Will It Ever Snow Again?

December arrives with the dreaded "Rex Block" firmly in charge over the western United States. 

A Rex Block is a large scale pattern characterized by high pressure "over" low pressure as is evident in the GFS forecast valid for 0000 UTC 3 December (5 PM MST Wednesday) with an upper-level ridge centered over Oregon and a weak trough off the coast of Baja California.  

Rex blocks are named for meteorologist Daniel Rex who published an early article on blocking in 1950. 

Given the stability of this pattern, all members of the Utah Snow Ensemble are snow free at Alta-Collins through 0600 UTC 6 December.  After that, you can find a few members that try to bring a storm or two to Alta, but most keep us dry.   

It's easier to forecast the onset of a block than its demise, so perhaps the odds will ever be in our favor as we approach 10 December, but for now anticipate dry conditions, valley inversions, and faceting snow on shady aspects.  

Different Storm Pathways

What will certainly be the last storm of November and possibly the last storm for some time is now in the books.  

As a recap, I thought we would do a comparison of the total storm depth change at Deer Valley Ontario (9100 feet) and Alta-Collins (9662 feet) as they illustrate two different pathways for increasing the total snow depth about 15 inches.  I've take the liberty of stretching the Alta-Collins plot from MesoWest so that the y-axis scale increments are comparable in scale to those at Deer Valley Ontario.

Most of the snow at Deer Valley Ontario fell from about 0000–1400 MST Thursday (red shading).  This was during the pre-frontal storm stage discussed in the prior post (see Can You Help Explain the Overnight Snows).  

The Deer Valley Ski Patrol was caught mocking the Alta Ski Patrol on their snow-stake web cam.

Of course, Alta is Mother Nature's favorite son, and with and following the passage of the front late yesterday afternoon and last night (blue shading), she decided to give Little Cottonwood the goods.  

Two different pathways to 15ish inches.  In the end, everyone is happy. 

Can You Help Explain the Overnight Snows

Sometimes those with investment portfolios and trophy homes get the last laugh and that was the case last night with Deer Valley the big winner in the central Wasatch.  Below is a look at the Ontario Snow Stake Web Cam at 8:05 AM showing a solid 8".

Source: https://www.deervalley.com/explore-the-mountain/webcams

Meanwhile, on the other side of the Wasatch, at a ski area also frequented by people with investment portfolios and trophy homes but better known for deep powder, pickings were much slimmer.  

Source: alta.com

This is a pattern that does sometimes bless the Deer Valley side, although I confess I don't exactly know why, in part because of poor radar coverage, poor radar estimates, and limited observational data.  

The issues with radar coverage are apparent in the plot below, which shows the accumulated precipitation estimated from the National Weather Service Radar (KMTX) for the 6-hour period ending at 1400 UTC (7 AM MST).  The radar thinks the heaviest precipitation is in upper Big Cottonwood, in Brighton Basin, rather than to the east in the Deer Valley Ontario area. 

Source: https://mrms.nssl.noaa.gov/

In part, this reflects differential orographic blocking of the radar, which results in weaker returns (all else being equal) in the Deer Valley area.  It could also reflect overshooting by the beam if the growth of snow crystals in this situation is shallow.  Finally, the correlation between radar reflectivity and snowfall rate (including water equivalent rate) is much lower than it is for rain, so there are times when radar estimates are simply out to lunch.  In any event, the National Weather Service radar is not all that helpful for understanding what is happening in these events.  

There is also a complete lack of upper-air observations near Deer Valley, so we have to make due with the sounding from the Salt Lake City International Airport.  This morning's sounding shows southerly winds at low levels and westerly flow at 700 mb, roughly 10,000 feet. At issue is whether or not the flow in the Heber Valley in such a pattern is lifted and produces local, shallow snowfall enhancement on the Deer Valley ridgeline.  Some have speculated this is the case, but the hypothesis has not been carefully evaluated. 

Source: SPC

Perhaps a conflicting piece of evidence in this case is that the flow direction on Mount Baldy was not southerly overnight but southwesterly.  

If it was southerly or southeasterly, it would fit this hypothesis a bit better.  Of course, there's always the possibility that wind direction is affected by local conditions and the overall flow in that area is actually ascending out of the Heber Valley.  

I have another hypothesis, although it might not be as compelling as the flow direction one.  As shown in the sounding above, the crest level flow in this case was westerly and we had near saturated conditions through a deep layer, with strong flow in the upper troposphere.  On the other hand, the low level atmosphere in the Salt Lake Valley was dry with a relatively high cloud base.  A look at radar echoes for this period showed that they were not evident right over the immediate western face of the Wasatch, but somewhat downstream. 

