Snowmelt Modelling Approaches in Watershed Models: Computation and Comparison of Efficiencies under Varying Climatic Conditions

• Published in: Water resources management Vol. 28; no. 11; pp. 3439 - 3453
• Main Authors: Verdhen, Anand, Chahar, Bhagu R, Sharma, Om P
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer-Verlag 01.09.2014; Springer Netherlands; Springer; Springer Nature B.V
• Subjects: Algorithms; Altitude; Atmospheric pressure; Atmospheric Sciences; Civil Engineering; Climatic conditions; climatic factors; Computation; Computer simulation; Earth and Environmental Science; Earth Sciences; Earth, ocean, space; Efficiency; Energy balance; Environment; Exact sciences and technology; Geotechnical Engineering & Applied Earth Sciences; Hydrogeology; hydrologic models; Hydrology; Hydrology. Hydrogeology; Hydrology/Water Resources; Mathematical models; meteorological data; Precipitation; prediction; Runoff; Simulation; Snow; Snowmelt; Snowpack; spring; Stations; Stream flow; Studies; temperature; uncertainty; Vegetation; Water resources; Water resources management; Watersheds; Weather; Wind; Asia; Himalayas; Pakistan, Himalayas; Inter-comparison; Energy budget; Temperature index; Snowmelt model; Climate variability; algorithms; altitude; models; efficiency; probability; energy balance; digital simulation; climate; drainage divide; meltwater; drainage basins; winds; snow; water resource management; prediction; temperature
• ISSN: 0920-4741; 1573-1650
• DOI: 10.1007/s11269-014-0662-7

Most of the watershed models contain snowmelt-computing options but there are modelling difficulties in snow-covered watersheds either due to paucity of data or in addressing snowmelt computation weakly. The temperature index (TI) and/or energy balance (EB algorithms of HEC-1, NWSRFS, PRMS, SHE, SRM, SSARR, SWAT, TANK, and UBC models have been investigated. The performance has been evaluated at the point (station specific) snowmelt computation with and without snowpack accounting. The computations have been performed for Solang station at 2 485 m altitude located in the western Himalayas. Springtime weekly snow and meteorological data of 1 983, 2003, and 2008 have been used. Data year 2008 has been used for weekly simulation with the observed snowpack ablation. The probability of success in simulating the snowmelt using TI/EB of all the models in average is 0.77. Nash-Sutcliffe (NS) efficiency coefficients for simulation with snowpack accounting are found to vary between 0.84 and 0.97. Although NS coefficients for verification year 2003 are satisfactory (0.5 to 0.88) but snowmelt prediction/verification efficiency at an interval of 25 years (1983) is below average. However, verification on probability criteria for data year 1983 in the case of TI/EB is 0.63/0.48. Results from EB approach show wind dependent fluctuations. Uncertainty arises due to inter-decadal variability of the snowpack/snowmelt. The approach applied in this paper is valuable in order to have a quick evaluation of snowmelt algorithm before integrating it with any operational watershed model.