Calibration of a Distributed Hydrologic Model Using Streamflow and Remote Sensing Snow Data

The Variable Infiltration Capacity (VIC) hydrologic model is applied to the headwaters of the Arkansas River in the USA for the purpose of water supply evaluation. Modelling the hydrologic regime of the Arkansas River is a challenge due to the large number of diversions and regulations that might impact the natural streamflow. Meanwhile, most of the headwaters in the Arkansas River are snow-melt dominated, further complicating the modelling procedure. A snow cover dataset can provide additional information during the model calibration process. In this study, fractional snow covered area (SCA) data was acquired from the Moderate Resolution Imaging Spectrometer (MODIS) satellite, and two independent calibrations were performed for streamflow and SCA using a Genetic Algorithm (GA) to understand the benefits from employing SCA as another calibration target. The calibration results for streamflow show good model fits to the observation (Nash Sutcliffe Efficiency (NSE) of 0.91 and 0.78 for calibration and validation, respectively). For the calibration focusing on SCA, the results show that it is possible to calibrate a VIC model without using observed streamflow (NSE = 0.58 and 0.65 for calibration and validation, respectively); this result supports the potential use of SCA for the calibration of distributed hydrologic models when observed streamflow values are unavailable or impaired.