Important Differences Between the EC and the GFS

In yesterday's post, I asked the question, "should we buy in to the latests forecasts?" Motivating that question was the very optimistic forecasts for snow being produced by the GFS and the NAEFS, as well as the less optimistic forecasts being produced by the ECMWF Integrated Forecast System (a.k.a., the ECMWF, EC, or Euro model and ensemble).

Given the restricted nature of data provided by the ECMWF, I've struggled with how to show these differences, but someone pointed out to me that San Jose State University has a comparison between the EC and the GFS on their web site.  I love it when my job gets easier.

So, here's the basic issue.  Yesterday mornings 1200 UTC GFS (green 500-mb height contours below) produced a much weaker and progressive trough than the EC (red contours) for later this week and weekend.  These result in dramatically different forecasts for precipitation in the central Wasatch, with the former being wetter and more favorable for orographic precipitation enhancement following the anticipated late Thursday/Thursday night frontal passage, and the latter much drier.

Ensembles smooth these differences out a bit, but the GEFS and EC Ensembles generally show similar differences.

Last nights 0Z runs show that the GFS has shifted a bit toward the EC solution, but the trough remains weaker.  Note that the forecast below is for 12-hours earlier than the one above since forecasts are available only every 24 hours.

San Jose State hasn't processed up the 12Z EC yet, but I've looked at it elsewhere and it shows a pretty similar solution to the previous forecasts with precipitation accompanying the front and then not much thereafter.

Bottomline: It looks like we'll get something with the frontal passage, but a major post-frontal snow event, as has been advertised by the GFS, remains somewhat suspect for Friday and Saturday.  I lean toward the EC solution because it has long been the better performing model and the trough evolution it is predicting is consistent with the parade of troughs that have moved through western North America over the past 6 weeks or so.

It will be interesting to see how this all plays out, but I'm betting on a final score that is EC 1 GFS 0, although I would be happy to lose (i.e., big snows Friday and Saturday) or take a tie (i.e., we get some but not a lot of post-frontal snow on Friday and Saturday).

Should We Buy In to the Latest Forecasts?

Snowfall has been scarce in the central Wasatch so far this season and since mid-November Alta has reported only 13" of snow.   Storms that have looked good in the GFS at Day 7 have consistently fallen apart.  The latest GFS forecast looks enticing, but at issue is whether or not to get excited.

In the short-term, we have two brush-by systems that might bring a bit of snow to the mountains tonight and late Wednesday.  Emphasis is on a bit, but even an inch or two would help.  Then, the GFS is advertising a fairly significant pattern shift later in the week with a deep, digging trough pushing into the western U.S.  As currently advertised by the GFS, there's a healthy frontal passage late Thursday and then a series of weak troughs that give us some additional shots of precipitation in the post-frontal northwesterly flow.

Although past events have sometimes featured NAEFS ensemble members that have been fairly dry, all the members are giving us something decent (i.e., 1.5 inches or more of water at Alta Collins) during the 3.5 day period ending 0000 UTC 14 December, although there are differences in timing and intensity of the various snow periods.  

That's all fine and dandy, but I took a looksee at the 0000 UTC Euro and it gives us a decent frontal passage, but digs the trough further south and is far less excited about the post-frontal action.

Given that I've seen this movie a few times already this season, I'm hopeful we'll see something later this week, but am not yet buying into the optimistic GFS and NAEFS forecasts.  They are a possibility, but not a certainty, and I'd rather be a realistic pessimist and be pleasantly surprised than a realistic optimist and be disappointed.  Low expectations always yield better powder days.

Avalanche Talk Wednesday Afternoon

There's some hope for snow in the latter half of the upcoming work week, but not much will happen between now and then other than the continued rotting of the snowpack.  

Rather than being out in the field watching facets grow on Wednesday afternoon, those of you with an interest in snow science might want to attend a seminar that will be given at the U by Jordy Hendrikx.  Jordy is director of the Snow and Avalanche Laboratory at Montana State University and Chair of the Research Committee for the American Avalanche Association.  

Photo Credit: Karl Birkeland

Here are the details:

Snow Science: Taking a Multi-Faceted Approach to Understand Avalanches

Jordy Hendrikx, Montana State University

Wednesday, December 9, 2015

INSCC 110

University of Utah

Refreshments and Meet the Speaker at 3:00pm and Seminar at 3:15pm

Map: http://bit.ly/1ljehqL

Still Some Gunk Around

Last night's trough passage helped mix out some but not all of the pollution.  As can be seen in the photo above, there's still some gunk lingering around the Salt Lake Valley.

And check out some of the details of the partial mixout in the graphs below.  At our mountain meteorology lab on upper campus, PM2.5 levels dropped abruptly shortly after 1800 MST last night, with only a very small recovery.  In contrast, at Neil Armstrong Academy in West Valley City, which is about 750 feet lower in elevation, the mixout was about 2 hours later and only temporary as PM2.5 levels rebounded to around 15 ug/m3.  

Source: MesoWest

This is a great example of a partial mixout and it is a fairly common occurrence with weaker trough passages, which often fail to completely scour out the air pollution.

