We're trying to use snowmelt in the runoff module of SWMM. We never used it, so we have no idea of the range of all the coefficients that we need to use. We want to use realistic values, to have realistic output. The simulation we have to make is for a city near Ottawa, for an event having a return period of 5 years.
Could anyone help ?
The SWMM documentation (available from the OSU website
includes a file called SWDAT4.DAT that includes sample of continuous snowmelt parameters. The number of parameters is quite overwhelming, but we've found it does produce reasonable estimates of snowfall/melt. In order to simulate a '5-y' condition, you should run the full winter season in order for the snowpack to accumulate. You should also keep in mind that there is no consideration of frozen ground and resulting reduced infiltration rates. To simulate true winter conditions, you'd probably need to reduce the soil infiltration rates from values you calibrated for another season. Alternatively, you could consider using the groundwater module to simulate the reduced infiltration capacity of the soils.
I am trying to categorize SWMM/RUNOFF input parameters for snowmelt into: (*) those which are absolutely necessary when the process is simulated in
impervious/perv areas and those which are used when continuous simulation is performed or in particular situations. I listed the parameters in the the C1-3 lines and I1-2 lines bellow. I personally have never used this rountine before and I was wondering if someone who has could quickly double-check the list for me.
The snowmelt routine has some serious flaws as has been pointed out by many over the years. Not least among the flaws is the ridiculously large number of parameters. I'd suggest you organize your list a bit better, as I have trouble finding the parameters on it. Furthermore, you might inquire as to whether anyone ever uses some of the features. For instance, I run continuous snowmelt in some of my models, but have never run a single event simulation and can't imagine doing it. Thus, all the initial condition variables are useless as far as I'm concerned (i.e. WSNOW 1-3 and FW 1-3). Other features I've never used are reading snow from a data file (I let SWMM simulate it based on temperature), and plowing from one catchment to another (or for that matter plowing within a catchment). For me, its main use is to improve wintertime CSO estimation by holding the precipitation on the ground longer and draining it slowly into sewers.
Here is another list of the snowmelt parameters in SWMM. If you are doing a single storm the intial conditions are most important. If you are doing continuous simulation the melt coefficients and the redistribution coefficients are most important. I find in Florida modeling that the parameter "Dividing temperature between snow and rain" is the most important. In Pittsbugh the melt coefficients are most important. Snowmelt does work on a seasonal basis. SWMM actually needs more parameters or other data such as solar radiation to better model the actual melt events. Maybe SWMM should incorporate some of the temperature simulation capability that Bill James and his students have added in the 90's and 00's. That way we can model warm pavements and other important factors.
The initial condition variables are used in continous simulation, BTW.
Global Parameters ================= Description Dimension Value ------------ --------- ----- Average watershed elevation ft, msl [m, msl]. 900 Ratio of free water holding capacity to snow depth (in. or mm w.e.water equivalent) 0.03 Ratio of free water holding capacity to snow depth on snow covered pervious areas (in. or mm w.e.= water equivalent) 0.03 Ratio of free water holding capacity to snow depth on normally bare impervious area (in. or mm w.e.water equivalent) 0.03 Dividing temperature between snow and rain Degrees F or C 34 Snow gage catch correction factor Dimensionless 1 Weight used to compute antecedent temperature index Dimensionless 0.1 Ratio of negative melt coefficient to melt coefficient Dimensionless 0.6 Average latitude of watershed, degrees north Degrees 40.4 Longitude correction, standard time minus mean solar time Minutes (of time) 20 Watershed Specific ================== Description Value Fraction of impervious area with 100 percent snow cover 0.4 Fraction of pervious area with 100 percent snow cover 0 Initial snow depth of impervious area that is normally snow covered 0 Initial snow depth on pervious area 0 Initial free water on snow covered impervious area 0 Initial free water on snow covered pervious area 0 Maximum melt coefficient for impervious area 0.006 Maximum melt coefficient for pervious area 0.006 Snow melt base temperature for snow covered impervious area 31 Snow melt base temperature for snow covered pervious area 32 Initial snow depth on impervious area that is normally bare 0 Initial free water on impervious area that is normally bare 0 Maximum melt coefficient occurring on June 21 for snow on normally bare impervious area 0.007 Snow melt base temperature for normally bare impervious area 31 Minimum melt coefficient occurring on December 21 for snow covered impervious area 0.002 Minimum melt coefficient occurring on December 21 for snow covered pervious area 0.002 Minimum melt coefficient occurring on December 21 for snow on normally bare impervious area 0.003 Snow depth above which there is 100 percent cover on snow covered impervious areas 0.3 Snow depth above which there is 100 percent cover on snow covered pervious areas 6 Snow depth above which there is 100 percent cover on snow covered impervious areas 6 Snow depth above which there is 100 percent cover on snow covered pervious areas Redistribution (plowing) depth on normally bare impervious area 0.7 Fraction transferred to snow covered impervious area 0 Fraction transferred to snow covered pervious area 0 Fraction transferred to snow covered pervious area in the last watershed 0.1 Fraction transferred out of watershed 0 Fraction converted to immediate melt on normally bare impervious area 0