Search

Drought is one of the most challenging climatic events. Recently, the drought influence in East Africa (EA) total water storage (TWS) is a serious problem, particularly in arid areas with modified natural vegetation relying on water deficit, garnered extensive research interest. Hydro-climatological and vegetation indices and remote sensing datasets derived from Gravity Recovery Climate Experiment (GRACE) mission datasets reveal good performance in analyzing hydrological drought influences in water storage. Over the last decades, studies were considered successful in monitoring the drought influence in the region TWS potential. However, several challenges remained unsolved, hindering the hydrological drought mitigation strategies. This review deals with an overview of drought impact monitoring targeted at the TWS variation with the response of vegetation change for sustainable drought mitigation. To improve the flexibility and adaptive capacities of the water deficit problem, we aim to provide an overview of drought impacts on TWS in the region to redefine the hydro-climatological and vegetation drought indices and improve the understanding of drought impact through remote sensing datasets. This review presents the challenges and prospects and offers a conclusion. Although, we hope that the review can facilitate further study regarding future hydrological drought projection in the development of several scientific research in the field.

Soil salinization is recognized as a key issue negatively affecting agricultural productivity and wetland ecology. It is necessary to develop effective methods for monitoring the spatiotemporal distribution of soil salinity at a regional scale. In this study, we proposed an optimized remote sensing-based model for detecting soil salinity in different depths across the Yellow River Delta (YRD), China. A multi-dimensional model was built for mapping soil salinity, in which five types of predictive factors derived from Landsat satellite images were exacted and tested, 94 in-situ measured soil salinity samples with depths of 30–40 cm and 90–100 cm were collected to establish and validate the predicting model result. By comparing multiple linear regression (MLR) and partial least squares regression (PLSR) models with considering the correlation between predictive factors and soil salinity, we established the optimized prediction model which integrated the multi-parameter (including SWIR1, SI9, MSAVI, Albedo, and SDI) optimization approach to detect soil salinization in the YRD from 2003 to 2018. The results indicated that the estimates of soil salinity by the optimized prediction model were in good agreement with the measured soil salinity. The accuracy of the PLSR model performed better than that of the MLR model, with the R2 of 0.642, RMSE of 0.283, and MAE of 0.213 at 30–40 cm depth, and with the R2 of 0.450, RMSE of 0.276, and MAE of 0.220 at 90–100 cm depth. From 2003 to 2018, the soil salinity showed a distinct spatial heterogeneity. The soil salinization level of the coastal shoreline was higher; in contrast, lower soil salinization level occurred in the central YRD. In the last 15 years, the soil salinity at depth of 30–40 cm experienced a decreased trend of fluctuating, while the soil salinity at depth of 90–100 cm showed fluctuating increasing trend.