Sunday, April 7, 2013

It Ain't Over Yet, but Mother Nature Is Playing Dice

Over?  Did you say over?  Nothing is over until we decide it is!  Was it over when the Germans Bombed Pearl Harbor?  Hell no!
- John "Bluto" Blutarski (Animal House, 1978)

Yup, eventual Senator Blutarski was known for his intellectual insights, but in this case he was wrong.  Nothing is over until Mother Nature says it is over, and she still hasn't closed the door on this winter.

The forecast models are bringing a deep trough across the Intermountain West, ushering in colder weather for the first part of the coming work week.  The trough is forecast to be moving across northern Utah tomorrow afternoon.

Once again, the change in temperature is an easier forecast than snowfall amounts.  In fact, the difference in forecast precipitation for Alta between the various models is mind boggling.  Here's a summary based on the forecasts initialized at 0600 UTC (midnight) 7 April.  Accumulations are for the period from midnight tonight through noon Tuesday:

GFS*:                     .04" SWE; 0.4" Snow
12-km NAM:        0.93" SWE; 11.5" Snow
4-km NAM:          2.66" SWE; 35.7" Snow

Yeah, you got that right, the model spread ranges from less than an inch of snow to almost 3 feet [*The GFS forecast is technically for the nearest grid point to Salt Lake City, but there's not much difference between that and what the model is producing at Alta given the coarse 27-km grid spacing].  

There are two fundamental differences between these model forecasts, which I'll illustrate using the forecasts for 1800 UTC (1200 MDT) Monday 8 April.  Like real estate, the first difference is related to location, location, location.  All the models bring the trough across Utah and generate a surface cyclone over southern Utah with a precipitation band forming in an area of large-scale confluence to the north of the low center.  The GFS, however, has this band just a bit farther to the south than the 12- and 4-km NAM, so Alta just misses out on the action.

GFS (top), 12-km NAM (middle), and 4-km NAM (bottom) forecasts valid
1800 UTC (1200 MDT) 8 April 2013.  GFS precipitation for 6-h period,
NAM precipitation for 3-h period.  Note subtle difference in position of
precipitation band over Wasatch Front.
The second difference is related to resolution and terrain representation.  The high resolution 4-km NAM generates a much stronger precipitation band, and also produces localized precipitation maxima over high terrain throughout the forecast period, which results in more precipitation than generated by the 12-km NAM.  

All of that is fine and dandy, but for that forecast to verify, the 4-km NAM pretty much has to nail the position of the precipitation band, and it is clear from the forecasts above that a small error in that position could yield a dramatically different result.  Betting on the 4-km NAM to verify is perhaps like betting on snake eyes.  It's not impossible, but it is a low probability outcome.  Something closer to the 12-km NAM, which generates about a foot, is perhaps the highest probability event, much like rolling a seven, but it's no guarantee either. 

Just to further highlight the uncertainty in this forecast, below is the probability of more than 0.5" of snow water equivalent during the 24-hour period ending at noon Tuesday based on the Short Range Ensemble Forecast System (SREF).  Of the 21 members that comprise this forecast system, only about 30% generate 0.5" or more of SWE in the area around Alta.  The rest generate less.

So, the bottom line is that this forecast is pretty much a crap shoot, with a wide range of potential outcomes in the central Wasatch.


  1. NWS just issued a high wind watch for the Davis area, and it looks eerily like those from the downslope event that came through in 2011. Any thoughts on it? Are we to worry about a repeat in Centerville again?

  2. Some of the same predictability issues apply with regards to the intensity of the potential downslope windstorm. I'd keep an eye on the forecast and secure items in the yard just in case.