Alps Weather 2023

jamesdeluxe

Administrator
By the end of the incoming storm, I'm going to start a new Alps weather/conditions thread with a less unpleasant subject line. [Done, TC] Here's Fraser's report:


The much-anticipated storm has arrived in the Alps, but it won’t be until later tonight that all areas get in on the action. Significant snow has already fallen today in some south-western parts of the Alps and will spread to most other regions later today and overnight.

With the angle of attack initially from the south-west, it is the southern Alps (especially the south-west) that will get first dibs on the heaviest snow. At least 15cm has already fallen in Auron/Isola 2000 in the far southern French Alps, for example, with a rain/snow limit around 1400m.


Further north, there have been bits and pieces of snow in the northern French Alps (e.g. Portes du Soleil) and western Swiss Alps (e.g. Zermatt) but with a higher rain/snow limit, which started at around 1500-1700m but is set to drop lower during the middle part of the day.

This storm will reach the eastern Italian Alps (Dolomites), the eastern Swiss Alps (e.g. Davos) and the Austrian Alps later today and overnight. The rain/snow limit will drop to 1000m or lower in many areas by Monday morning but will remain slightly higher for a time in the far south-east (Dolomites).

Monday will see snow continuing at times across many parts of the Alps though precipitation will be more sporadic and lighter in the south (e.g. Sestriere, Monte Rosa, Livigno) than in the north (e.g. Avoriaz, Engelberg, Lech). The rain/snow limit will continue to fall, especially in the northern Alps where it may drop as low as 400-600m in places later in the day.

By Tuesday morning, between 25-50cm of fresh snow is expected across a good portion of the Alps, with a bit less in the extreme south-west (e.g. Isola 2000) and extreme north-east (e.g. Schladming). The greatest accumulations of snow are likely to be at altitude in the northern French Alps with 60cm possible at higher altitudes close to the Massif du Mont Blanc, in resorts such as Chamonix and Flaine.

Lower down in the north-western Alps, where the snow is most needed, around 10-20cm is expected at 1000m by Tuesday morning in resorts such as Morzine. Megève, Villars, Gstaad and Grindelwald.

After a better weather day in the Alps on Tuesday, a new storm will arrive on Wednesday, this time mostly affecting the northern and especially north-western Alps. The rain/snow limit will briefly rise towards 1600-1800m before falling to 1000-1300m later in the day.
 
Last edited by a moderator:
The weather/snowfall sites Open Snow and WePowder have become very optimistic about the coming Alps storms.
(I believe their models are mostly built off of the Euro and American models - plus local knowledge/elevation data. The Storm Journal interview with OpenSnow founder Joel Gratz was very interesting - supposedly they can now automatically model snowfall for any GPS location on earth).

Fraser seems to look at models and add insight - like where Joel of OpenSnow was in his early days of 2010 when he just did an email list for Colorado.

I like the new Mont Blanc forecasts. Guess there are 2 areas of Europe where turnarounds can happen more quickly: NW Alps and Voralberg/ West Austrian Alps. Obviously not as drastic as the Sierra, but I will take it. More importantly, there is snow in the valleys/base.

94DD29D7-8D82-4E69-908B-0BD6CB8A566D.jpeg

D53BDBB7-FA23-4B00-8D06-A1D9181CD898.jpeg


Open Snow

A454A263-8679-427F-B999-E8A2FA82011E.jpeg
26D27D2F-88F6-4EBC-B3C3-2601AA5DC833.jpeg
50462AC1-0B27-44F7-80BD-6FDD984203D2.jpeg
 
Last edited:
There's the market correction we've been expecting! :eusa-dance:

I want to say something about weather will generally start to revert to standard climatic averages - unless you just get locked into a pattern. However, I am afraid of a possible forthcoming statistical treatise/analysis. So I am going to leave it at I like the new forecasts - however not sure I trust the OpenSnow all Earth models. Some were way too optimistic over Christmas week.

