基于Sentinel-1和Sentinel-2协同反演干旱区地表土壤水分——以格尔木河中下游为例

Based on Sentinel-1 and Sentinel-2 Synergistic Inversion of Surface Soil Moisture in Arid Areas ——A Case Study of the Middle and Lower Reaches of Golmud River

中文关键词:  土壤水分  水云模型  Sentinel数据  NDWI

英文关键词:Soil moisture  Water cloud model  Sentinel data  NDWI

中文摘要:

      土壤水分是全球水循环的重要组成成分,对研究水分的空间分布、农作物长势和产量、气候变化、水资源时空特征等有着重要意义。利用Sentinel(哨兵)系列主动微波雷达卫星SAR(Sentinel-1)结合光学卫星(Sentinel-2)对格尔木中下游低矮植被覆盖下的地表土壤水分进行反演研究,探讨不同极化组合方式和水云模型前后的土壤水分含量反演方法的适用性。结果表明,VV(VV Polarization)极化比VH(VH Polarization)极化更加适用该区域,VV极化结合归一化水指数(NDWI)反演地表土壤水分精度达到42.6%,拟合精度最高；VH极化仅为22.6%；利用水云模型去除植被覆盖后对地表土壤水分的反演精度有所提升,VV极化精度提高约3.5%,VH极化提高1.5%；Sentinel系列卫星影像对于干旱区的土壤水分的反演具有较好的适用性。探索了适用于该研究区乃至柴达木盆地土壤水分大面积实时监测的可靠手段和依据。

英文摘要:

      Soil moisture is an important component of the global water cycle.It has important significance for studying the spatial distribution of soil moisture,crop growth and yield,climate change and so on.In this paper,the Sentinel series of active microwave radar satellites SAR(Sentinel-1)combined with optical satellites(Sentinel-2)are used to invert the surface soil moisture under the cover of low vegetation in the middle and lower reaches of Golmud River,and then to discuss the applicability of different polarization combinations and water-cloud models for the inversion of soil moisture content.The results show that the VV(VV Polarization)polarization is more suitable for this area than the VH(VH Polarization)polarization.The VV polarization combined with the Normalized Difference Water Index(NDWI)inverts the surface soil moisture with an accuracy of 42.6%,the accuracy of the fitting is the highest;while the accuracy of the VH polarization is only 22.6%.After using the water cloud model to remove the vegetation cover,the inversion accuracy of surface soil moisture has been significantly improved;the VV polarization accuracy is increased by about 3.5%,and the accuracy of the VH polarization is increased by 1.5%；Sentinel series of satellite images have good applicability for inversion of soil moisture in arid areas.This paper aims to explore a reliable basis and suitable method for real-time monitoring of large-scale soil moisture in this study area and even the Qaidam Basin.