Quantifying Drought Impacts on Reservoir Operations with DWAT: The Obong Reservoir Water Crisis

In September 2025, Obong Reservoir(supplying ~87% of the city’s domestic water) in Gangneung-si, South Korea experienced a severe drawdown that triggered citywide rationing. Reported effective storage fell to about 11% on 12 Sep 2025, with the water level near 99.5 m, only 7 m above the implied dead-water line. This case study applies the Dynamic Water Resources Assessment Tool (DWAT) to (i) reproduce the observed 09/2025 drawdown, (ii) diagnose dominant drivers of the low-water crisis, and (iii) quantify the rainfall threshold required for short-term recovery. DWAT is a hydrological modeling framework designed for water-resources assessment across diverse regions worldwide. It allowing detailed characterization of both short- and long-term hydrologic behavior. DWAT represents key processes such as surface runoff, groundwater flow, and human water use (e.g., irrigation and municipal withdrawals), supporting integrated evaluation of water availability and its movement through a watershed.

For reservoir operation, simulations incorporated spillway/outlet/intake characteristics, the stage–area–storage relationship, and time-varying withdrawal data reflecting operational conditions. The simulation period spanned from January 2023 to September 2025, including a one-year warm-up period and the major drought period affecting Gangneung. Results confirm that DWAT accurately reproduces the progressive water-level decline over multiple seasons, the sharp drawdown in late summer 2025, and the transition into the near dead-storage zone in both timing and magnitude. Water-balance diagnostics indicate that the Obong watershed is strongly storage-dependent (surface runoff is less than 3.4%), such that prolonged drought markedly reduces event-driven inflow, depletion of soil moisture and groundwater weakens baseflow support, and continued pumping accelerates reservoir water-level decline. Recovery experiments using the calibrated model show that reservoir stage can return to the normal operating range and that restoration of soil moisture and groundwater storage requires at least ~200 mm of rainfall.

Overall, the DWAT-based drought simulation demonstrates that DWAT is well suited for integrated drought assessment and reservoir operation analysis, providing a practical tool for diagnosing low-water crises and for identifying actionable recovery thresholds that can support emergency response planning and adaptive water-supply management under increasing hydroclimatic variability.

This work was supported by Korea Environment Industry & Technology Institute(KEITI) through Water Management Program for Drought Project, funded by Korea Ministry of Climate, Energy and Environment(MCEE).(2022003610002)