中国中西部地区年降水量空间分布数据集（2010）
收藏资源简介：
中国中西部地区年降水量空间分布数据集（2010）是以中国中西部地区（25°-35°N，105°-115°E）121个气象站2010年年降水量数据、TRMM 3B43同期降水估值产品，采用基于HASM（High Accuracy Surface Modeling）的混合插值法，即首先使用TRMM数据代表降水在空间中变化平稳的部分，即趋势面；然后结合地面站点观测值，计算去除趋势面后的残差值，利用HASM对变化剧烈的残差场进行插值，最后将趋势面与插值后的残差场进行相加，得到插值结果。经验证，该方法模拟精度明显高于基于传统插值方法（IDW、Kriging、Spline）生成趋势面的混合插值结果，适应性较好。在全局尺度上，平均绝对误差（MAE）和均方根误差（RMSE）分别为125.15mm和155.80mm。该数据集为基于TRMM数据和不同内插方法获取趋势面后得到的2010年降水量数据。空间分辨率为0.25°x 0.25°（单位：mm）。数据集存储为.tif格式，由20个数据文件组成，数据量为64.2 KB（压缩为1个文件，48.8 KB）。基于该数据集的研究成果发表在《地球信息科学学报》2015年17卷第8期。
Spatial Distribution Dataset of Annual Precipitation in Central and Western China (2010) was developed using the hybrid interpolation approach based on HASM (High Accuracy Surface Modeling). The data sources include annual precipitation records from 121 meteorological stations located in Central and Western China (25°N–35°N, 105°E–115°E) in 2010, and the contemporaneous TRMM 3B43 precipitation estimation products. Specifically, TRMM data is first utilized to represent the spatially smooth component, i.e., the trend surface, of precipitation. Then, combined with ground-based station observations, the residual values after removing the trend surface are calculated, and the sharply varying residual field is interpolated using HASM. Finally, the trend surface and the interpolated residual field are summed to obtain the final interpolation result. Verified results demonstrate that the simulation accuracy of this method is significantly higher than that of hybrid interpolation methods that generate trend surfaces via traditional interpolation approaches (IDW, Kriging, Spline), with favorable adaptability. At the global scale, the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) are 125.15 mm and 155.80 mm, respectively. This dataset consists of 2010 annual precipitation data obtained by deriving trend surfaces using TRMM data and various interpolation methods, with a spatial resolution of 0.25° × 0.25° (unit: mm). The dataset is stored in .tif format, comprising 20 individual data files with a total size of 64.2 KB (compressed into a single file of 48.8 KB). Research findings based on this dataset have been published in the *Journal of Geo-information Science*, Volume 17, Issue 8, 2015.