My hypothesis is that in this event, we are seeing a situation where there is weak orograhic lift over the western Wasatch, but it is is deep, resulting in ice crystal generation aloft. Those ice crystals are carried downstream and fall out preferentially downstream of the Wasatch Crest over Deer Valley.  

There are examples of this happening over other ranges.  The best example I can think of is a case examined by Geerts et al. (2015) in the Range of Wyoming.  They flew through the storm in an aircraft with upward and downward pointing cloud radars (the dashed line below is the aircraft flight level).  These radars don't scan, but instead collect a continuous curtain of radar data above and below the flight track, allowing the detailed vertical structure o fthe storm to be observed.  The top panels are two different flight flight tracks during the storm.  In both cases, there are no low echoes upstream of the mountain and on the windward (left) side of the crest, reflectivities are highest aloft.  They calculated the streamlines of ice crystals and showed that those generated in this windward area aloft were carried downstream and fell out on the lee side of the mountain. 

Source: Geerts et al. (2015)

That paper is a favorit of mine because it shows how you sometimes need to think beyond where the mountains are forcing rising motion. You also need to think about transport and fallout.  This is particularly important when there is crystal generation aloft, possibly well above the crest.  

Anyway, that's my story and I'm sticking to it.  Perhaps you have other ideas and can help explain the overnight distribution of snowfall.

The Atmosphere Is a Complicated Place

Introductory meteorology textbooks depict a world of cold front, warm fronts, and occluded fronts. The cold front separates two airmasses, one the colder "polar" airmass, the other the warmer "tropical airmass."  The cold front is a long-lived feature in that intrudes into the warmer airmass, lifting it and producing a band of heavy precipitation.  Chance are you have seen conceptual models of this type.

If only the world were so simple.

The reality is that the atmosphere is a dynamic, complicated place in which troughs, fronts, and other atmospheric features are constantly evolving.  You can't put a line on a map in one area and expect it to move continuously, without evolution into another.  The developing storm for tonight and tomorrow is a prime example. There are a lot of moving parts, to take this discussion for what it's worth: A summary of a complicated atmosphere.

The GFS forecast valid at 1200 UTC 26 November (0500 MST Tuesday) is below.  I've identified some of the primary large-scale features of concern for the forecast.  The first is an atmospheric river (AR), characterized by an elongated filament of high integrated vapor transport (IVT) above 250 kg/m/s, that extends from the eastern Pacific across southern California and southern Utah.  The second is an upper-level trough at 500 mb (dashed line upper left) and 700 mb (dashed line upper right).  There is another also a developing trough downstream of the Sierra Nevada, evident at 700 mb, which I've identified with a solid line at lower left. 

Below is the total precipitation produced by the High Resolution Rapid Refresh (HRRR) through this time (0500 MST Tuesday) early tomorrow morning just to highlight the higher precipitation amounts tonight in the central mountains.  From Provo north, precipitation is heaviest around Provo Peak, Cascade Ridge, and Mount Timpanogos. By and large, this reflects the position of the strongest IVT accompanying the AR over southern Utah.  

Precipitation over the Salt Lake Valley though is fairly limited.  During this period, drier air fills the valley at low levels and causes precipation sublimation or evaporation.  Eventually we get some precipitation, but it will largely be a "cloud storm" or "virga storm" tonight. 

By 0300 UTC 27 Nov (8 PM MST Tuesday), the AR has "penetrated" across the Rockies and into the central US.  In this case, don't think of the leading edge of the AR as a material surface.  The IVT across the southern Great Plans was already close to AR level earlier and the strengthening IVT in that part of the world led to IVT values ≥ 250 kg/m/s rapidly extending all the way to eastern Missouri.

Meanwhile over the Great Basin the trough downstream of the Sierra has acquired frontal characteristics.  This occurred ahead of the approaching 700-mb trough, as depicted in the lower left-hand panel below.  This is an example of discrete frontal propagation in which a new front forms ahead of the approach 700 mb trough, as often happens over the Great Basin.

With this front moving through, precipitation over the Salt Lake Valley becomes more widespread, as indicated by the 6-h accumulated precipitation forecast valid 0300 UTC 27 Nov (8 PM MDT Tuesday).  

