Category Archives: Forecasting

Core Dump on Our Pending Storm

The Doppler on Wheels (DOW) has been in town now for two weeks, and we've yet to have a significant storm during that period.  At Alta, the best we've done is 4 inches on November 4th, which isn't much of a storm.  We've kept ourselves busy with various educational and outreach activities, but are in desperate need of a storm.

Fortunately, it looks like Mother Nature will give us something tomorrow afternoon through Friday.  It's been an interesting storm to follow in the forecast models for a number of reasons.

1. It's taking forever to get here.

The GFS forecast initialized at 0000 UTC 12 November (5 PM MST Saturday) showed the upper-level trough making landfall onto the Pacific Northwest coast at 1800 UTC 16 November (11 AM MST Thursday), with precipitation across the Wasatch Mountains, Uintas, and even western Colorado.  Under this scenario, we'd want to be out storm chasing early tomorrow (Thursday).

During the past three days, however, the GFS has really slowed the progression of the trough.  The forecast from 0000 UTC 15 November (5 PM MST Tuesday) has the trough much farther west and well off the coast at 1800 UTC 16 November (11 AM MST Thursday).  Precipitation is just sneaking into the northern Wasatch and Bear River Range area, and there's no precipitation over Colorado.  Instead, we'll be able to sleep in tomorrow!

2. The Sierra Nevada really take a bite out of storms

Through flow blocking and water vapor depletion in mountain-induced rain and snowfall, the Sierra Nevada have a dramatic impact for the worse on moisture transport into the Great Basin.  This can be seen in the GFS forecast pannels immediately above.  Note in particular the how the column-integrated relative humidity in the lower left panel decreases abruptly across the southern "High" Sierra, with moisture only able to sneak in across the lower northern Sierra north of Lake Tahoe.  This effect is also apparent in theNAM forecast for the same time and, in this case, it is a contributor to the delay of precipitation spreading into northern Utah.  Without the High Sierra, moisture would penetrate more easily into the Great Basin and the Wasatch would light up even earlier.  Pity.

The time-height section from the NAM shows a classic "cloud-storm" environment tomorrow over the Salt Lake Valley.  Cloud storm is a phrase we jokingly call events with high clouds and virga, but little precipitation reaching the valley floor.  There's copious moisture at mid levels, but dry environment down low.  Deep moisture doesn't penetrate into the Salt Lake Valley until Friday night.  More evidence of further delays in the storm really getting going over the Salt Lake Valley.

Add all this up — the delay in the arrival of the trough, the drying influence of the High Sierra, and the dry low levels over northern Utah — and you have a recipe for restless natives anxious for the arrival of a storm that has been promised for Thursday.

3. Many storm chasing options

We of course have a mobile radar, so we can put it wherever we want and don't have to necessarily wait in the Salt Lake Valley for weather.  That being said, it takes time to move the DOW around and configure a reasonable scanning strategy to do real science.   We have a number of possible targets through Friday afternoon, including the spillover of precipitation across the northern Wasatch Mountains and into the Ogden Valley, multi-ridge interactions between the Stansbury and Oquirrh Mountains or Oquirrh and Wasatch Mountains, a cold frontal passage presently forecast for Friday morning, and post-frontal convection in northwesterly flow in the wake of the cold front.  It's going to be a busy time!  We'll have an interesting planning session this afternoon and then will need to keep a close eye on things in the field to maximize our opportunities.

4. Mountain snow possibilities

The situation this week has been pretty grim for skiers.  Not only has it been dry, but it has also been warm.  I don't follow the snowmaking activities at the resorts, but I suspect they were limited at best.  That situation will continue today and even tomorrow looks to be pretty warm.  In addition, the early phases of the storm when it does arrive look quite warm.  For example, the NAM forecast for 1200 UTC (5 AM MST) Friday morning has 700-mb temperatures of around -2ºC.  That equates to a snow level around 7000 feet or so.

Thus, while the upper elevations of Snowbasin are likely to get a pasting, the base may see rain for at least a portion of the storm.  PCMR may also see rain at the base during the early stages.

However, the snow level will be lowering during the period, especially on Friday.  Friday has some potential to be productive in the Cottonwoods due to cold, unstable, northwesterly flow.  Overall, the NAM-12km is generating about an inch of water and 9 inches of snow at Alta Collins through late Friday.  The numbers, however, vary widely across models and ensemble members.  At this point, I'd lean toward 6-12 inches at upper elevations in Little Cottonwood, with the potential for more if the post-frontal environment is highly productive.  Water totals in the northern Wasatch should be higher.

