Assessing the effect of rating curve uncertainty in streamflow simulation on Kulfo watershed, Southern Ethiopia

• Published in: Journal of water and climate change 2024-09, Vol.15 (9), p.4199-4219
• Main Authors: Bekele Mena, Nahom, Ayele, Elias Gebeyehu, Chora, Henok Gubula, Dada, Teka Tadesse
• Format: Article
• Language: English
• Subjects: Accuracy; Artificial intelligence; Basins; Decision making; Discharge; Flood peak; High flow; Hydrologic data; Hydrology; Investigations; Neural networks; Rainfall; Resampling; Resource management; Simulation; Simulation models; Stream discharge; Stream flow; Streamflow data; Time series; Uncertainty; Water resources; Water resources management; Watersheds
• ISSN: 2040-2244; 2408-9354
• DOI: 10.2166/wcc.2024.645

Accurate streamflow simulation and comprehending its associated uncertainty are crucial for effective water resource management. However, the uncertainty of rating curves from which streamflow data is derived remains poorly understood. This study aims to simulate streamflow under rating curve uncertainty conditions. The bootstrap resampling technique (BSRT) was used to establish the rating curve and estimate associated uncertainty. Furthermore, it integrated with standalone and hybrid models (GRU, Bi-LSTM, and Conv1D-LSTM), to assess the effect of this uncertainty on streamflow simulation. Different lag times of rainfall and discharge are used as input for DL streamflow simulation models. Despite the complexity, the Conv1D-LSTM model did not outperform the Bi-LSTM model. However, it slightly outperforms the GRU model. Moreover, the rating curve uncertainty significantly propagates to streamflow simulation, particularly in high-flow regions. Consequently, the uncertainties related to rating curves on the Kulfo River led to a streamflow uncertainty of about 17.8 m3 s−1, representing 22% at peak discharge. The performance of the DL models was evaluated using different metrics (RMSE, MAE, NSE, and R2). The findings underscore the importance of considering rating curve uncertainty in streamflow simulation to enhance water resource management practices and support informed decision-making in the study area.