Water Level Reconstruction Based On The Drought Indices And ENSO Indices:

Spatial and temporal water level variation is essential for hydrological extremes monitoring and the establishment of integrated water management system,which is helpful for productivity and regional development.It is extremely important for our research area(the Mekong River Basin and Yangtze River Basin),where people mainly make living by agriculture.Duo to the uneven spatially and temporally distribution of water level observation stations,it is necessary to use the satellite remote sensing method with wide spatial coverage to monitor river water level.However,the water level monitoring based on satellite altimetry directly or high resolution satellite images indirectly have some defects.Thus,the precipitation derived from Tropical Rainfall Measuring Mission(TRMM)and the terrestrial water storage(TWS)from Gravity Recovery And Climate Experiment(GRACE)are applied to estimate the water level of two basins in this study.And their corresponding drought indices(TRMM Standardized precipitation Index(TRMM-SPI)and GRACE Drought Severity Index(GRACE-DSI))are obtained by standardized procedure,which are then used to correlate with standardized water level.Based on the correlated relationships,estimated standardized water level are obtained,which can be transformed into water level(water level reconstruction)by inverse standardized process.In order to further improve the water level estimations of the TRMM precipitation and GRACE TWS and their corresponding drought indices,a weighting factor is introduced to calculate a weighted basin-averaged precipitation and TWS,and then the water level can be re-estimated according to the abovementioned steps.A traditional drought index(Palmer Drought Severity Index(PDSI))is also adopted to reconstruct water level,which can be used for a comparative analysis with satellite remote sensing drought indices.In addition,since both the hydrological regimes of Mekong and Yangtze River Basin are influenced by the El Nino-southern oscillation(ENSO),ENSO events are used to analyze and clarify the water level reconstructions derived from drought indices.And then ENSO indices including the sea surface temperature(SST)in the Nino 3.4 region,the southern oscillation index(SOI)and multivariate ENSO index(MEI)are directly applied to reconstruct water level according to the same procedure as drought indices.The conclusions of this paper are as follows:(1)The water level reconstruction in both Mekong and Yangtze River Basin based on PDSI are in good agreement with the observed water level,and PDSI performs better in the Mekong River Basin.In addition,PDSI derived water level reconstruction in the Mekong River Basin have been improved substantially with the assistance of ENSO information.(2)The water level estimation of TRMM precipitation and GRACE TWS and its corresponding drought indices(TRMM-SPI and GRACE-DSI)are consistent with the water level observations.The water level reconstruction results of TRMM-SPI and GRACE-DSI show a significant improvement compared with precipitation and TWS fitting results.In addition,the weighted process has substantially improved the water level estimation results for all variables.(3)The ENSO phenomenon has a significant effect on the all the above water level estimations,and its effect on the Mekong River Basin and the Yangtze River Basin is the opposite.The water level reconstruction based on ENSO indices and the observed water level match well,and ENSO indices perform better in the Mekong River Basin.In addition,for three kinds of ENSO indices,SST outperform in both Mekong and Yangtze River Basin.