摘要
降水量是构成生态系统的重要气候参数之一。然而，由于降水量观测站布设的局限性，如站点稀少，空间分布不均匀等，绝大多数空间位置上的降水数据无法获得，估算降水量的空间分布难以得到理想的结果。地形特征对降水量空间分布的影响非常重要，通常利用降水量与地形因子之间的相关关系评估降水量的空间分布。近年来，GIS技术中数字高程模型的引入和精确的地形分析能力大大提高了降水量空间分布研究的准确性，已成为降水量空间分布研究的重要手段之一。利用GIS技术建立了新疆降水量的空间分布模型，其中，地理和地形因子的选取来自lkm×1km的DEM数据，主要包括经度、纬度、海拔高度以及不同范围内不同方向上的最小和最大海拔高度等。此外，降水量资料来自1971—2005年新疆131个降水量测站（包括气象站和水文站），其中，112个站点的资料用于建立模型，剩余的19个站点的资料用于验证模型，研究结果表明模型是合适的，验证的模拟值与实际值之间的相关系数分别为在0．7870—0．9263之间，夏季较好，冬季较差。在新疆这个地形复杂的区域，利用GIS技术建立模型研究降水量的空间分布，对于研究新疆降水资源、农业气候资源区划以及生态环境变化等有一定的实际意义。
Precipitation constitutes one of the most important meteorological and climatological parameters for our e- cosystems. However, in most cases, the network of the precipitation measuring stations is sparse and available data are insufficient to characterize the highly variable precipitation spatial distribution. This is especially true for moun- tain areas, where the complexity of the precipitation distribution is combined with the measurement difficulties. Therefore, it is necessary to develop methods to estimate precipitation in areas where precipitation has not been measured, using data from the surrounding meteorological stations. Topography has the important effect on the spa- tial distribution of precipitation. Spatial correlations between precipitation and geographical and to tures can be used to estimation spatial distribution of precipitation. Our aim in this study is to develop an objective mapping and interpolation method that takes into account geographical variables, and uses GIS techniques to obtain spatial distribution of precipitation in Xinjiang in northwestern China. In this paper, an empirical methodology for modeling the spatial distribution of precipitation in Xinjiang using GIS was proposed. This methodology is based on a multiple regression modeling followed by kriging of residuals. In regression modeling , as dependent variables, we used annual and seasonal precipitation, whereas as independent variables, we used altitude, latitude, longitude, slope, aspect, etc. Relationship between precipitation and geographical and topographical variables （longitude, latitude, elevation, minimum and maximum elevation with different direction within given radius, etc） were used to improve the spatial distribution of precipitation estimation. The geographical and topographical variables were de- rived from the 1 km resolution digital elevation model （DEM）. Precipitation data, for the observation period 1971 to 2005, were assembled from 131 observation stations. Precipitation data from 112 observation stations were used to initialize the model. The data from the other 19 stations were retained to valid it. In this study, the results are satisfactory in the case of mean seasonal and annual precipitation over the period 1971 to 2005, with the correlation coefficients （R）, between the estimated and observed valued, of 0.926 3 （year）, 0. 805 9 （spring）, 0. 867 7 （summer）, 0. 807 5 （autumn）, and 0.787 0 （winter）. Through GIS technology, the spatial distribution maps were generated for mean seasonal and annual precipitation in Xinjiang. This paper investigates the application of a GIS to construct DEM-derived terrain variables that are used to interpolate seasonal precipitation data for every 1 km by 1 km grid cell in Xinjiang. The results of validation suggest terrain - precipitation model building using terrain variables, followed by kriging of the regression residuals, is useful for interpolating precipitation data. Interpolation results can be transferred to any grid point in the domain for which the characteristics of the terrain variables are known.
出处
《干旱区地理》
CSCD
北大核心
2010年第6期868-873,共6页
Arid Land Geography
基金
国家自然科学基金(40701148)
新疆维吾尔自治区自然科学基金(200821176)
中央级公益性科研院所基本科研业务费项目(IDM200604)
新疆气象局科学研究与技术开发项目(200718)
关键词
DEM
地理和地形因子
降水量
空间分布
新疆
DEM
geographical and topographical variables
precipitation
spatial distribution
Xinjiang
作者简介
李军（1974-），男，天津杨柳青人，博士，副研究员，主要从事农业遥感和地理信息系统应用研究．E-mail：zju_lijun@163．com