By 1500 UTC 27 November (8 AM MST Wednesday) we are well behind the front and 700mb trough, which have merged into one feature that extends from California into the southern Great Plains.  At this time, mountain precipitation would be associated with unstable, postfrontal, northwesterly flow (a bit being produced by the GFS is in the red circle). 

So, there's a lot going on.  True AR conditions remain to the south of the Wasatch during this period, although we will get some mountain snow on the fringes of it.  Then we have the frontal passage late Tuesday and Tuesday evening, and the post-frontal period Tuesday night into Wednesday.

Let's look at some totals from the models. The HRRR is generating 1.79" of water and 19.3" of snow.  The first part of the storm is relatively warm, with the wet-bulb zero level reaching about 7500 feet  early Tuesday morning (call it a 6500-7000 foot snow level give or take at that time) before it falls late Tuesday into Wednesday with the frontal passage.  Snow through mid day Tuesday looks to be relatively high density (snow-to-liquid ratios between 8 and 11 to 1 at Alta Collins), after which we transition slowly into lower density snow. 

The GFS (not shown) is one of the drier models, putting out only 0.84" of water and 11" of snow.  

Below is a plan-view plot from the Utah Snow Ensemble for the total accumulated snowfall through 1200 UTC 28 Nov (5 AM MDT Thursday, although most of this falls through Wed evening).  The mean of the 82-member ensemble is at upper right, minimum lower left, and maximum lower right.  The mean for Alta-Collins is about 16", with a minimum of 8" and a maximum of 27". 

The large contrast between the low-end models and the high end models at this stage is a bit ulcer inducing.  The HRRR at 19.3" is nearly double the GFS at 11".  I'm inclined to be cautious in a situation like this and lean toward a storm total of 12–24" at Alta Collins.  That means this will likely come in as the biggest storm of the season so far. It will help a lot, but probably not be truly transformative.  A best case scenario would be for the AR to shift a bit northward and for the post-frontal period late Tuesday night and Wednesday to be highly productive.  

Catskills and Poconos for the Win

Looking for powder?  Head east to the Catskills and the Poconos.  Some decent 24-hour totals out there, including 16.5" near Delhi. 

It's a bit of a strange pattern, resulting from the downstream development that we've talked about in prior posts and which also led to the bomb cyclone in the eastern Pacific.  In the northeast, that eventually led to a deep closed low over the mid Alantic states. 

Even Cleveland got in on the action for a bit. 

Meanwhile, looking to our west, the upper elevation site (7617 ft) at the Mount Shasta Ski Park had a pretty good run the past couple of days for snowfall, but appears to be either in or just below the melting layer now.  Observations from that site show total snow depth increasing from 15 to 63 inches in about 24 hours,but temperatures also steadily increasing through the period. 

Currently it is 35F, so I suspect they are seeing either rain, slush, or wet snow.  Quite a recipe for a deep, upside down snowpack and rain-on-snow avalanches.  Such a waste.

It's About to Hit the Fan

Satellite imagery for the north Pacific Basin this morning is simply incredible.  I could teach an entire class based on it.  One can see all of the features discussed in the previous post, How to Break the Jet Stream, the amplifying ridge over the Bering Sea, the amplifying trough over the east Pacific, the explosively deepening cyclone off the northwest coast, and the developing atmospheric river to the south of the low center.  

Source: College of DuPage

The National Weather Service Ocean Prediction Center surface analysis for 0600 UTC 19 November (11 PM MST Monday) showed the nascent cyclone upstream of the California coast.  At that time it was what we call an "open wave" cyclone with a warm front, cold front, and intervening warm sector, with a central pressure of just under 1004 mb.  

Source: https://ocean.weather.gov/unified_analysis.php

However, in the satellite imager above, you can see the development of a clear comma-cloud signature overnight, an indication of rapid deepening.  The GFS forecast called for the low center to deepen from what we'll call 1003 mb at the time above to an unbelievable 941 mb by 0000 UTC 20 November (5 PM MST Tuesday).  I'm sitting here right now wondering if I've done something wrong.  That is a drop of 62 mb in 18 hours.  The weather.utah.edu products don't include the central pressure of cyclones, so I'll use the GFS forecast from Tropical tidbits for 0000 UTC 20 November (5 PM MST Tuesday) to illustrate this incredible bomb cyclone. 