Snowmakers had better be ready to release the torrents Friday, Friday night, and Saturday morning.  After that, ridging returns and more marginal snowmaking conditions return for a couple of days.

Addendum at 10:15 AM 15 November

The SREF plume diagram below was unavailable when I wrote this post, but I've added it here and it shows remarkable spread for the event at Alta Collins.  Talk about forecaster heartburn!  Hope for the high members to verify.

Curb Your Enthusiasm

We've reached the time of year where every wiggle in the jet stream becomes enticing. 

Call it what you want.  The November doldrums.  Purgatory.  There's not enough snow to ski in the central Wasatch and we need more.

I'm already getting inquiries about the storm later this coming work work week.  My advice?  Curb your enthusiasm.

That advice is not because I think the storm will be a bust, but because we are still a few days out and the ensembles are showing a wide range of possible outcomes, from relatively light accumulations to up to 14 inches at Alta-Collins.  Typically, ensembles are underdispersive, meaning that range probably underestimates the full range of possible outcomes.  In other words, we don't really know much yet except that a trough will be coming through and we'll probably see some snow. 

What I do know is that today will be a lovely mid November Sunday.  Enjoy it. 

Perplexing Probabilities

There are a host of challenges posed in the forecast communications business.  One that I thought of this morning as I surveyed the ensemble forecasts is the low probability, high impact weather event.

Fair weather looks to predominate over northern through Thursday, but on Friday, an upper-level trough moves across the northwest U.S. with the trailing cold front racing across Utah.  The NAM calls for precipitation accompanying the front to be relatively light.  Perhaps some valley rain showers and mountain snow showers, but nothing for skiers to get excited about.  

If we look at our downscaled forecasts for Alta based on the Short Range Ensemble Forecast System (SREF) we find that most members are producing very light accumulations of 0.25" of water equivalent or less through 6 PM Friday (0000 UTC 21 October).  Again, nothing to get excited about.  However, 2 of the 26 members are going bigger and putting out about 0.7" of water or so.  
If we look at our downscaled NAEFS forecasts for Alta, most members producing light accumulations, a few in the 0.4" to 0.7" range, but then two outliers that go absolutely huge, generating about 2.5 inches of water and around 25 inches of snow.  Skiing anyone?

Such outliers are unusual, but not unheard.  However, I don't know of any studies that have attempted to look specifically at the reliability of such low probability, high impact forecasts.  The NAEFS forecast above, if taken literally, would yield about a 10% chance of 20" of snow or more on Friday, but a 90% chance of 7 inches or less.  Is that a reasonable forecast of the probabilities?  In addition, if that was a reasonable forecast of the possible outcomes, how best to communicate that to the public and forecast customers?  "Well, we think that there will be some snow showers.  Odds are it won't add up to much, but there's a slight chance of 20."  That should go over well.

I don't have answers for these questions.  We need better validation studies of our ensembles and, as ensembles improve, better ways to both extract and communicate probabilistic forecast information in a way that is useful to the end user.  

The Waiting Game Begins

Last night's trough passage brought a couple inches of snow to the upper elevations of the central Wasatch.

Source: Snowbird
Even at my place, there was a trace of snow. 

The waiting game now begins for the start of the ski season.  As things stand now, the next five days look dry.  Although cold today, we should see marvelous fall weather beginning tomorrow through at least Wednesday and probably Thursday as well.  

After that, we shall see.  The good news is that the GEFS is calling for troughiness over the western US late next week.  

However, snowfall in the Wasatch is greatly dependent on all sorts of factors that cannot be nailed down so far in advance.  Some of those solutions would probably give us a pretty good dump, others next to nothing.  

Thus, at this stage, it's best not to buy into any click bait based on extended range forecasts. 

Need a Recommendation

I've discovered some minor damage to my carbon fiber mountain bike frame.  Recommendations for affordable repair shops or individuals greatly appreciated. 

The Last Hours of Summer

After a blistering hot warmest summer on record, the last few hours of summer giving us a blissful kiss of winter with a fresh coat of of snow in the mountains. 

Source: Snowbird
It's a bit difficult to gauge accumulations from what automated sensors are running right now in the Wasatch, but my guess based on the Alta-Collins total snow depth sensor is that they got about 3 inches or so overnight.  We'll see some additional snow showers today, although accumulations will perhaps be an angry inch. 