Guess I will make some final calls next weekend. Really, I am game for anything - as long as I am adding some unskied resorts or mountain sectors to my ski resume. Adding Sestriere, Alpe d"Huez, Puy St. Vincent, and more parts of Serre Chevalier + Les 2 Alpes would be great. Or getting to ski more off-piste in 3 Vallees, Val d'Isere, and checkout Ste. Foy would be equally brilliant.

I am almost ready to give up my seat on the heli to one of the other guys and apply the money to my own personal guide for the day.
Give the guide my itinerary/goals:
  • Toula Glacier here at least to mid-station AM
  • Marbree Glacier here less likely since it looks like you ski to valley
  • Courmayeur resort off-piste after lunch PM
Think something like this is possible since my brother and I in Chamonix were able to first cable car, play tourist at Aguille du Midi, ski advanced itinerary, eat lunch and ski to the base/town by 1130 am. Quick enough that the guide suggested another run or to go down to Grand Montets. Not in the cards since traveling to Zermatt with road closures.

Been reading my guide book too much - it's mostly about Courmayeur. I know, there is still no snow below 2000m or so ....yet.

1673236176295.png
 
However, I am afraid of a possible forthcoming statistical treatise/analysis.
Nah. As a big picture, I see it exactly the same way. You ask me what Mammoth will end up at the end of the season, and I'll always say assume average after the one week in advance we can predict. I've been reading OpenSnow in detail for a long time, and it seems the deterministic models that produce the numbers you see when you look up those 10 day forecasts for individual resorts overproject. I'm fairly sure these average a lot of model forecasts, including optimistic outliers. I'd guess they should use medians instead of averages. When you read the narrative forecasts by the local meteorologists they are more conservative, and in general they won't quote numbers at all beyond 5 days.

We are fairly much on the same wavelength as ChrisC on resorts we are considering, and have as yet no reservations for anything other than the car rental.
 
I've been reading OpenSnow in detail for a long time, and it seems the deterministic models that produce the numbers you see when you look up those 10 day forecasts for individual resorts overproject.

Just looked at the OpenSnow Euro numbers this AM...Wow! They have every major resort on the North-to-West fringe of the Alps forecasted to receive 60-70"+ in the next 10 days (Alpe d'Huez/La Grave to Chamonix/Flaine to Verbier/Crans Montana to Engelberg to Lech/St. Anton). We will see about any accuracy. Assume a Northern Stau is coming?... Perhaps the best weather pattern for snow in the N and NW Alps. WePowder has a great description of this phenomenon.

Also, I cannot tell how OpenSnow takes elevation into account. For example, often Avoriaz=Morzine snow forecasts. And Zermatt, Verbier, and Engelberg have a single number for 5-6k of vertical?

(I've also emailed Joel a lot about his Euro numbers in the past - primarily about snowfall reports and forecasts. They never equate. Always off by a factor of 2-5x or more).

1673278795626.png


1673278852867.png

1673278897904.png

1673278956182.png

1673278997915.png
 
Last edited:
But even wePowder has become more optimistic. I think its estimates for snow are likely more accurate.

1673279189469.png
 
Just looked at the OpenSnow Euro numbers this AM...Wow! They have every major resort on the North-to-West fringe of the Alps forecasted to receive 60-70"+ in the next 10 days (Alpe d'Huez/La Grave to Chamonix/Flaine to Verbier/Crans Montana to Engelberg to Lech/St. Anton). We will see about any accuracy. Assume a Northern Stau is coming?... Perhaps the best weather pattern for snow in the N and NW Alps. WePowder has a great description of this pattern.

Also, I cannot tell how OpenSnow takes elevation into account. For example, often Avoriaz=Morzine snow forecasts. And Zermatt, Verbier, and Engelberg have a single number for 5-6k of vertical?

(I've also emailed Joel a lot about his Euro numbers in the past - primarily about snowfall reports and forecasts. They never equate. Always off by a factor of 2-5x or more).

View attachment 33225

View attachment 33226
View attachment 33227
View attachment 33228
View attachment 33229
Fairly speculative at this stage as there is huge model volatility. The last GFS operational run was looking at apocalyptic levels of snow for the north-western Alps around the middle of next week. The very latest one has removed that scenario entirely. The only thing we can be sure of at this stage is that after this storm, which finishes early tomorrow with snow falling to around 600m, it will turn milder with further precipitation at times over the next few days, but also with milder air in the mix, which means that the rain/snow level could 1700-1900m by Friday/Saturday. Next week though is anyone’s guess, so don’t get hung up on the automated forecasts. That's not to say it couldn't happen, just urging caution...
 