Source: TropicalTidbits.com

We're fortunate that storm is a bit offshore as it means the worst of the winds will be a maritime issue (but still a threat that will alter shipping routes).  However, the atmospheric river accompanying the system has its sights set on northern California and it appears it will be a long-lived AR event as we discussed in the prior post.  Below is the forecast for 1800 UTC 20 Nov (11 AM MST Wednesday).  The lower-left hand panel shows the magnitude of the integrated vapor transport (IVT) as colorfill and IVT vectors.  IVT is a measure of the amount of water vapor passing over a square meter of the earth's surface every second.  High values, indicative of an atmospheric river, extend from the eastern Pacific into northern California.  

Going out another 24 hours, there isn't much change in location, although the intensity is higher.  

Below is the Utah Snow Ensemble Forecast for Mt. Shasta Ski Park in the southern Cascades of northern California.  There is strong agreement in the ensembles for substantial precipitation at this location with the lowest amounts for this system (i.e., through 0000 UTC 24 November) of about 5" and the highest around 11" (see upper-left diagram).  I've used a red line to indicate an important transition point in the storm.  Prior to that time, the wet-bulb 0.5°C level is below the site elevation and the precipitation falls as snow.  In fact, there is a very tight clustering of the snowfall amounts through about 0900 UTC 21 November near about 25 inches.  Through that time, the wet-bulb 0.5C level in all the ensemble members rises (lower left panel) and the snow-to-liquid ratio falls (lower right) so this will be some high density, upside down snow. 

Around 0900 UTC 21 November, the wet-bulb 0.5°C level begins to rise above station elevation in somemembers and eventually it rises above station elevation inall members.  The net result is that all members call for rain in the latter part of this storm period.  Precipitation in the upper-left panel keeps increasing, but snowfall is flatlines and the snow-to-liquid ratio goes to zero.

The saddest five words in the English language are "the snow turned into rain." That looks to happen in this case.

Of course the more serious issues may involve flooding.  There is a flood watch issued for much of northwest California, including the northern and central Sacramento Valley, mountains of southwest Shasta County, and areas to the west. Let's hope the precip numbers for this event come in lower than advertised. 

How to Break the Jet Stream

A major transition in the structure of the jet stream will occur over the next few days, resulting in the development of an omega block over the Norh Pacific Basin and high-impact weather for northern California and the Northwest United States.

The plot below is a combined sea level pressure (black contours) and dynamic tropopause (jet-stream level) forecast for 0000 UTC 18 November (5 PM MST Sunday). I have identified the jet stream over the North Pacific Basin and North America with a black line.  The forecast shows a deep low pressure system over the Sea of Okhutsk (at the tip of the L1 arrow).  Downstream of it, there is strong southerly flow at the surface.  L1 and the strong southerly flow ahead of it don't look at that disruptive, but combined with warming due to condensation in the precipitation system accompanying them (not shown), they serve as the proverbial straw that break's the camel's back. 

By 0900 UTC 19 November (2 AM MST Tuesday), L1 has weakened, but the ridge downstream of it has amplified substantially.  Concurrently, the surface high pressure system, H1, has also amplified.  This is an example of the mutual amplification of an upper-level ridge and surface high pressure system.  However, that's not all that is going on.  The trough downstream of that ridge is also amplifying, and another low pressure system, L2, is starting to develop over the eastern Pacific.  

By 0000 UTC 20 November (5 PM MST Tuesday), the jet stream pattern over the Pacific Basin is highly amplified (call it wavy if you want) and L2 has turned into a monster as it and its accompanying upper-level trough mutually amplify.  L2 is what we call an explosively deepening cyclone, or bomb because its sea level pressure drops so fast.  Before numerical weather prediction and our ability to anticipate explosively deepening cyclones, such a storm would have been a shipwrecker, coming out of nowhere to  produce dangerous winds and seas that are a marine nightmare.  Today, we know it is coming, but it will still be a beast. 

Finally, by 1800 UTC 20 November (11 AM MST Wednesday), the omega pattern is fully developed with the high-latitude ridge and mid-latitude troughs forming a clear "omega" pattern covering the North Pacific Basin. 

This "breaking" of the jet stream results from the progressive amplification of a series of upper-level ridges and troughs (and accompanying surface high and low pressure systems) through a process called downstream development.  The end result is a high-amplitude flow pattern and in many cases high-impact weather.  