Tomorrow may provide a bit of a cold break, but given the instability, the possibility of snow showers remains.  Although accumulations will likely be limited, if I get out hiking, my plan will be to be ready for anything.  Even a brief period of snow squalls can be damn uncomfortable if you are traveling light. 

The big question mark for the weekend is what will happen as the trough swings Saturday night and Sunday.  The models are hinting that a more organized band of precipitation may develop during that period and affect the Wasatch.  See, for example, the NAM forecast below. 

Most members of our downscaled SREF generate about 0.2-0.5" of water during that period, which would probably equate to another 2-5" of snow.  There are, a couple members that are more enthused, including one ARW member that generates over 2 inches of water by Monday morning.  Ah, one can always find a solution to their liking in the ensembles, but recognize it is a low probability outcome. 
The bottom line for this weekend is keep an eye on forecasts and be prepared for cold unsettled weather if you are heading out and trying to do some hiking this weekend.  I'm already assuming that mountain biking is probably a nonstarter outside of the lowlands.

Irma Track Troubles

Hurricane Irma remains an extremely powerful and dangerous category 5 hurricane, after passing over St. Martin this morning.

Source: NOAA/NWS/National Hurricane Center
Here's a shot as the system passed over Barbuda overnight.

Source: NOAA/NWS/National Hurricane Center
Wind observations from Barbuda show the remarkable increase in wind speed with the approach of the eye, reaching a maximum sustained velocity of nearly 120 mph and a peak gust of 155 mph before the sensor gave out.

Source: MesoWest
These are simply incredible numbers, especially that the winds at the time were out of the north and the site is on the south side of Barbuda.

From now through Saturday, the center of Irma is expected to track just to the north or along Puerto Rico, the Dominican Republic, Haiti, and eastern Cuba.

Source: NOAA/NWS/National Hurricane Center
Forecasts for Florida and the southeast US are remarkably tricky and depend strongly on when Irma takes the anticipated northward shift in track.  To highlight the abrupt nature of the anticipated track shift, I've put together a loop of the GFS forecast below (based on lower resolution grids).  Note the very steady WNW progression of Irma until it approaches the Florida Peninsula, when it decides to alter course and move northward.

The loop above is just one model solution.  Below shows the tracks produced by the GEFS ensemble, which fan out from the eastern Gulf of Mexico to tracks that don't even reach the US mainland.
Source: Tropical Tidbits
Forecasters will have these and additional tracks from the ECMWF and Hurricane models to ponder.  Today's forecasts are especially critical given the long time scales needed for hurricane preparation and evacuation from the Keys and South Florida. Keep in mind that we are still talking about a 4-5 day forecast from now until Irma is flirting with Florida.  The issues at play at these lead times are well summarized by the 5 AM AST forecast discussion from the National Hurricane Center.

Source: Source: NOAA/NWS/National Hurricane Center
The bottom line is that there is a wide range of possible tracks, and as a result, details of the timing and magnitude of Irma's impacts on Florida and other southeast US states remain uncertain.

Morning Eclipse Nowcast

Morning has broken and things are looking good along the path of totality over the western U.S.  It's early for a visible loop, but as can be seen below, a band of mid and upper-level clouds is slowly but surely exiting the totality path area of far western Wyoming and Idaho.  

Here's a look at the GOES-16 geocolor imagery from CIRA at 13:32 UTC (7:32 AM) and it looks pretty good too.  There are some thin patchy clouds over portions of the Idaho panhandle and western Montana moving southward that are just barely discernible in the imagery.   

Source: CIRA
You couldn't ask for a better forecast from the operational HRRR than the one below, valid about a half hour after totality.  No clouds predicted along the entire track from Wyoming to the Pacific Coast. 

Source: ESRL
That forecast might be a bit optimistic as there may be a few high clouds around over Idaho and the remnants of the cloud band evident that is over southern Idaho and western Wyoming this morning could linger near and along the totality track over central Wyoming, as indicated by the NAM forecast below.  

Nevertheless, conditions look quite good for the totality track from Jackson to the Willamette Valley with just the threat of some thin clouds spilling down from the north.  You can see these clouds, for example, in the Montana Snowbowl web cam image below.  

Good viewing to those of you along the path.

Addendum @ 8:25 AM MDT

Now that the sun is a little higher, visible satellite imagery shows quite a bit of smoke in the valleys of the central Idaho Mountains.

Based on web cams, I don't think this smoke will obscure the sun, although it may redden it.  Hopefully it won't spoil any views.  