As things stand this morning we have a shallow inversion over the Salt Lake Valley.

Source: Storm Prediction Center

The strength and depth of that inversion is such that I think we will probably burn it off with surface heating and help dilute things a bit further over the valley.

The New GEFS and Mountain Precipitation over the West

The National Centers for Environmental Prediction upgraded the Global Ensemble Forecast System on Tuesday.  The changes are fairly extensive and include increases in horizontal and vertical resolution, improvements in initialization methods, and upgrades to the model dynamics and physics.  For true weather wonks, details are available here.  The GEFS is a component of the North American Ensemble Forecast System (NAEFS), which is available on weather.utah.edu.

NCEP has been running this new version of the GEFS in what is known as reforecast mode, which involved running it in hindsight mode for the past two years.  This has allowed us to evaluate its performance for the past two cool-seasons (October to March).

The GEFS is an ensemble modeling system, meaning it produces multiple forecasts (known as members), but I find it helpful to begin an evaluation effort focusing on a single member (in this case the Control) to understand the capabilities and limitations of the model before examining how well it predicts probabilities.  Below is one measure of model performance, known as an equitable threat score (the higher the better), for 12-36 hour (i.e., day 1) precipitation forecasts of the largest 25% of precipitation events at mountain SNOTEL stations in the western U.S.  Although there is considerable spatial variability, the highest equitable threat scores are found in general in the Cascades and Sierras, with lower values at sites in the interior.  This implies a general decrease in model accuracy as one moves into the interior.  If we use some simple statistical techniques to try and account for unresolved terrain effects (the same used for our NAEFS ensemble products on weather.utah.edu), there is a significant improvement in model performance (right image) at nearly all sites, although one can still see the general decline as one moves inland.  If you live in the mountains of Colorado and think your forecasts suck, you're right!

Equitable threat scores for Day 1, 24-h precipitation forecasts at SNOTEL sites 

For the west as a whole, there is a general decrease in model performance with increasing lead time.  This isn't too surprising and it results from a number of factors, including a decrease in the number of times the GEFS control correctly forecasts an event combined with an increase in the number of false alarms.  In fact, at day 7, there are about 3.5 false alarms for every hit (not shown).

Equitable threat scores for daily precipitation forecasts with increasing lead time (forecast day) at SNOTEL sites.  

I suspect that the ratio of false alarms to hits is nearly as bad for the GFS and the Euro, simply because at day 7, there's a lot of uncertainty in the forecasts.  If your modus operandi is to look at  one of these models and go with it, you are going to get burned a lot.  Long range forecasts are about probabilities rather than certainties and it's better to understand the odds than hope for a single forecast to come through.

Speaking of odds, we're starting to look at how well the full GEFS ensemble does generating probabilities.  We should know soon if it is a card shark or an irrational gambler.

Special thanks to Wyndam Lewis, Trevor Alcott, and Jon Rutz for their contributions to this post.

Friends of the Marriott Library Lecture

Instead of skiing Sunday afternoon, catch me at the Marriott Library and learn about the Greatest Snow on Earth.

Bloody Cold in Some Spots

With generally calm, clear, dry weather in place over northern Utah, overnight minimum temperatures are getting way down there in some spots.

Cache Valley: Lewiston -1ºF, Evans Farm 0ºF, Logan Airport 3ºF
Bear River Valley: Randolph -13ºF, Pixley -14ºF
Rush Valley: Faust 0ºF
Uinta Basin: Duschesne 1ºF, Maeser 3ºF

Meanwhile in the friendly confines of the Salt Lake Valley, we hit minima of 22ºF at the University of Utah and 19ºF at the airport.

The Cache Valley, Bear River Valley, Rush Valley, and Uinta Basin are well known for low minimum temperatures.  These are basins or valleys that are largely enclosed, resulting in very light nighttime winds during periods of weak flow.  The lack of flow and associated turbulence reduces the mixing of air near the surface at night, allowing temperatures near the basin or valley floors to drop very low.  Note that this has nothing to do with "cold air flowing to lower elevations" as is commonly claimed.  Studies show repeatedly that shortly after sunset the flow in these very cold low-elevation areas typically becomes quite weak, which limits mixing and allows the air near the surface to cool dramatically.

In contrast, the Salt Lake Valley is more open, enabling a steady nocturnal drainage flow that helps keep the surface layer a bit more mixed and warmer.  The urban heat island helps as well.

There is one other factor presently at play in the Cache Valley, Bear River Valley, and the Uinta Basin: Snow cover.  Yesterday's modis image shows extensive snow cover in these areas, in contrast to the Salt Lake Valley, which has only spotty coverage.

Snow affects the surface energy balance in several ways.  During day it reflects more sunlight back to space than bare ground (in most instances).  At night, however, it reduces the flux of heat from the ground to the atmosphere, which allows the air near the surface to cool more than over bare ground (in most instances).

Ultimately, the presence of snow leads to stronger inversions within the valleys and basins of Utah. We're fortunate to be snow free in the Salt Lake Valley at the moment.