In this case, the high impact weather is in the form of L2, an explosively deepening cyclone, and it's accompanying atmospheric river, which looks to bring heavy rainfall to portions of northern California and the Pacific Northwest. Looking at the GFS forecast for 1800 UTC 20 November (11 AM MST Wednesday) shows L1 off the coast of Washington and British Columbia (upper right panel in the figure below), but the accompanying atmospheric river, identified by the high integrated vapor transport (IVT) values in the lower right plot below, is aimed straight at northern California.  This is a flow configuration that can produce heavy rainfall in the coastal mountains and southern Cascades.

The IVT associated with this system is high, but not exceptionally strong.  In the forecast above it tops out between 750 and 1000 kg/m/s, whereas extreme values can reach over 1500 kg/m/s.  However, due to the high-amplitude nature of the flow, this is a slow moving system, so the AR will be pointed at that area for a long time.  The Center for Western Weather and Water Extremes (C3WE) has created an AR scale to rate the intensity of atmospheric rivers and their impacts that considers both the strength and persistence of high IVT.  On a scale of 1-5, this one rates a 4 on the coast of northern California based on the control run of the Global Ensemble Forecast System (GEFS), another model used for weather prediction.

By now you are probably wondering what all this means for Utah.  It probably means a bit of a break after the weak system that moves through on Monday.  After that system, we look to be to the south of the AR, in an area that is relatively dry.  As a result, the Utah Snow Ensemble plume for Alta-Collins shows a bit of snow around 0000 UTC 19 November (5 PM MST Monday), and then 3 days of what looks to be dry weather.  After that, there is a wide range of possibilities.  

To summarize, the next few days will provide an example of how Mother Nature can break the jet stream and produce a high-amplitude flow pattern, explosive cyclogenesis, and high impact weather.  I'd say buckle up, but Utah is not in the crosshairs, at least for the next few days.

Book Review: The Darkest White

 

Perhaps no modern figure has had a greater impact on a snow sport than Craig Kelly.  He was the first truly big legend in snowboarding, ushering the sport from its nascent fringe to mainstream, although he might object to the word "mainstream." Many people consider him the greatest snowboarder of all time. 

Kelly's life is the subject of Eric Blehm's book, The Darkest White, which was released earlier this year. It covers his transition from BMX to snowboarding and his ascent to world champion at a time when snowboard racing and freestyle competitions were not well structured.  Kelly excelled in all competitions.  It was the early days, but imagine if Mikaela Shiffrin, in addition to racing everything from slalom to downhill, was also throwing in some half pipe for the hell of it.  Kelly walked away from all of that to pursue freeriding and eventually train to become a certified mountain guide at a time when there was considerable prejudice against snowboarders. 

The book covers those aspects of Kelly's life, although I found that part of it a bit chronological and maybe too focused on competitions.  I suspect snowboarders will appreciate it more. 

The real page-turning part of the book is the description and diagnosis of the avalanche that sadly took his life and the lives of six other people in the Selkirk Mountains.  While working on his guideship skills, Kelly joined a group at Sellkirk Mountain Experience to work with the famed mountain guide Reudi Beglinger.  While ascending La Traviata in two groups, an avalanche was triggered catching 13 tourers.  Several were recovered successfully, including Ken Wylie, an apprentice guide, who was rescued after 35 minutes of full burial.  However, seven died including Kelly who was buried 9 feet deep and recovered after abount an hour. 

I was quite aware of the avalanche as well as another that season that killed seven others because I spent a week on a ski touring trip nearby with Golden Alpine Holidays.  It was on our collective minds during our tours and we spent a good deal of time skiing in low-angle terrain and not disturbing the monster in the basement.

It's difficult for me to say if the book (or all of the discussion you can find online) paint a fair and accurate assessment of what happened and why.  Hindsight is of course 20/20, but I spent a lot of time reflecting on the snowpack assessment and human decision-making in the account that Blehm presents.  There is a lot to be learned and I often reflected upon my own experiences and lapses of judgement, of which there are many.  I suspect Craig would appreciate everyone learning from what happened, especially if it results in safer backcountry experiences. 

To conclude, I share the photo below that I took of a memorial to Craig at Seki Onsen ski area in Japan in 2017.  I don't know what it says, but I was impressed to find it at a small (but very snowy) resort in the Myoko Kogen.

Déjà Vu All Over Again

I had a mild bike ride to the office this morning for November 11 and the forecast high for this afternoon is 65 for KSLC. 

Tomorrow will be different. 

However, it is a bit of a case of déjà vu all over again as it is a system that reminds me of the others we have had so far this fall with the strongest part of the trough and the so called "dynamics" moving to our south.  We get a frontal passage and some precipitation, but looking at the models, the system is just not put together great to give us a really big dump. 

