Daily flow simulation in Thailand Part I: Testing a distributed hydrological model with seamless parameter maps based on global data

Abstract

Study region: Upper region of the Greater Chao Phraya River (GCPR) basin in Thailand. Study focus: This study presents a (∼1 km resolution) distributed hydrological model, wflow_sbm, with global spatial data and parameterization for estimating daily streamflow in the upper GCPR basin, with the aim to overcome in situ data scarcity often occurring in Southeast Asia. We forced the model with the MSWEP V2 precipitation and eartH2Observe potential evapotranspiration datasets. Seamless distributed parameter maps based on pedotransfer functions (PTFs) and literature review were applied to bypass calibration. Only the KsatHorFrac parameter determining the lateral subsurface flow was calibrated. A target storage-and-release-based reservoir operation module (ROM) was implemented to simulate reservoir releases. We compared the simulated daily streamflows obtained from different PTFs and evaluated the model performance in the period 1989–2014. New hydrological insights for the region: The global-data-driven wflow_sbm model can reconstruct daily streamflow in the upper GCPR basin, especially for natural catchments (KGE = 0.78). The ROM can capture the seasonal variability of reservoir releases, but not very accurately at the daily timescale (KGE = 0.43) since the actual reservoir operations are too complex. Different PTFs and KsatHorFrac values only introduce little uncertainty in the streamflow results. Therefore, the proposed model provides an opportunity for streamflow estimation in other ungauged or data-scarce basins in Southeast Asia. Nonetheless, the difficulty in the reservoir system modeling reflects the necessity of better understanding of human intervention on daily streamflow.

Keywords

• Wflow_sbm
• Reservoir modeling
• Data scarcity
• Global forcing data
• Pedotransfer function
• Chao Phraya basin