A New Automatic Hydrological Station Relocation Algorithm (ASRA) for Moving Hydrological Stations Onto a Simulated Digital River Network

Determining the precise placement of hydrological stations on a simulated digital river network is crucial for constructing hydrological models applicable to process simulation, water resource management, and flood forecasting endeavors. To solve this problem, we categorized and scrutinized deviations between the simulated and their actual station locations, and proposed a novel automatic hydrological station relocation algorithm (ASRA). The algorithm was first validated in the Amazon Basin using Global Runoff Data Centre (GRDC) hydrological stations and 90 m x 90 m Shuttle Radar Topography Mission (SRTM) data, successfully correcting the spatial position and corresponding catchment area (CCA) of each station. Findings revealed that CCA inaccuracies were notably decreased, transitioning from an initial 7.62% when employing a conventional 5-km search radius to 5.43% after adopting an iteratively optimized, objective, and rational 8-km search radius. The ASRA method was subsequently applied to GRDC stations within the HDMA and HydroSHEDS data sets, successfully repositioning 8,339 and 8,026 stations respectively, all with catchment area deviations of less than 5%, thus either exceeding or at least equaling the precision of prior research efforts. A Python program was developed and incorporated into an ArcGIS toolbox that features user-friendly attributes, enabling swift computation and accurate rectification, as a result of building upon our method. In short, our study presents a fresh approach and a robust tool for tackling the inconsistencies of hydrological station locations. The updated global GRDC hydrological station locations specifically tailored for both HDMA and HydroSHEDS data sets, together with the toolbox developed, were accessible for download on the figshare platform. As a result of the inherent limitations posed by DEM resolution, hydrological stations' actual positions frequently diverge from their corresponding simulated locations on the digital river network derived from DEM data. These discrepancies can vary across different DEM data sets. While traditional manual methods have been reliable in terms of accuracy, they suffer from inefficiency, while automatic approaches, conversely, have offered speed but at the cost of precision. Hence, this study introduces an automated method and ArcGIS-based toolbox that combines both high accuracy and efficiency, exemplifying its application in the Amazon basin as well as globally. The deviation of hydrological station locations on a digital river network were categorized into three types according to the catchment area and location A new ASRA was developed to deal with all deviation types with the advantages of convenience, efficiency, and easy integration with ArcGIS This method provides an objective and reasonable search radius to decrease the catchment area deviation based on an empirical search radius