中国中西部地区年降水量空间分布数据集(2010)

Spatial Distribution Dataset of Annual Precipitation in Midwestern China(2010)

中国中西部地区年降水量空间分布数据集(2010)是采用混合插值法进行高精度曲面建模(High Accuracy Surface Modeling,HASM)区域降水模拟的数据结果。该方法包括三个步骤:首先使用TRMM数据(或者站点观测值空间插值结果)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,并利用HASM对残差进行插值,得到空间中变化不平稳的残差场;最后将趋势面与残差场进行加和,完成插值计算。结果表明,TRMM作为背景场的HASM模拟精度明显高于传统插值方法,适应性较好。采用平均绝对误差(Mean Absolute Error,MAE)和均方根误差(Root Mean Square Error,RMSE)作为精度评价指标完成精度验证。在全局尺度上,MAE和RMSE分别为125.15、155.80 mm;在局部尺度上,MAE和RMSE分别为167.53、228.81 mm。与本数据集相关的科学发现论文已发表在《地球信息科学学报》2015年第17卷第8期。

The dataset on the spatial distribution of annual precipitation in midwestern China(2010)is a simulation of regional precipitation with High Accuracy Surface Modeling(HASM).The method includes three steps:First,use the TRMM data(or station observations spatial interpolation results)to represent the stable change of precipitation in space,that is,the trend surface;second,calculate residue combined with station observations after removing the trend surface,and then incorporate residue in HASM to form unstable changed residual field in space;finally,add the trend surface and residual field to complete incorporation.The results show that HASM model with TRMM as the background field is significantly more precise and adaptive than traditional interpolation methods.Mean absolute error(MAE)and root mean square error(RMSE)are the accuracy evaluation indicator.The MAE and RMSE of TRMM based HASM simulation were 125.15 mm and 155.80 mm on a global scale and 167.53 mm and 228.81 mm on a local scale,respectively.Related research results were published in Journal of Geo-Information Science(Vol.17,No.8,2015).