Journal of Water and Climate Change (Dec 2021)

Quantifying the effects of watershed subdivision scale and spatial density of weather inputs on hydrological simulations in a Norwegian Arctic watershed

  • Minh Tuan Bui,
  • Jinmei Lu,
  • Linmei Nie

DOI
https://doi.org/10.2166/wcc.2021.173
Journal volume & issue
Vol. 12, no. 8
pp. 3518 – 3543

Abstract

Read online

The effects of watershed subdivisions on hydrological simulations have not been evaluated in Arctic conditions yet. This study applied the Soil and Water Assessment Tool and the threshold drainage area (TDA) technique to evaluate the impacts of watershed subdivision on hydrological simulations at a 5,913-km2 Arctic watershed, Målselv. The watershed was discretized according to four TDA scheme scales including 200, 2,000, 5,000, and 10,000 ha. The impacts of different TDA schemes on hydrological simulations in water balance components, snowmelt runoff, and streamflow were investigated. The study revealed that the complexity of terrain and topographic attributes altered significantly in the coarse discretizations: (1) total stream length (−47.2 to −74.6%); (2) average stream slope (−68 to −83%); and (3) drainage density (−24.2 to −51.5%). The spatial density of weather grid integration reduced from −5 to −33.33% in the coarse schemes. The annual mean potential evapotranspiration, evapotranspiration, and lateral flow slightly decreased, while areal rainfall, surface runoff, and water yield slightly increased with the increases of TDAs. It was concluded that the fine TDAs produced finer and higher ranges of snowmelt runoff volume across the watershed. All TDAs had similar capacities to replicate the observed tendency of monthly mean streamflow hydrograph, except overestimated/underestimated peak flows. Spatial variation of streamflow was well analyzed in the fine schemes with high density of stream networks, while the coarse schemes simplified this. Watershed subdivisions affected model performances, in the way of decreasing the accuracy of monthly streamflow simulation, at 60% of investigated hydro-gauging stations (3/5 stations) and in the upstream. Furthermore, watershed subdivisions strongly affected the calibration process regarding the changes in sensitivity ranking of 18 calibrated model parameters and time it took to calibrate. HIGHLIGHTS The annual mean PET, ET, and lateral flow slightly decreased, while rainfall, surface runoff and water yield slightly increased with the increase of TDAs.; The fine TDAs produced finer and higher ranges of snowmelt runoff volume across the watershed.; All TDAs had similar capacities to replicate the observed tendency of monthly mean streamflow hydrograph, except overestimated/underestimated peak flows.; Spatial variation of streamflow was well analysed in the fine schemes compared to the coarse ones.; The scales of watershed subdivisions affected model performances, and sensitivity ranking of 18 calibrated model parameters in five hydrological subgroup processes (e.g. surface runoff, lateral flow, snowmelt, channel water routing, and goundwater) and time taking for calibration.;

Keywords