基于Sentinel-1及 Landsat 8数据的黑河中游农田土壤水分估算

土壤水分是陆地表层系统中的关键变量。利用主动微波遥感,特别是合成孔径雷达(Synthetic Aperture Radar,SAR)的观测,在监测和估计表层土壤水分时空分布方面已开展了诸多研究。然而,SAR土壤水分反演仍存在诸多挑战,特别是地表粗糙度和植被的影响。因此,本文提出了一种结合主动微波和光学遥感的优化估计方案,旨在同步反演植被含水量、地表粗糙度和土壤水分。反演算法首先在水云模型的框架下对模型中的植被透过率因子(与植被含水量密切相关)采用3种不同的光学遥感指数——修正的土壤调节植被指数(Modified Soil Adjusted Vegetation Index,MSAVI)、归一化植被指数(Normalized Difference Vegetation Index,NDVI)和归一化水体指数(Normalized Difference Water Index,NDWI)进行参数化估计,用于校正植被层的散射贡献。在此基础上,构造基于SAR观测和Oh模型的代价函数,利用复型洗牌全局优化算法进行土壤水分和地表粗糙度的联合反演。采用Sentinel-1 SAR和Landsat 8多光谱数据在黑河中游开展了反演试验,并利用相应的地面观测数据对结果进行了验证。结果表明反演结果与地面观测具有良好的一致性,其中基于NDWI的植被含水量反演效果最佳,与地面观测比较,土壤水分决定系数(R 2)在0.7以上,均方根误差(RMSE)为0.073 m^ 3/m^ 3;植被含水量R 2大于0.9,RMSE为0.885 kg/m 2,表明该方法能够较准确地估计土壤水分。同时发现植被含水量的估计结果,以及植被透过率的参数化方案对土壤水分的反演精度有一定的影响,在未来的研究中需要进一步探索。

Estimation of Soil Moisture of Agriculture Field in the Middle Reaches of the Heihe River Basin based on Sentinel-1 and Landsat 8 Imagery

Soil moisture is a key variable in land surface system. Using active microwave remote sensing observations, especially Synthetic Aperture Radar (SAR), has been proven a promising way on the estimation of spatial-temporal distribution of surface soil moisture by a lot of studies. However, there is still challenging in this field, because of the impacts caused by surface roughness and vegetation cover. In this context, this paper proposes an optimal estimation approach combined using SAR and optical remote sensing imagery, in order to retrieve vegetation water content, roughness and soil moisture simultaneously. First, water-cloud model is used to correct vegetation effect on microwave scattering process. In this step, vegetation transmittance factor (closed related to vegetation water content) is estimated by using three optical remote sensing indexes, namely, Modified Soil Adjusted Vegetation Index (MSAVI), Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). Second, a cost function is constructed based on SAR observations and Oh model simulations, then soil moisture and surface roughness can be estimated through global optimization by shuffled complex evolution algorithm. The proposed method is performed by using Sentinel-1and Landsat 8 data in the middle researches of the Heihe River Basin, retrieved results are validated against ground measurements. Results show a good agreement between remote sensing estimates and ground measurements, which indicates the proposed method can retrieve soil moisture accurately. For soil moisture, the determination coefficient (R ²) is higher than 0.7, the root mean square error (RMSE) is 0.073 m³/m³. With respect to vegetation water content,R ² is higher than 0.9 and RMSE is 0.885 kg/m². In the meantime, it is found that the result of estimated vegetation water content and the parameterization scheme of vegetation parameters have pronounced influence on the accuracy of soil moisture estimates, which need to be further addressed in future research.