Fairly speculative at this stage as there is huge model volatility. The last GFS operational run was looking at apocalyptic levels of snow for the north-western Alps around the middle of next week. The very latest one has removed that scenario entirely. The only thing we can be sure of at this stage is that after this storm, which finishes early tomorrow with snow falling to around 600m, it will turn milder with further precipitation at times over the next few days, but also with milder air in the mix, which means that the rain/snow level could 1700-1900m by Friday/Saturday. Next week though is anyone’s guess, so don’t get hung up on the automated forecasts. That's not to say it couldn't happen, just urging caution...

Oh the Americans.....and their models

By 2015, the GFS model had fallen behind the accuracy of other global weather models. On March 22, 2021, the NOAA upgraded the GFS model, coupling it with the WaveWatch III global wave model, which will increase the GFS's resolution from 64 to 127 vertical levels, while extending the WaveWatch III forecasting window from 10 to 16 days. This left some meteorologists hopeful that the GFSv16 upgrade would be enough to close the accuracy gap with the ECMWF's model, which was considered to be the most accurate global weather model at the time.
 
As if to reinforce my point about model volatility, the ECM European model offers an entirely new solution for next week with the surge of arctic air (and accompanying cyclogenesis) way further west then before (see attached). This would mean snow for the south/south-western Alps but very little for the NW. Basically I have almost zero confidence in any of the models past about 5 days right now! This chart would be absolutely perfect for Portugal's one tiny resort though!
 

Attachments

  • ECM1-168.gif
    ECM1-168.gif
    54.7 KB · Views: 79
In defense of OpenSnow, all of their live forecasters will not quote forecast snowfall numbers more than 5 days out. This tells me that they don't trust the models beyond that time.

sometimes you can say with 99% certainty that it won't snow for 10 days.
During last year's midwinter drought from hell, I noticed that the forecasters WERE willing to project dry spells beyond 5 days even though they won't project precipitation forecasts any farther out.
(I've also emailed Joel a lot about his Euro numbers in the past - primarily about snowfall reports and forecasts. They never equate. Always off by a factor of 2-5x or more).
Andes numbers on snow-forecast.com are often in the optimistic fantasyland range too. If you just run models and don't curate the results, this is often what you'll display.

OpenSnow is a North American company. Anything for Europe is likely to be just what the models spit out with no review. This is the first season OpenSnow has had a Euro narrative forecaster. He's calling for 40-80cm in the snowiest locations for each of the upcoming two storms.

Assume a Northern Stau is coming?..
My understanding is that a Northern Stau is very cold and hits mainly the north to northeastern Alps, January 2019 being a classic example. The weather forecast over the next week is coming from the NW off the Atlantic with variable rain/snow line but thankfully not nearly as warm as before Christmas.
 
Andes numbers on snow-forecast.com are often in the optimistic fantasyland range too. If you just run models and don't curate the results, this is often what you'll display.

I have given up on snow-forecast. I like that they try to bring elevation into the mix, but everything is always so far off.

OpenSnow is a North American company. Anything for Europe is likely to be just what the models spit out with no review. This is the first season OpenSnow has had a Euro narrative forecaster. He's calling for 40-80cm in the snowiest locations for each of the upcoming two storms.

I tried to explain that a reliable Euro report might be more lucrative for subscriptions than North America since Euro resorts are closer together and it's easy to road trip between them in pursuit of snow. You just need accurate forecasts and reports - aggregated.

Storm Chaser Steve is always on a plane in North America. Unrealistic, but fun to love thru virtually. In Europe, it's just a car.

I've met quite a few Brits on a month-long Euro ski road trip just following the snow - living off of storm forecasts.

The various valleys and exposures on the W/NW/N sides plus SW/S/SE sides of the Alps all result in very different outcomes. But you can easily road trip from Chamonix to Gressoney/Alagna in 2 hrs if the storm comes from the South.

