土壤湿度指数在黄土高原的适宜性评价

利用1992—2000年间黄土高原地区逐月降雨量、实测土壤湿度数据,结合GIS遥感技术、相关分析等方法,对基于TU-Wien变化检测算法,从ERS散射计数据反演获取的土壤湿度指数(SWI)或表层土壤湿度(SSM),从点和面2个尺度进行了验证。结果表明,在不同的土地利用、土壤质地和地形条件下,SWI或SSM与降水呈正相关(P<0.01)。时间序列方面,降雨、实测值和SWI或SSM基本上呈现出相同的变化趋势;空间分布上,降水和SWI具有相似的分布特征。在点尺度上,选取的7个站点的SWI与降水呈正相关(P<0.01),SWI和实测值的相关性随土层深度加深而下降,仅在10cm处达到显著水平(P<0.01)。对表层10cm的实测数据而言,降水和SWI的相关性要好于其与实测值的相关性。在黄土高原区域尺度上,SWI或SSM能够较为准确地揭示该地区表层土壤湿度的演变和空间分布特征。有长达20年(1992—2011年)之久的数据积累,并且可以免费及时获取的SWI或SSM,对于大范围实测土壤湿度数据匮乏的黄土高原地区而言,是一种具有较高利用价值的土壤湿度监测数据产品。

Evaluation of the Soil Water Index and Surface Soil Moisture on the Loess Plateau

In this paper,the capability of the European Remote Sensing Satellite(ERS) scatterometer-derived soil water index(SWI) and surface soil moisture(SSM) data which are retrieved by a change detection method,developed at the Institute of Photogrammetry and Remote Sensing(IPF),Vienna University of Technology(TU-Wien) to disclose soil moisture and precipitation distribution on Loess Plateau is investigated with the help of GIS,remote sensing and correlation analysis.Monthly averaged SWI/SSM data for the years 1992—2000 are analyzed to evaluate the use of the SWI/SSM as an index to monitor soil moisture on Loess Plateau and to investigate whether it can reflects general precipitation distribution characteristics in this area or not.Monthly averaged in situ relative soil moisture and monthly precipitation data from Chinese meteorological gauge stations are employed to perform comparisons with SWI/SSM on point and regional scales.The result is that in different land use,soil texture and topography conditions SWI/SSM correlate well with precipitation and in situ relative soil moisture,most of them reach significant level,they have same change tendency,SWI and precipitation have similar spatial distribution characteristics.Among the seven stations,there is a good correlation between precipitation and soil water content(SWI and in situ relative soil moisture),especially for first 10 cm layer,the correlation between precipitation and SWI is better than that of precipitation and in situ moisture.The results show that SWI/SSM can relative accurately disclose soil moisture variation and spatial distribution characteristic on the Loess Plateau.Furthermore,the freely available 20-yr(1992—2011) time series SWI/SSM data are thus a valuable tool to overcome the scarcity of in situ soil moisture observation,which are usually not available on regional scale.