基于Aster数据的新疆渭-库绿洲及其周边地区土壤盐分的定量反演

Soil Salinity Retrieval in the Weigan and Kuqa River Delta Oasis area with Aster date

以新疆渭干河-库车河地区为研究区域,在野外调查采样的基础上,对土样进行实验室光谱测量并重采样与Aster波段相匹配,利用偏最小二乘回归建模方法建立了土壤盐渍化定量反演模型,其精度满足大区域的土壤盐渍化监测要求,表明该建模方法具有较好的普适性和稳定性。用10景Aster图像数据实现了该区域的土壤盐渍化定量反演与制图,反演的盐分分布与实地调查较为一致,为大面积区域性土壤盐渍化的遥感定量调查与监测提供了较为有效的技术方法。

In this study, based on field investigation, chemical collected from Weigan and Kuqa River Delta Oasis region were analysis and spectral measurements ot sahne soil samples conducted in laboratory. The spectra of soil samples were resampled first to match Aster spectral resolution. Then the soil salinity retrieval model was established between soil salinity contents and spectra of soil samples using partial least squares regression （PLSR） method. The model was applied to 10 scenes Aster reflectance images to quantitatively map soil salinity in the region. The distribution map of soil salinity consistent well with field investigate. The result demonstrated that combining multispeetral remote sensing data and PLSR method is a suitable technique for quantitative retrieve soil salinity in a large region.