青藏高原逐月多年平均积雪密度格点数据集  

dataset of monthly multi-year average snow density grids on the Tibetan Plateau

作　　者：曹晓云 史飞飞 刘致远 张娟[1,2,3] 李弘毅 肖建设 CAO Xiaoyun;SHI Feifei;LIU Zhiyuan;ZHANG Juan;LI Hongyi;XIAO Jianshe(Institute of Qinghai Meteorological Science Research,Xining 810001,P.R.China;Key Laboratory of Disaster Prevention and Mitigation of Qinghai Province,Xining 810001,P.R.China;Northwest Institute of Eco-Environment and Resources,Chinese Academy of Sciences,Lanzhou 730000,P.R.China;School of Electronic Engineering,Chengdu University of Information Engineering,Chengdu 610225,P.R.China)

机构地区：[1]青海省气象科学研究所,西宁810001 [2]青海省防灾减灾重点实验室,西宁810001 [3]中国科学院西北生态环境资源研究院,兰州730000 [4]成都信息工程大学电子工程学院,成都610225

出　　处：《中国科学数据(中英文网络版)》2025年第1期305-317,共13页China Scientific Data

基　　金：长江流域开放基金项目(CJLY2022Y12);国家自然科学基金(U22A20556,U21A2021,41761078);青海省科技厅科技计划项目(2024-ZJ-740);第二次青藏高原综合科学考察研究项目(2019QZKK0105);青海省防灾减灾重点实验室项目(QFZ-2021-Z01)。

摘　　要：积雪密度是表征积雪特性的一个重要参量,也是将积雪深度转换成雪水当量的重要指标,在山区雪水资源估算、融雪洪水等水资源管理、自然灾害预报以及气候研究等方面具有重要作用。以1960–2020年青藏高原132个逐日国家气象站资料、中国区域地面气象要素驱动数据集、卫星融合雪深数据集为主要数据源,分不同地表类型比较几种机器学习模型在积雪密度模拟中的性能,选取最优模型,综合地面、卫星和再分析资料,制作了青藏高原逐月积雪密度数据集。与132个青藏高原国家气象观测站的逐月多年平均积雪密度数据进行精度检验发现,逐月多年平均积雪密度的平均均方根误差为0.019g/cm^(3),平均相对误差为11.88%,表明数据集具有较高精度。本数据集将为青藏高原水资源评估、水文过程模拟等提供数据支撑。Snow density is an important parameter for characterizing snow cover and functions as a key index for converting snow depth into snow water equivalent.It also plays a vital role in the estimation of snow water resources in mountainous areas,the management of water resources such as snowmelt floods,natural disaster forecasting,and climate research.Sourcing data from 132 day-by-day national meteorological stations on the Tibetan Plateau from 1960 to 2020,China’s regional surface meteorological element-driven dataset,and the satellite-fused snow depth dataset as the main data sources,we compared the performance of several machine learning models in the simulation of snow density by different surface types,and selected the optimal model.Integrating the ground,satellite,and reanalysis data,we finally produced the dataset of monthly multi-year average snow density grids on the Tibetan Plateau.An accuracy evaluation using month-by-month multi-year average snow density data from 132 national meteorological stations on the Tibetan Plateau showed that an average root mean square error of 0.019 g/cm^(3) and an average relative error of 11.88%,indicating the high accuracy of the dataset.This dataset is expected to provide data support for water resource assessment and hydrological process simulation on the Tibetan Plateau.

关 键 词：青藏高原 积雪密度 气象站点 机器学习 

分 类 号：P426.635[天文地球—大气科学及气象学]