甘肃省近30年平均降水空间模拟方法的研究

生态环境治理以及全球变化研究的需要强化了部分自然地理要素空间模拟研究的重要性。区域多年平均降水量是影响农业生产的主要气象因子,是合理规划种植制度的重要指标。降水资源的研究是有限的降水资源合理利用和配置的依据。本研究利用ArcMap地理信息系统,对国际主要的4种插值算法(距离权重法、趋势面法、样条函数法、普通Kriging法)进行分析,建立甘肃降水空间数据库,计算并比较分析了甘肃近30年平均降水量的时空变化。对各方法进行比较研究,然后根据插值精度选择其中精度最高的一种进行甘肃近30年的降水模拟。研究结果表明:在平均降水量模拟的4种算法中,普通Kriging的计算精度最高。利用最优的插值方法的栅格降水空间数据库,建立的多年平均降水资源信息系统,可快速计算研究区内任一地域单元中降水的总量及其空间变化,对降水资源模拟、工程规划、发展战略及决策管理,均具有重要的意义。

Simulation Methods on Annual Rainfall in Recent 30 Years in Gansu

The spatial simulating methods of some physical geographical element is becoming increasingly important nowadays in environment improvement and the research of global change.Study on the spatial changes of Average Annual Rainfall(AAR) is not only a basis of the engineering planning for rain-harvesting agriculture,but also an important index for the reasonable utilization of limited rainfall water resources.Using ARCMAP GIS technology,the average annual rainfall databases of recent 30 years in the study areas were established.By means of 4 methods(inverse distance weighted,trend,spline,ordinary Kriging),the spatial and temporal changes of AAR were calculated and comparatively analyzed.Results indicated that simulating precision of ordinary Kriging method is the best one in the approaches.The AAR information system,established by raster rainfall spatial databases using optimized methods,can quickly and timely calculate the total quantity and its spatial changes on any scales in the study areas,which has an important role in runoff simulating,engineering planning,strategy developing,and decision making as well as management for rainfall harvesting agriculture.