Forecast Outlooks and Products for Eclipse Planning

As I write this, we are now about 3 days from the 2017 Eclipse and the ultimate test of transportation and communications infrastructure in rural areas of Idaho and Wyoming, where the vast majority of Utahns and University of Utah students hope to view the eclipse (Question: Will anybody attend class on the first day of the semester, which is also Monday?).

The large-scale forecast for Monday seems to be stabilizing, but I still consider highly specific cloud forecasts to be difficult given the weak large-scale forcing.  A high-amplitude ridge parked over our area would be a godsend for forecasters, but that's not what we're looking at for Monday.

Instead, the GFS calls for a weak upper-level shortwave trough to sweep across Idaho and Wyoming from 1200–1800 UTC (0600–1200 MDT), and be accompanied by some mid- and high-level clouds.  A short-wave ridge further west builds along the Pacific Northwest coast.

Eclipse time is approximately 1721 UTC (1121 MDT) in Redmond, OR, 1730 UTC (1130 MDT) in Stanley, ID, 1736 UTC (1136 MDT in Jackson, WY, and 1741 UTC (1141 MDT) in Riverton WY.  That's just before the bottom image above, which given the GFS forecast would yield the lowest cloud cover odds and fractions during the eclipse over eastern Oregon and increasing cloud cover odds and fractions as one moves eastward to western Wyoming.  

The NAM agrees with the basic synopsis being advertised by the GFS, but note that the shortwave trough orientation is more from SW to NE (positively tilted in meteorological vernacular, and that the structure and characteristics of the clouds varies when one examines the gory details.  

That variation in the structure and characteristics of the clouds represents the dilemma for forecasts along the path of totality over Idaho and Wyoming.  This is a weak shortwave trough, so a routine "public" forecast would be pretty straightforward.  Probably mostly sunny given the fact that some mid and high level clouds aren't going to be a big deal.  

However, exactly where and when clouds will be at the time of eclipse is difficult to ascertain.  Will one have a clear view of the sun in Jackson, but have an untimely patch in Driggs?  Impossible to say.  In part, this reflects the unpredictability of such cloud cover at such long lead times, but also the fact that present day forecast models do not explicitly resolve cloud processes, adding to the forecast uncertainty.  Timing will also matter.  For example, if you just happen to be underneath a local area of clouds at the time of eclipse, that's a bummer.  

Based on current forecasts, the greatest likelihood of clear skies over the interior mountain west eclipse path is eastern Oregon.  The potential for some mid or high clouds exists as one moves eastward, especially over eastern Idaho and Wyoming.  The timing, location, and extensiveness of that cloud cover remains uncertain.  

For your planning purposes, here are a few products for your consideration:

1. NWS Digital Forecasts. You'll need to use the drop-down menu to request "Sky Cover (%)" and select the appropriate time.  12 PM is the closest available.  One disadvantage of these forecasts is that they are "deterministic" and don't show the full range of possibilities.  Numbers represent percent of cloud cover.  

Source: NWS
2. Experimental High Resolution Rapid Refresh (HRRRX). HRRRX is an experimental version of the HRRR that is being developed and tested by the NOAA Earth Systems Research Laboratory for future operational implementation.  Forecasts are available out to varying lead times (probably dependent on computer time availability), including cloud products.  At 3-km grid spacing, this is the model to go to for short-range cloud-cover guidance.  In addition, they have added sun-obscuration mods to account for reductions in solar radiation during the eclipse (details here).  I've been told that a more "crash proof" web access to the HRRRX is available here.

Keep in mind that even at short time scales, errors in cloud cover are to be expected.  Use the HRRRX (and other model forecasts) as guidance, but not absolute truth.

3. GOES-16 Imagery.  I'm a huge fan of these high-frequency, geocolor loops from CIRA.  Use for "eclipse chasing" the morning of the event and fine tuning during the event, if traffic permits.

Hopefully, the forecast verifies with minimal clouds and not an unfortunate veil of thick cirrus.  Good luck!

Potentially Harsh Realities of Eclipse Chasing and Cloud Forecasting

A lot of assumptions, many quite poor, are being made concerning sky cover for Monday's eclipse.

Meteorologists who have to produce very detailed cloud forecasts, particularly those in which even a thin layer of clouds can make or break military mission success, alternative energy projections, or civilian aviation operations, know just how difficult it is to predict the existence of a thin cloud layers.  I once met the director of Air Force Weather and asked him what is number one forecast problem was.  He said instantly and without hesitation, "clouds."

