黄河沿岸陕豫区土壤水分的空间变化特征及其驱动力因子分析

Analysis of Spatial Variability of Soil Moisture and its Driving Force Factors in the Shaanxi-Henan Region along the Yellow River

基于黄河沿岸陕西和河南境内的33个农业气象站从1992-2005年的实测土壤水分资料，计算了各站表层10、20和50cm各土层的多年平均植被可利用土壤水分储量，发现土壤水分储量存在明显的空间变异性。多年平均土壤水分储量呈现西北（陕北）低以及南、东南和东部（陕南和河南）高的变化态势。其中，在陕西绥德站出现水分的最低值，与黄土高原属于水分贫瘠区相对应。在河南信阳站出现水分最高值，与河南南部和陕南水分比较充沛相对应。土壤水分储量垂向分异不明显，各土层的土壤水分储量之间存在较好的相关性。通过简单线性相关分析发现，土壤水分储量与田间持水量呈现较强的正相关，与纬度呈较强的负相关。而与土壤容重、地理位置、气象要素等其他因子相关关系不显著。灌溉与土壤水分储量的相关系数比降水与土壤水分储量的相关系数大，说明该区灌溉是土壤水分的重要调节因子。通过主成分分析发现，对大多数站点而言，土壤水分变化受下垫面、气候和灌溉等的综合影响，其影响程度存在明显的空间分异性。

Based on the observed soil moisture data at 33 agro-meteorological stations in the Shaanxi Henan region along the Yellow River from 1992 to 2005, the vegetation available soil moisture in the unit of depth at 10, 20 and 50 em soil layer is calculated. It is found that soil moisture shows strong spatial variability with the values increasingfrom northwest, Shaanxi to south, east and southeast, Henaru Soil moisture is low in the loess plateau, with the minimum value at Suide station, Shaanxi Province. Soil moisture is rich in the south of Shaanxi and Henan, with the maximum at Xinyang station, Henan Province. The vertical variation of soil moisture is not obvious. By using linear correlation analysis between the soil moisture and the influential factors, it is found that there is an obvious positive correlation between soil moisture and field capacity and negative correlation between soil moisture and latitude. The correlation between soil moisture and irrigation is stronger than the correlation between soil moisture and precipitation The correlations between soil moisture and density, longitude and elevation above sea level are weak. By using principal component analysis, it is found that the driving force of soil moisture is not just as vegetation, soil or climate factor. It is the combination of such multiple factors to drive the spatial variability of soil moisture. The comprehensive effect of all the influential factors on soil moisture shows obvious spatial variability.