Below is the GFS forecast valid 1200 UTC 12 November (6 AM MST Tuesday).  The strongest part of the 500-mb trough is basically over Northern Arizona and southern Utah and passing to our south.  The surface front is pretty much over Salt Lake City, but the 700-mb temperature contrast with it is somewhat diffuse and the precipitation with the front scattered and not all that organized.  

The GFS time height section shows very dry air ahead of the front and then a period of about 6-hours with deep moisture as the front moves through.  We do get into northwesterly post-frontal flow Tuesday night, but it is fairly dry at low levels (below 800-mb).  This is not the kind of forecast that causes my heart to flutter. 

The latest models are calling for a small storm at Alta.  Through late Tuesday the HRRR generates 0.25" of water and 3.5" of snow and the GFS  0.37" of water and 5" of snow.  The experimental RRFS has a remarkable amount of spread with one member producing essentially a trace and another up at 10".  

I'm not sure what to make of that, but I view that 10" as something close to the upper limit of what I'd expect under the best of circumstances, with a significant boost from lake-effect interactions with the terrain behind the trough.  

Finally, the Utah Snow Ensemble is generally in the under 12" range with means from the two ensemble systems at about 4" and 6".  Yes, I know there's more snow after this system in this forecast but I'm not going to address that here.  Whatever you do, don't be biased by the highest members! (See Anchoring Bias and Ensembles for why). 

It is what it is, another modest system that will add a bit more to our November snowpack.  I'm thinking 5-10" for Alta Collins. Although it is clear that the ensembles are saying the range of possible outcomes for this event is pretty big, there isn't much here to get me thinking about more than a foot.  That said, none of these models are particularly good at dealing with the post-frontal northwesterly orographic snowfall enhancement in the Cottonwoods.  Then again, they also aren't advertising a great environment for that.  

Colorado and (gasp!) Texas Running Away With It

 Looking for snow?  Forget the Collins glacier.  Go to Alta or maybe Texas.  

Over the last four days, Colorado, New Mexico, far western Oklahoma, and the upper northwest corner of Teas were absolutely pounded.  This includes the high planes.  Texline, TX recorded 24" of snow.  Boise City, OK, not Idaho, recorded 26".  

The big winner was a site 12.9 miles ENE of Fort Garland with 53.3".  The NWS does not provide specific locations for privacy reasons, but that looks to be a site in the low pass through the Sangre de Cristo between the San Luis Valley and the I-25 Corridor.  Some big numbers as well in the high plains of Colorado and the mountains of New Mexico.  A look at Angel Fire (40") this morning.

https://www.angelfireresort.com/weather/

I've always wanted to ski there just for the name.  One of the best in the business. 

The fattest snowpacks in the Utahrado region are now in the San Juans and Sangre De Cristo Mountains. The big winner (blueish dot) is Beartown at 11,600 feet which is sitting at 7.7 inches.  This is the equivalent of their median snowpack on December 20th, so they are running about 6 weeks ahead of median. 

Source: NRCS

Hayden Pass in the Sangre De Cristos now sits at 5.8 inches, the equivalent of their median on December 28 and way above anything on record, although observations at this site start only in 2008.  

There are, however, other sites at record levels for this date in the Sangre de Cristos, Pikes Peak, and Buffalo Peaks.  These are historically dry areas, so an extreme event like this is really exceptional. 

It's a weak La Nina year and this has guided seasonal forecasts.  Here's one for Nov-Jan from the Climate Prediction Center that now looks on track to bust for at least parts of southern Colorado and northern New Mexico.  

This one event has produced about 2/3 of the average Nov-Jan precipitation showing how an extreme weather event can strongly contribute to seasonal precipitation and snowfall.  This is a characteristic of the cool-season snow climate of some regions of the western United States that is often overlooked when seasonal forecasts are being issued.  It introduces an element of randomness to the year-to-year variability in western precipitation that can limit the reliability of seasonal forecasts based on long-term means and similar analyses.  There is a good paper by Lute and Abatzoglou (2014) showing that 20-38% of the annual snowfall water equivalent and about 2/3 of the year-to-year variability in that metric can be attributed to the top ten decile (10% largest) snowfall events in portions of the western United States.  Basically, a handful of big events, sometimes just one or two, make or break the season.

Congratulations to the early season snowfall winners.  

The RRFS Snow Ensemble

We are excited to share that the RRFS Snow Ensemble is now available on https://weather.utah.edu and has replaced the old SREF product.

