Alternative Approach for Hydrological Predictions in Watersheds with Intermittent or Aggregated Observations

Código

I-EBHE0003

Autores

Nikunj Kalubhai Mangukiya, Ashutosh Sharma

Tema

WG 1.05: Comparative understanding of runoff generation processes

Resumo

Water resources management requires precise estimates of streamflow. However, due to limitations in monitoring infrastructure, many regions lack high-temporal-resolution data and often rely on intermittent or monthly aggregated observations. Hydrological predictions in such data-sparse watersheds remain challenging. This study proposes an alternate approach utilizing advanced data-driven models and their capabilities of learning long-term dependencies within hydrological variables and processes to enhance streamflow predictions in data-sparse regions. The proposed approach is evaluated for simulating daily flow patterns from monthly or weekly intermittent observations and monthly aggregates for two contrasting hydrological settings: near-natural watersheds and human-influenced watersheds with varied anthropogenic activities in each. The results demonstrate the capabilities of the proposed approach in reliably predicting daily flows from monthly or weekly intermittent observations, achieving a median Nash-Sutcliffe efficiency (NSE) of 0.70 for near-natural watersheds and 0.55 for human-influenced watersheds. The proposed approach also demonstrated reliable performance for daily flow predictions from monthly aggregates, with median NSE values of 0.61 for near-natural watersheds and 0.48 for human-influenced watersheds. The findings show that the proposed approach is robust across different seasons and hydrological conditions, with a median percentage bias within ±5%, except in arid regions. Moreover, the proposed model trained on intermittent and aggregated observations reasonably predicted daily streamflow in ungauged watersheds. Overall, this study demonstrates that the proposed approach provides a robust and accurate representation of daily streamflow patterns from aggregated or intermittent observations, offering hydrological insights and promising solutions for improving water resource management in regions with limited monitoring infrastructures.