Plus there is no competition. No one really forecasts anything reliably for individual Alp resorts, reports snowfall accurately, factors in elevation, etc. It's a mess! In North America, this is mostly standardized by the resorts: Europe/Alps - not so much. The Euros just seem to care about how many pistes and km of pistes are open.
 
Will have to inquire about where these totals came from......but a healthy storm? the last few days. Can't tell if there is some double counting in there....

1673381831208.png

1673381926178.png




And snow-Armageddon ....or Snowmegoddon model

1673382289343.png
 
Fraser's report today. Looking good for big snow next week but he's still non-committal:

As for the rest of this week, Thursday will be mainly dry then, on Friday, a few more centimetres of snow are likely to fall in the north-western Alps with a rain/snow limit around 1200-1600m. From Saturday onwards, the weather will become very unsettled indeed as a succession of storms invade the Alps from the west then north-west, lasting well into next week.

It will be mild at first on Saturday, with the risk of rain to 1800-200m in the north-western Alps, before turning colder everywhere with a chance of significant snow for much of the Alps from Sunday onwards. Indeed, some weather models are predicting huge amounts of snow next week, almost anywhere, although the detail is still hard to pin down at this stage.
 
Fraser needs to have a talk with OpenSnow.

OpenSnow sucks for Europe ....

More powerful together, dominate.

Although I was really negative about Joel saying he only wanted to do subscriptions, no more advertising?! He continues to suck with reporting.....don't know what he is doing.
 
Last edited:
Fraser's report today. Looking good for big snow next week but he's still non-committal:

As for the rest of this week, Thursday will be mainly dry then, on Friday, a few more centimetres of snow are likely to fall in the north-western Alps with a rain/snow limit around 1200-1600m. From Saturday onwards, the weather will become very unsettled indeed as a succession of storms invade the Alps from the west then north-west, lasting well into next week.

It will be mild at first on Saturday, with the risk of rain to 1800-200m in the north-western Alps, before turning colder everywhere with a chance of significant snow for much of the Alps from Sunday onwards. Indeed, some weather models are predicting huge amounts of snow next week, almost anywhere, although the detail is still hard to pin down at this stage.

This OpenSnow forecast.....is in Lala land like Snowe-forecast.
 
Also, I cannot tell how OpenSnow takes elevation into account. For example, often Avoriaz=Morzine snow forecasts. And Zermatt, Verbier, and Engelberg have a single number for 5-6k of vertical?

(I've also emailed Joel a lot about his Euro numbers in the past - primarily about snowfall reports and forecasts. They never equate. Always off by a factor of 2-5x or more).
If you click through open any Euro resort you'll see an elevation number at upper left. The computer model is presumably using that single elevation for projections. My educated guess is the elevation is the average of highest and lowest lift service. For several of the low places in the NW that got deluged with rain at Christmas which I checked, the projection elevations are in the 5,700 feet/1,700 meter range. Verbier projects at 7,241 feet/2,207 meters. To get lower you can look at Bruson, which projects at 5,135 feet/1,566 meters.

If you're going to set up an automated algorithm, midpoint elevation is a logical choice. But you're going to miss some rain events at base elevations. Remember Joel started doing this in Colorado, where it essentially never rains in ski areas during ski season. And when OpenSnow expanded in North America, the verticals as noted are for from what they are in Europe. I can assure you that the narrative forecasters in the PNW, Tahoe, SoCal and New England put a lot of effort into forecasting the rain/snow line during storms.

I'm sure that there is much wider dispersion of model forecasts for 6-10 days than for 1-5 days. That dispersion is limited to zero at the low end but has no upper limit. Since that is a skewed distribution, averaging the individual model runs is likely to result in overprojecting more often than underprojecting.

We would all like to know what's happening 6-10 days out, but model accuracy is poor. I pay more attention to the narratives by Fraser, WePowder and the new guy at OpenSnow this year. These people concentrate on the 1-5 day forecasts for good reason. And they also pay attention to the rain/snow line.
 
Back
Top