Hidrolojik Model Kalibrasyonunda Uydu Tabanlı Aylık Buharlaşma ve LAI Verilerinin Kullanılması

Hydrological model parameters are usually calibrated based on the performance of the daily river discharge data recorded at the basin outlet. In the unconvetional approach, satellite-based remote sensing data, which is open to everyone, is utilized in addition to the discharge data. The latter approach is aslo called spatial calibration in hydrology. The objective of this study is to assess the utility of remotely sensed actual evapotranspiration (AET) and monthly leaf area index (LAI) maps on the calibration of the hydrological model. For this, six different calibration cases (scenarios) are designed using a physically-based hydrologic model for the Vienne basin in France. It should be noted that LAI is used to estimate interception and correct the PET in the model i.e. both affecting the simulated AET. The daily discharge simulation performance of the model is assessed using KGE and the spatial performance of the model is assessed using SPAEF i.e. between mHM's long-term (2002-2014) monthly average AET raster output maps and reference MODIS-AET maps. According to the results, the KGE for scenario 1 (single AGI) was 0.91, SPAEF was below zero; for scenario 2 (with 4 AGI), the KGE was 0.37 while the SPAEF was positive; for scenario 3 (Single AGI and MODIS-AET), the KGE was 0.90 while the SPAEF were ~0.70; For the 4th scenario (MODIS-AET only), the KGE was -0.24, SPAEF was around 0.60, for scenario 5 (12 flow data and MODIS-AET) the KGE was 0.67, SPAEF was around 0.75; for scenario 6 (one-year daily flow and MODIS-AET) KGE 0.72 and SPAEF was around 0.75. Our results are promising even for poorly gaged basins as we could reach reasonable performance with satellite based AET and only 12 discharge measurements.