The RRFS Ensemble is a 6-member ensemble that is under development for future operational use by the National Weather Service.  Based on the FV3 dynamical core (i.e., the software that solves the atmospheric equations of motion), it is run at 3-km grid spacing and provides forecasts out to 60 hours.  The RRFS Ensemble is projected to go operational in 2024, although there have been a number of issues and challenges identified during testing that may affect that, in particular related to the forecasting of convective storms in the midwest.  The RRFS Ensemble has not been carefully evaluated over the western US, so one of the reasons we are producing this product is to evaluate its fidelity for orographic precipitation.  

We are also interested in testing our techniques for snow prediction.  Thus, what we call the RRFS-Snow Ensemble is basically an ensemble in which we plug in new techniques to predict snow-to-liquid ratio (SLR) and snow amount.  A summary of the data and methods and graphics is provided below. 

Precipitation Downscaling

Although we downscale the Utah Snow Ensemble from the lower resolution global ensemble grids, the RRFS is providing forecasts at 3-km grid spacing. Thus, we are currently doing no precipitation downscaling and just using the raw model grids.  There might be some advantage to downscaling the RRFS eventually, but for now, we're not doing it.  

Snow-to-Liquid Ratio (SLR)

Snow-to-liquid ratio (SLR) is based on a new random forest algorithm developed using SLR observations from more than 900 Community Collaborative Rain, Hail & Snow Network (CoCoRaHS) observing sites.  Thank you to all the volunteer observers and the CoCoRaHS team!  In particular we are using a subset of CoCoRaHS observing sites at which the observers are taking manual cores of the snowfall data, which we hope will reduce issues related to precipitation undercatch, which is a problem with the water equivalent measured by many gauges.  The random forest is trained using data from across the contiguous United States and in testing has performed better than existing operational techniques in both the western and eastern United States.  

Snow Level

Identifying snow level in the west or precipitation type in the east is a bit of a thorny issue.  The so-called "wet-bulb" technique that we use in the western United States works fairly well when the temperature decrease with height is close to what meteorologists call a wet-adiabatic lapse rate.  It doesn't work well if the atmosphere is stable and/or has a warm nose above freezing aloft.  

As a result, we decided to use a more physics-based approach to identify if snow is occurring.  At each model grid point, we calculate the melting energy in the model soundings.  This is the amount of energy available to melt snow in the sounding.  The technique is based on Bourgouin (2000), although we use wet-bulb temperature rather than dry-bulb temperature to calculate melting energy (special thanks to Kevin Birk of the National Weather Service for providing some of the initial code for this work).  Currently we are applying our random forest SLR without adjustment if the melting energy is ≤ 2 J/kg and assuming the precipitation is all rain if the melting energy is ≥ 9 J/kg.  If the melting energy is between those values, we reduce the SLR between the random forest value and 0 based on linear interpolation between the two thresholds.  

Those thresholds are based on published values, but admittedly, the data is not comprehensive.  It may require some modification over time.  However, they do give results similar to the wet-bulb method when the lapse rate is near moist adiabatic and can deal with more complicated temperature profiles.

Note that we are not attempting here to diagnose freezing rain or sleet.  The melting energy approach we are using will basically give us a SLR of 0 in those instances.  So, our plots only show forecasts of accumulated snow.  

We are working with another group to possibly incorporate a machine learning technique for precipitation type in the future, but it may be a while before we get to that.  

Four-Panel Plots

We provide loops of four-panel plots of the following variables for several regions, including over the central and eastern US: 

• Total precipitation (water equivalent) since the beginning of the forecast period
• Total snow since the beginning of the forecast period
• 24-h precipitation (water equivalent)
• 24-h snow
• 6-h precipitation (water equivalent)
• 6-h snow
• 1-h precipitation (water equivalent)
• 1-h snow
• Wet-bulb 0.5°C height above ground level (based on the lowest wet-bulb 0.5°C level)
• SLR

SLR and snowfall are calculated in 1-h intervals, with the resulting 1-h accumulations summed to provide accumulations over longer periods.  Thus, the 24- and total snowfall should not be confused with the change in snow depth on the ground over long time periods, which would be affected by settlement.  

Each four panel plot includes the the RRFS control forecast at upper left, the ensemble mean at upper right, the ensemble minimum at lower left, and the ensemble maximum at lower right.  Below is an example of the total snowfall through 60-h over the Wasatch Front.

