wxsim precipitation estimate

Hi Tom

The snow amounts were way off with the latest storm. I mean like three time the amount of snow that would be proper. Any thoughts please let me know. We picked up 15 inches of new snow and wxsim was forecasting 40 inches.

I must have something set improperly.

Thanks,

John

I am having sort of the same issue.

For instance, Environment Canada is forecasting for Friday a mix of sun and cloud and my forecast is saying 40 to 50 cm of snow. I have had other complaints about the precipitation being not only off but unreasonably high.

I wonder what could be going on that is having the precipitation so far off.

Thanks for your insight into this

Gary


EnvironmentCanada.JPG

Hi Gary,

Something is set up wrong I think. It’s happening now with wxsim forecasting 16 inches and NWS forecasting 3 to 5 inches. The heavy snow is well west of us.

Tom will see the post and maybe offer a resolution.

Thanks,

John

Hi Everyone,

I just checked my wxsim forecast and it now shows 5 inches of snow which is inline with NWS. It was forecasting 16 inches in the last forecast.

Thanks,

John

I’m seeing the same thing, I’ve been going in manually changing the .txt output on my website to reflect a more realistic figure.
3-5 inches overnight
1 inch tomorrow
2 inches tomorrow night
2-4 Tuesday

Most models are showing 1 maybe 2 inches overnight tonight for our region.

I decided to reboot my server this morning after the morning forecast and the results after the reboot were more inline with what we are expecting.
The WxSim forecast went from 10+ inches of snow over the next 48 hours to no snow. :slight_smile:

The snow amounts are dependent on the following things (in order of increasing importance):

  1. Settings within WXSIM regarding modification of intensity of precipitation, as per some settings under Preferences/Settings (fourth tab)

  2. Temperature and other factors affecting snow depth:liquid water depth ratio, compaction, melting, etc.

  3. The GFS and or NAM model data being imported.

To comment on each:

  1. These settings should be close to default values, and only different from that if you’ve done a careful study (over several months at least) to establish actual biases. This should be a non-factor for most people. In fact, the settings can be changed only in professional mode, because great care should be taken to establish that they would need changes. For most users, then, this is a non-issue.

  2. The algorithms here have been very carefully developed and are pretty widely applicable. Problems with this stage of the forecast process are usually minor, and there’s not much to do about them. It would account for slight differences here and there, but not huge errors or biases.

  3. THIS is the main issue. Snow accumulations depend critically on the exact path of storms. One a month ago here in Georgia gave me 4-5 inches, but 20 miles southeast of here there was practically non, and 30 miles northwest, they had 12 inches! Errors of 20-30 miles in storm tracks are common in all of the big models. In fact, that’s doing pretty well! WXSIM users almost always import GFS data, which is available worldwide and is one of the best around, though the European (ECMWF) model does somewhat better on average (it’s also expensive to get the data)! U.S. users also have NAM (North American Mesoscale) importable in the FOUS format. If both models are imported, they are generally blended by WXSIM, for a better result than either alone.

The NWS humans look at all these models and apply their judgement, hopefully adding some value to the raw products. In my experience, WXSIM’s output has been, on average, just about exactly as good as the NWS, but sometimes one’s better, sometimes the other. If you’re seeing big differences between WXSIM and NWS forecasts for this, it’s most likely (though I can’t say for sure) a matter of the ECMWF, UKMET, CMC, or some mesoscale models differing from GFS and NAM, and the humans giving those others more credence. Let’s see how it works out! :slight_smile:

Tom

Hi Tom,

I had made changes to the fourth tab. I reset them to default except the Upslope enhancement which is set to 40 with a direction of 320.

Thanks for the support.

John

Hi Everyone,

After several days of operation the precipitation amounts are back in line.

Thanks for the help Tom.

John