Ensemble machine learning paradigms in hydrology: A review

Recently, there has been a notable tendency towards employing ensemble learning methodologies in assorted areas of engineering, such as hydrology, for simulation and prediction purposes. The diversity techniques available implementation hydrological sciences led to the development utilization different strategies implementation. This review paper explores refers advancement methods, including resampling methods (e.g., bagging, boosting, dagging), model averaging, stacking viz. generalized stacked, application fields hydrology. main topics this study cover subjects surface river water quality, rainfall-runoff, debris flow, icing, sediment transport, groundwater, flooding, drought modeling forecasting. general findings survey demonstrate absolute superiority using over regular (individual) In addition, boosting AdaBoost, extreme gradient boosting) have more frequent successfully implemented problems than stacking, dagging approaches.