基于实测高光谱和Landsat 8 OLI影像的土壤盐化和碱化程度反演研究

Inversion of Soil Salinity and pH Degree Based on Measured Hyperspectral and Landsat 8 OLI Image

针对宁夏银北地区土壤盐碱化定量监测的需要,利用实测土壤高光谱和Landsat 8 OLI多光谱影像数据采用多项式、多元线性回归等方法进行土壤含盐量和pH值反演研究,并对影像光谱反演模型进行校正,以提高遥感定量反演精度。结果表明:(1)基于实测光谱的土壤含盐量反演精度均高于基于OLI影像反演精度;基于实测光谱敏感波段反射率反演精度高于实测盐分指数反演精度,其中实测光谱经平滑后敏感波段建立的模型效果最佳(R^2=0.695)。(2)基于实测光谱平滑后敏感波段建立的pH值反演模型精度最高且最稳定(R^2=0.545),基于OLI影像光谱反演精度低于实测光谱,但也通过了显著性检验和精度验证。(3)经实测光谱模型校正后的Landsat 8 OLI影像光谱的土壤含盐量反演模型R^2从0.347提高到0.623。研究结果可以为准确、快速地定量监测当地土壤盐分含量、pH值的变化提供科学依据和技术手段。

The purpose of this study was to improve the precision of salinity predicting model in Yinchuan,Northern Ningxia.The measured Hyper-spectral and Landsat 8 OLI multi-spectral image data were used to carry out the inversion of soil salt content and pH based on salinity index and sensitive wavelength,and then OLI image inversion of soil salinity was verified by using the measured data.The results showed that:(1)The inversion accuracy of soil salt content based on the measured field spectral model was higher than that based on the OLI image model.The inversion accuracy of the measured spectral reflectance at the sensitive wavelength was higher than that of the measured salinity index.The model smoothing with measured sensitive wavelength showed the best accuracy(R^2=0.695).(2)For soil pH,the prediction model established at 572 nm and 573.5 nm as sensitive wavelengths with measured smoothing was the most accurate and stable(R^2=0.545).The accuracy of spectral inversion based on OLI image was lower than that of measured spectral data,but it was also tested and verified with significance analysis.(3)After the measured Hyper-spectral verification,the inversion model of soil salt content based on Landsat 8 OLI multi-spectral image increased the coefficient of determination(R^2)from 0.347 to 0.623.The results can provide a scientific basis and technique for accurately,rapidly and quantitatively monitoring local soil salinity and pH.