Alternative solutions for long missing streamflow data for sustainable water resources management
Buket Mesta, Omer Burak Akgun, Elçin Kentel
Abstract
Sustainable water resources management requires long time series of streamflow data. In this study, a Takagi–Sugeno fuzzy rule-based (FRB) model is developed to reconstruct long periods of missing daily streamflow data which is a common problem in developing countries. The FRB model uses observations of neighbouring stream gauges, and thus is advantageous regarding data and time requirement compared to physical models. With the proper set of inputs, the FRB model provides better estimates than the hydrological model at two of the studied four stream gauges in the Meric–Ergene Basin. Filling long data-gaps with FRB models will facilitate the development of realistic water management strategies.
Topics & Concepts
StreamflowWater resourcesMissing dataComputer scienceData setSet (abstract data type)Environmental scienceHydrology (agriculture)Data miningDrainage basinGeologyGeographyArtificial intelligenceMachine learningGeotechnical engineeringCartographyProgramming languageEcologyBiologyHydrological Forecasting Using AIHydrology and Watershed Management StudiesWater resources management and optimization