Determining the basic parameter in snow melt process for estimating flood hydrograph in Karoun basin

Mountainous area has a considerable share in supplying water demand in the arid regions. Because, snow consists a large part of the rainfall in such regions which could be saved in the area located in the high elevations for a long time. Therefore, snowmelt runoff consists a large part of the total runoff in the mountain watersheds. The snow hydrology is much complicated than rainfall hydrology since there is always a lag time in occurring snowmelt runoff in the basin's outlet. Thus, finding out simple methods for calculating such runoff is so much important. In this research, after calibrating and validating the critical temperature and degree-days coefficient, the HEC-HMS model was applied to simulate the snowmelt runoff in Karoun basin, which is one of the most important mountain watersheds in Iran.
In this research, a large part of Karoun watershed upto PolShaloo outlet was selected. This region is located in the southern-west of Iran which lies between 50o23'37" to 51o54'29" E longitude and between 30o16'50.4" to 31o32'18.3" N latitude. About 50 percentage of total precipitation rains as snow and usually occurs during November-May. HEC-HMS model was applied for runoff simulation. In this regard, first the whole basin was divided to eight sub-basins enttled Marghak, Barez, Pataveh, Kata, Solegkan, Armand, DazakAbad and PolShalou. Then, based on the Digital Elevation Mode, the sub-basin's physiographic characteristics were extracted in Arc-View environment and using HEC-GeoHMS extension. The values of curve number, lag time and initial loss was calibrated and validated for each of the sub-basins using the observed sonw-free runoff events. Thereafter, the hypsometric maps with 100-meter elevation intervals was prepared by which the area, the required thermal data, and snowmelt pattern was determined for each of the elevation bands. Moreover, the snow water equivalent was calculated in the first day of the selected snowy events based on the daily calculations of snowmelt. After providing the required input data, the HEC-HMS mode was run for the snowy events and the critical temperature and degree-day coefficient was calibrated and validated for the study area.
The hypsometric maps with 100-meter elevation intervals revealed that 22, 70 and 8 percentage of the study area is, respectively, located in area with the elevations ranges of 693-2000, 2000-3000 and >3000 meter. Based on the observed data in the snow survey stations in the study area, the correlation coefficient between snow water equivalent and elevations was higher than 90 percentage. Therefore, this equation was applied for determining the snow water equivalent in the beginning of the selected snowy events. The daily calculations of snowmelt based on this equation showed that in Mach 17, 1998, which is the first day of the first selected snowy event, all the regions with elevations higher than 1943 meter is covered by snow and the non-snow regions covers less than 20 percentage of the study area. The results of the sensitivity analysis of the HEC-HMS model for calibrating the basin's physiographic parameters in the non-snow events revealed that the curve number, initial loss and lag time parameters had the most effect on the outlet discharge. The calibrated values for these parameters ranges among 76.6-91, 5.02-13.96 millimeter and 155-571.12 minutes, respectively. After calibrating and validating the basin's physiographic parameters, the snowmelt parameters were calibrated and validated base on the observed runoff during March 17-19, 1998, which is one of the largest snowy event in the study area. In this stage, the results of the sensitivity analysis demonstrated that critical temperature and degree-day coefficient had the most effect on three characteristics of flood hydrograph including the peak discharge, the average flood depth and the time of peak discharge occurrence. The increase in the critical temperature usually led to the peak runoff delaying. Except was for Armand sub-basin due to its high area in which, the increase in the critical temperature, the peak discharge of the flood hydrograph was increased mainly due to increasing the snow melt rate and the flow rate. The best values for critical temperature and degree-day coefficient was obtained, respectively, 2.5 oC and 3.2 millimeter per oC-day through the calibration process. Based on the calculated criteria indices in the validation process, applying these values led to the acceptable erors in estimating snowmelt runoff for the flood event occurred through March 25, 1998 to April 8, 1998.
Conclusions and Recommendations: Results of this research showed that among different physiographic characteristics, the curve number, initial loss and lag time parameters had the most effect on the average flood depth, the peak discharge and the time to the peak discharge for the non-snow events. Moreover, the snowmelt runoff is highly influenced by critical temperature and degree-day coefficient and therefore, the HEC-HMs results for simulating runoff in the mountain watersheds will be reliable if the suitable values are selected for these parameters. Based on the results of this study, 2.5oC and 3.2 millimeter per degree-days will be, respectively, the best values for critical temperature and degree-day coefficient in the study area and these values could be used in the further researches in Karoun Basin.