青藏高原3km雪水当量、雪粒径与融雪量数据集（2002.8-2022.7）

1、数据内容：2002年8月1日-2022年7月31日青藏高原高分辨率季节性积雪雪深、雪水当量、积雪状态（雪粒径、雪密度、雪温度、雪湿度）和积雪通量（融雪量、升华量、降雪量、雪上降雨量）数据集。范围：70-105°E, 25-40°N。时间分辨率：逐小时（积雪状态量逐日）。空间分辨率：3km （0.1/3°）。
2、数据来源及加工方法：针对青藏高原地形复杂、积雪分布异质性高、雨浇雪现象突出的问题，使用光学与被动微波结合的方式提高反演空间分辨率，利用积雪过程耦合辐射传输模型建立MODIS表征的积雪累积消融过程与AMSR-E/AMSR2表征的干雪体散射水平之间的内在联系。利用过程模型能量与水量平衡的特点输出各类参数，形成逐小时（逐日）、多参数积雪数据集。
3、数据质量描述：根据研究区内310个气象台站20年日值雪深检验表明，0-120 cm范围内，均方根误差为6.1 cm，相关系数为0.79、平均偏差为-2.1 cm。受建站条件限制，气象数据在部分区域分布稀疏且未涵盖更高的雪深范围。
4、数据应用成果与前景：当前数据集空间分辨率高，参数多样。雪深与现有台站验证结果较好，对湿雪对反演的影响有一定抗性，其它参数与雪深的关系符合积雪物理过程机理。适用于支撑与积雪相关的气候变化、陆-气交互、水文生态和工程性研究。
5、其它声明：1）算法使用的驱动数据包括第三极地区长时间序列高分辨率地面气象要素驱动数据集（TPMFD）（https://cstr.cn/18406.11.Atmos.tpdc.300398）、ERA5-Land降水数据和IMERG降水数据，并假设气温和降水有一定偏差；2）算法从每年8月1日起算，未区分冰川区与非冰川区，因此部分冰川区会在第二年7月31日残余未融化的积雪累积量。

1. Data Content: This dataset contains high-resolution seasonal snow cover parameters over the Qinghai-Tibet Plateau from August 1, 2002 to July 31, 2022, including snow depth, snow water equivalent, snow cover states (snow grain size, snow density, snow temperature, snow moisture content), and snow cover fluxes (snowmelt volume, sublimation amount, snowfall amount, rainfall on snow). Spatial coverage: 70°E–105°E, 25°N–40°N. Temporal resolution: hourly (daily for snow cover state parameters). Spatial resolution: 3 km (0.1/3°).
2. Data Source and Processing Method: Aiming at the challenges of complex terrain, high heterogeneity of snow cover distribution and prominent rain-on-snow events over the Qinghai-Tibet Plateau, a combination of optical and passive microwave remote sensing data was adopted to improve the spatial resolution of snow cover inversion. A snow process-coupled radiative transfer model was utilized to establish the intrinsic relationship between the snow accumulation and ablation process characterized by MODIS and the dry snow volume scattering level represented by AMSR-E/AMSR2. Leveraging the characteristics of energy and water balance in the process model, various snow cover parameters were output to generate this hourly (daily for snow cover state parameters) multi-parameter snow cover dataset.
3. Data Quality Description: Validation based on 20-year daily snow depth data from 310 meteorological stations within the study area indicates that within the snow depth range of 0–120 cm, the root mean square error (RMSE) is 6.1 cm, the correlation coefficient is 0.79, and the mean bias is -2.1 cm. Restricted by station construction conditions, meteorological data are sparsely distributed in some regions and do not cover higher snow depth ranges.
4. Data Application Achievements and Prospects: This dataset boasts high spatial resolution and diverse parameters. The snow depth product shows good consistency with in-situ station validation results, and exhibits certain resistance to the interference of wet snow during inversion. The relationships between other parameters and snow depth conform to the physical mechanisms of snow cover processes. It is suitable for supporting snow-related research on climate change, land-atmosphere interaction, hydrology and ecology, and engineering applications.
5. Other Statements: 1) The driving data used in the algorithm include the long-term high-resolution ground meteorological forcing dataset for the Third Pole region (TPMFD) (https://cstr.cn/18406.11.Atmos.tpdc.300398), ERA5-Land precipitation data and IMERG precipitation data, with the assumption that there are certain biases in air temperature and precipitation. 2) The algorithm starts from August 1 of each year and does not differentiate between glacial and non-glacial areas, so residual accumulated unmelted snow by July 31 of the following year may exist in some glacial regions.

搜集汇总

背景与挑战

背景概述

该数据集提供了2002年8月至2022年7月青藏高原高分辨率季节性积雪的雪深、雪水当量、积雪状态和积雪通量等多参数数据，空间分辨率为3km，时间分辨率为逐小时或逐日。数据通过光学与被动微波结合的方式反演，验证结果显示较好的准确性，适用于气候变化、陆-气交互等研究领域。