基于Sentinel2A影像的干旱区土壤盐渍化监测研究

Research on Monitoring of Soil Salinization in Arid Area Based on Sentinel2A Image

以渭干河-库车河三角洲绿洲为例,在5种盐分光谱指数(NDSI、SI、SI1、SI2、SAVI)基础上,选择Sentinel2A多光谱影像中的3个红边波段和近红外窄波段进行扩展,提出了新的基于红边和近红外窄波段的光谱指数,用偏最小二乘回归(PLSR)方法构建土壤电导率遥感反演模型。结果表明:相对传统光谱指数,引入红边1波段(B_(5))和近红外窄波段(B_(8a))的光谱指数与土壤电导率相关性显著提高;基于Sentinel2A影像的PLSR的土壤电导率信息建模效果较好(R^(2)=0.77),模型准确、可靠。盐分反演结果表明研究区土壤盐分含量整体较高,盐渍化影响范围广泛。自绿洲内部向外围由非盐渍化向轻度-中度-重度盐渍化过渡,绿洲西部、南部轻度盐渍土分布较多,绿洲东部、东南及西南方向有重度盐渍土及极重度盐渍土分布,盐渍化程度较高。

Taking the Weigan-Kuqa River Delta Oasis as an example,based on five salt spectral indices(NDSI,SI,SI1,SI2,SAVI),three red edge bands and near-infrared narrow bands in Sentinel2A multispectral images were selected for expansion.A new spectral index based on red edge and near-infrared narrow bands was proposed,and a partial least squares regression(PLSR)method was used to construct a soil conductivity remote sensing inversion model.The results showed that compared to traditional spectral index,the correlation between the spectral index introduced in the red edge 1 band(B_(5))and the near-infrared narrow band(B_(8a))and soil conductivity was significantly improved;the modeling effect of soil conductivity information using PLSR based on Sentinel2A images was good(R^(2)=0.77),and the model was accurate and reliable.The results of salt inversion indicated that the soil salt content in the study area was generally high,and the impact of salinization was extensive.There was a transition from non salinization-mild-moderate-severe salinization from the interior to the exterior of the oasis.There were more mild saline soils distributed in the western and southern parts of the oasis,while there were severe saline soils and extremely severe saline soils distributed in the eastern,southeast and southwest directions of the oasis,with a higher degree of salinization.