A New Method for Long-Term River Discharge Estimation of Small- and Medium-Scale Rivers by Using Multisource Remote Sensing and RSHS: Application and Validation

Abstract

River discharge is an important hydrological parameter of river water resources. Especially in small- and medium-scale rivers, data deficiency is the biggest problem for studies of river discharge. In recent years, remote sensing has become a rapid and convenient method to estimate river discharge. However, remote sensing images still have some difficulty generating continuous long-term river discharge. To address this problem, we developed a new method coupling the remote sensing hydrology station method (RSHS) with statistical regression downscaling, using data from optical satellites (Landsat-8, Sentinel-2), radar satellites (Sentinel-1), and un-manned aerial vehicles (UAVs). We applied this method to monitor monthly river discharge for small- and medium-scale rivers from 2016 to 2020 on Yunnan-Guizhou Plateau and evaluated the accuracy of the results. The results show that (1) by applying the newly constructed method, the water body continuity index obtained by Landsat-8 increased by 7% and the average river length percentage in the channel reached 90.7%, a 40% increase; (2) there were only 10 river flow data points, on average, in the 5-year period obtained before this method was applied; after this method was applied, more than 50 river flow data points could be obtained, on average, extending the quantity of data fivefold; in addition, improper extreme values could also be avoided; (3) with better continuity of water body distribution, the images provided steadier river widths. The relative error of daily flow estimation from Landsat-8 images was reduced by 60% and the mean percentage error was reduced by one-fourth. The relative error of the multisource remote sensing composited flow was reduced by 37% with a reduction in the mean percentage error of over a half; (4) in addition, we found that when the threshold difference between water bodies and land in remote sensing images is more than 0.2, the impact of water body recognition error on flow accuracy can be ignored. This method helps to overcome the absence of remote sensing methods for the long-term estimation of flow series in small- and medium-scale rivers, improves the accuracy of remote sensing methods for calculating flow, and provides ideas for regional water resource management and utilization.

Detection and Analysis of the Variation in the Minimum Ecological Instream Flow Requirement in the Chinese Northwestern Inland Arid Region by Using a New Remote Sensing Method

“…To obtain river discharge data for the study area, we adopted the remote sensing hydrological station (RSHS) technique [24,25] to monitor the discharge and conduct inversion for the 7 main rivers in the Aiding Lake Basin. The main technical aspects were three-dimensional modeling of low-altitude remote sensing data, field data acquisition, remote sensing flow estimation, and hydrological model simulation analysis based on hydrology, hydraulics, and quantitative remote sensing [25,26]. A remote sensing flow estimation system was established in the interactive data language development environment, coupled with an ecohydrological model system, and a model establishment method was constructed suitable for medium and small rivers in areas lacking hydrological data [26].…”

Enhancing hydrological model calibration through hybrid strategies in data‐scarce regions

“…Both edges have increased attention in deriving streamflow approximation from remote sensing (Huang et al, 2020; Sichangi et al, 2016). However, remote sensing datasets have their own limitation while estimating the discharge of medium to small size stream due to its spatial resolutions (Anand et al, 2023; Lou et al, 2022).…”

Perspectives of human–water co‐evolution of blue–green water resources in subtropical areas

“…Currently, RSHS technology has been applied to the Aibi Lake Basin, Qinghai-Tibetan Plateau, Changbai Mountain, and desert rivers (Lou et al, 2020;Wang et al, 2020;Wufu et al, 2021;Yang et al, 2020Yang et al, , 2022Yang, Wang, et al, 2021). The accuracy of the digital river model obtained by UAV can attain the centimetre level (Zhang et al, 2018), and the verification results show that RSHS technology is suitable for runoff monitoring in medium and small ungauged basins (Lou et al, 2022;Yang et al, 2020;Yang, Wang, et al, 2021).…”