Hydrological Modelling using Satellite-Based Crop Coefficients: A Comparison of Methods at the Basin Scale

Abstract

The parameterization of crop coefficients (<i>kc</i>) is critical for determining a water balance. We used satellite-based and literature-based methods to derive <i>kc</i> values for a distributed hydrologic model. We evaluated the impact of different <i>kc</i> parametrization methods on the water balance and simulated hydrologic response at the basin and sub-basin scale. The hydrological model SPHY was calibrated and validated for a period of 15 years for the upper Segura basin (~2500 km²) in Spain, which is characterized by a wide range of terrain, soil, and ecosystem conditions. The model was then applied, using six <i>kc</i> parameterization methods, to determine their spatial and temporal impacts on actual evapotranspiration, streamflow, and soil moisture. The parameterization methods used include: (i) Normalized Difference Vegetation Index (NDVI) observations from MODIS; (ii) seasonally-averaged NDVI patterns, cell-based and landuse-based; and (iii) literature-based tabular values per land use type. The analysis shows that the influence of different <i>kc</i> parametrization methods on basin-level streamflow is relatively small and constant throughout the year, but it has a bigger effect on seasonal evapotranspiration and soil moisture. In the autumn especially, deviations can go up to about 15% of monthly streamflow. At smaller, sub-basin scale, deviations from the NDVI-based reference run can be more than 30%. Overall, the study shows that modeling of future hydrological changes can be improved by using remote sensing information for the parameterization of crop coefficients.