Which brings us to the eclipse.

If you are desperate to observe totality, you have a window of less than 3 minutes and one thin patch of cloud could totally ruin your experience.  It doesn't take much to generate a thin patch of cloud.  There are many days in which the sky is partly cloudy and there's a patch veil of altostratus or cirrostratus.  Many.  You don't think about the weather on these days, because the weather is just fine for regular activities, but you would if you were in the Air Force or the solar industry.

Today provides a great example of a fair weather day that might just ruin your eclipse experience.  Let's imagine that the eclipse was today in Salt Lake City.  The 500-mb pattern is fairly innocuous, but there is a weak trough over Nevada and California.

The National Weather Service forecast calls for mostly sunny skies and a high of 88ºF.  A beautiful summer day.  You'd probably be excited if that was the forecast for the eclipse.

Not so fast.  A look at the satellite imagery shows patch altostratus over portions of Nevada and northern Utah.  On a meteorological scale of 1 to 10, these rate about a 0, but on an eclipse risk scale of 1 to 10, they could be a 10 if you happen to be in the wrong spot for that 2 minute period of totality.  

Sound crazy?  What if the eclipse was happening when I walked out of the Student Life Center?  Sorry, you're partially obscured.

Well, that's not so far from the cloud edge, you could just drive a bit down the road, if it wasn't clogged with traffic.

Or maybe the eclipse happens a bit later?  Hmmmm....

Here's something for you to ponder.  You wake up Monday morning and there's altostratus around.  Those are benign clouds for garden variety weather, but a disaster for eclipse viewing.  Do you move elsewhere or wait it out and hope for the best?  Alternatively, it's later in the morning.  The moon is sliding in front of the sun.  Traffic is everywhere because a million people have converged on the totality strip.  There is a patch of altostratus between you and the sun.  Totality is 30 minutes away.  What do you do?  Do you try to move?  Do you sit it out and hope for the best?

I have dealt with these sorts of problems during field campaigns, especially those with mobile radars. We can't precisely predict where storms will develop and you're not always in the right spot.  It can be an agonizing decision about whether or not to stay put, hoping the storm shifts, or move.  Moving has consequences.  It takes time.  You can miss out while on the road (at least with the eclipse, you know exactly when it's going to happen).  Computer models offer no help under such a scenario.  It's just you and the radar and your best nowcast.

Such a scenario is not out of the realm of possibility for the eclipse, where it could be just you, your eyes, and the clouds.  Forecasts for Monday include solutions with weak flow and mid-level moisture over western Wyoming and southeast Idaho, prime territory for Salt Lake eclipse viewers.  The GFS forecast shows the potential for mid level cloudiness over eastern Idaho and Wyoming Monday morning (eclipse time is around 1730-1740 UTC over Idaho and western Wyoming).  Time height sections (not shown) show some mid-level moisture in the region.

I can't tell you if there will be midlevel clouds because we're still dealing with a 5+ day forecast with a lot of uncertainty.  I hope the morning dawns "severe clear" for you without a cloud in the sky.  But what if there are some clouds around?  What would you do?  Could a shift of a few miles do the job?  Would you need to go farther?  In times like that, as Dwight Eisenhower said, "plans are nothing, planning is everything." 

The "Official" 2017/18 Ski Season Outlook

Released yesterday with a tweet of Trumpian proportions, we now share via conventional blog channels the Official Wasatch Weather Weenies 2017/18 Ski Season Outlook.

The outlook serves as a friendly reminder of the limited value of extended outlooks for doing any real planning for your ski season, vacation, or adventures.  The skill of such outlooks is generally low and there's really little to help guide us this coming season since it is likely to be a "No Niño" winter (see No Niño Winter Likely Ahead).  There may be a slight loading of the dice for a warmer winter and below average snowfall in warmer, lower elevation regions, but that's about it.  There are equal chances of above average, average, or below average snowfall wherever you are planning to ski.

That also means that if you are looking for powder this winter in the contiguous U.S., the best odds are found, as usual, in the Cottonwoods, because of their highly favorable climatology.  Globally, the best odds for powder are found in snowy regions of Japan's Honshu and Hokkaido Islands in January  (see my book Secrets of the Greatest Snow on Earth).  The challenge there is often not finding deep, but finding steep (and clear, at least in January).  It's there though if you know where to look and are willing to earn your turns.

Professor Powder enjoying a rare January bluebird day in the Hida Mountains, Japan