Modeling of Areal Coverage of Snow of an Ungauged Catchment with ArcSWAT

Aim: The study aimed at modeling the aerial extent of snow cover of an ungauged mountainous Himalayan region using the temperature index-based method of ArcSWAT model.

Study Design: 20 year precipitation and temperature data along with elevation information were used in the simulation of accumulated snow.

Place and Duration of Study: Department of Civil Engineering, Indian Institute of Technology Delhi, India from January 2014 to June 2014.

Methodology: The basin was divided into 3 elevation bands and daily snow accumulation depths were obtained for each of the elevation zones. To account for the lack of measured snow depths, satellite imagery was used to calibrate the model. LandsatLook imagery taken on different dates in a year was visually interpreted for the presence of snow cover in the different elevation zones. In addition, image classification was used to identify snow covered region in each elevation band and to determine the percent area under snow cover. Temperature and precipitation lapse rates were alternately adjusted till the simulated results were in agreement with the results obtained from the imagery. Simulation was deemed to be acceptable whenever a non-zero snow depth was simulated by the ArcSWAT model for above 5 percent area under snow determined from the satellite imagery.

Results: Calibration resulted in a temperature lapse rate of -6°C/km and a precipitation lapse rate of 5mm/km for the region. Snow accumulation depths obtained from the calibrated model for all elevation zones agreed reasonably well with the results obtained from image classification.

Conclusion: ArcSWAT could be suitably used to model the snow cover of ungauged hilly catchments. Satellite imagery/remote sensing data can be a suitable aid to calibrate the snow model for ungauged regions. Division into greater number of elevation zones is expected to improve the calibration process.