For some of the caveats of interpreting these plots, see my blog post on The Utah Snow Ensemble. 

Plume Plots

We are also providing forecast plumes and violin plots for many locations, including several in the northeast US, to provide more information about SLR uncertainty.  These are identical to those for the Utah Snow Ensemble, so refer to The Utah Snow Ensemble blog post for information on interpreting these plots.  The RRFS ensemble, however, only has 6 members, so there are not a lot of forecasts and the violins are going to be based in part on interpolation fro sparse data.  Some groups use time-lagging (i.e, using older forecasts) to increase the ensemble members (but also decreasing the forecast period), but we're not bothering with that for now. Perhaps at some point we will change the lower left panel from wet-bulb 0.5 level to melting energy, but for now we're keeping it consistent with the Utah Snow Ensemble.  The wet-bulb 0.5°C level is based on the lowest level in the sounding, so it will not tell you where the top of a warm nose is and may not be a useful variable in situations where there is a warm nose aloft.  

Caveats and Disclaimers

This is an experimental product.  In fact, it is an experimental SLR product post-processing experimental ensemble modeling system!  Feedback is helpful to us as we are trying to find ways to better forecast snow and its characteristics and squeeze everything we can out of the operational model suite.  Tell us what works and what doesn't.  

The RRFS ensemble is based on data and products from the National Centers for Environmental Prediction (NCEP), University of Utah, and other groups.  These groups do not accept any liability whatsoever for any error or omission in the data and their availability, or for any loss or damage arising from their use. 

This blog post may be updated as needed.

The Oquirrhs Got the Snow

If you are looking for snow, perhaps you should look to the west to the Oquirrh Range instead to the Cottonwoods. The latest snow water equivalents from SNOTEL sites show the highest amounts at the three SNOTEl stations in the Oquirrhs.  These numbers do not include the lake-effect snow that fell out there last night and this morning.   

Source: NRCS

Shall we have a closer look?  The fattest snowpack is at Rocky Basin Settlement (8704 ft) where the snow water equivalent sits at 2.7". 

That's well above median for that site, although median so early in the season isn't a very robust statistic.  

For comparison, the Snowbird SNOTEL (9177 ft) is at only 1.2 inches.

Why are the Oquirrhs so blessed?  A big chunk of the snowpack at Rocky Basin Settlement fell on October 29.  They got some snow in the southwest flow like the Cottonwoods, but they did very well in the north-northwesterly flow following trough passage.  For example, the radar image below for 1821 UTC 29 October shows strong echoes just to the west of the Oquirrh crest.  These echoes persisted for a good chunk of the afternoon. 

It's hard to say exactly what the role of the lake was during that period, but the echoes are suggestive that both lake-effect and orographic (mountain lifting) processes contributed to snow enhancement over the Oquirrhs.

Then, in the evening, some lake-effect snow developed. 

For a while, as midnight approached, a very localized band developed.  

So, the Oquirrhs have done much better in the post-trough north-northwesterly flow than the Cottonwoods.  

Is this unusual?  It depends on what you use as a baseline.  It is unusual for the Oquirrhs to have more snow than the Cottonwoods.  At Rocky Basin Settlement, peak median snow water equivalent is 23.9 inches compared to 42.9 inches at the Snowbird Snotel. 

On the other hand, lake-effect periods produce about the same amount of snow in the Oquirrhs as in the Cottonwoods.  The figure below shows the water-equivalent produced by lake-effect periods during the 1998-2009 water years (adapted from Yeager et al. 2013).  During those water years, lake-effect periods produced an average of 2.12 and 2.37 inches of precipitation water equivalent at the Rocky Basin Settlement and Dry Fork SNOTEL stations in the Oquirrhs, respectively, compared to 2.06" and 2.38" at the Mill D North and Snowbird SNOTEL stations in the Wastach, respectively (see left figure below).  

However, since the Oquirrhs are otherwise drier, lake-effect periods constitute a greater fraction of the cool-season precipitation there (right figure above), including 6.3% and 8.4% of the cool-season precipitation at the Rocky Basin Settlement and Dry Fork Snotels.  For comparison, lake-effect periods produce 5.9% and 5.1% of the cool-season precipitation at Mill D North and Snowbird.  

So, such lake-effect snow is not unusual in the Oquirrhs.  They are currently ahead of the Cottonwoods because they were favored given the north-northwesterly flow in the wake of the trough on 29 November.  A similar situation occurred last night and this morning.  

As the saying goes, it's better to be lucky than good.