泊表层水温。数据由与卫星数据耦合的物理/统计混合模型Air2Water模拟得到。数据时间分辨率为逐日，时间范围为1985-01-01至2022-12-31，气象驱动数据为ERA5-Land。卫星观测数据为Landsat的LST产品。本数据与中国6个湖泊（太湖、千岛湖、呼伦湖、纳木错、泸沽湖、洱海）的原位观测数据进行了验证，相关系数均高于0.96，平均均方根误差为1.75℃，平均绝对误差为1.38℃，验证结果可信。本数据可为气候变化和湖泊生态研究提供基础。

This dataset encompasses the surface water temperatures of 2260 lakes across China, alongside the lake surface water temperatures after excluding extreme high temperatures. The data were simulated via the physical-statistical hybrid model Air2Water, which is coupled with satellite datasets. The data have a daily temporal resolution, spanning from 1985-01-01 to 2022-12-31, with ERA5-Land serving as the meteorological forcing dataset. The satellite-derived observations are the Land Surface Temperature (LST) products from the Landsat series. This dataset was validated against in-situ measured data from six lakes in China, namely Taihu Lake, Qiandao Lake, Hulun Lake, Nam Co, Lugu Lake, and Erhai Lake. All correlation coefficients exceeded 0.96, with a mean root mean square error (RMSE) of 1.75℃ and a mean absolute error (MAE) of 1.38℃, indicating reliable validation outcomes. This dataset can serve as a fundamental data resource for studies on climate change and lake ecology.

搜集汇总

背景与挑战

背景概述

该数据集提供了1985年至2022年中国2260个湖泊的日尺度表层水温数据，包括原始值和去除极端高温后的值，通过Air2Water模型结合ERA5-Land和Landsat数据模拟生成，并经过原位观测验证（相关系数高于0.96，平均误差约1.5°C），数据以CSV格式开放获取，适用于气候变化和湖泊生态研究。