 Poster: A snowHead
|
|
@Peter S, very interesting! Quite a drop in temp over last 10 days!
|
|
|
|
|
|
Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
|
That climate reanalyzer chart is for quite a specific section of the north atlantic, roughly the part north of spain to south of greenland. And while the western section was already below average, the big change is a reduction of warm anomalies to the W-SW of Ireland....albeit it's still above average. I think the cooling here is a result of the persistent troughing we've seen there over the last 2 weeks....several ex tropical storms have combined with the westerly jet and ended up stuck in this region as there is still a large block of high pressure over scandi / central europe preventing the lows moving east.
Low pressure systems lead to an upwelling of deeper cold to the surface, so while any cooling is welcome, it may just be a temporary surface change.
Then again the pattern of low's forming here looks set for another week or so at least....so could well see a further reduction in the warm anomalies.
One area that doesn't seem to be contributing is the melting greenland icesheet. We are right at the point now where arctic regions start to net gain mass of snow and ice as the summer melt season ends. And for the 5th year in a row, net summer loss of the central pack has been below the long term average. This is a 50 year average going back to when satelite data began to record arctic sea ice in 1979.
Could even make a case that at 5.5m sq km's end of August, we are close to / slightly above 2007 levels, but clearly still in a downtrend, with 2012 remaining the record melt year....so far.
Some early seasonal snowfall predictions are emerging.....UKMO august update is absolutely abysmal, projecting very low snow for europe, and even north america....despite la Nina. ECM however is projecting a huge Januray positive anomaly for the alps. All just for fun at this range....will get a first look at multi model autumn projections which come out on the 10th Sep.
Near term enjoy the settled warm weekend because it looks very short lived.....more westerly low's on the way next week with another dip in temps into mid month.
|
|
|
|
|
|
Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
|
|
Is ECM still considered the state of the art model or have others caught up in terms of tech?
|
|
|
|
|
|
You need to Login to know who's really who.
You need to Login to know who's really who.
|
@afterski, can't find the long term stats, but below for August is a good representation. Generally it's ECM top with UKMO, GEM (CMC), GFS (NCEP), JMA etc fighting for 2nd to 5th. Meanwhile AIFS often beats ECM, so the quicker AI models often outperform the traditional physical based ones. I'll find the stats another time.
In any case, all of these verification studies are short term, above is 5 days....haven never come across 5 months stats, as I suspect they don't have much skill.
Best we can do is take the average of the best long range models, which will be updated for Sep next week.
|
|
|
|
|
|
Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
|
Some reading on the european impacts of SST's around the sub polar North Atlantic hasn't cleared the air much for me.
Often referred to as the 'cold blob', just south of Greenland, its official name is actually the North Atlantic Warming Hole (NAWH). Turns out the official term refers to the fact that it's not a hole that is warming, but rather a cooler hole in the wider sea surface warming trend.
The clearest link to european winter seems to be a correlation with -NAO, so a weaker westerly jet bringing a more northerly sourced air flow. And in very simple terms you'd expect cooler seas to have less warming moderation on that cold air. All good, and ties into other drivers / expectations for this winter such as impacts from La Nina and -QBO. But a lot of the research seems circular, ie which one is driving the other.
The AMOC is the inter-decadal trend showing very gradual weakening in the gulf stream, as colder fresh water from Greenland interrupts the northward flow of tropical heat.
While 2023 and 2024 saw very warm north atlantic anomalies, it's presumed that El Nino and the general warming trend over-rode AMOC weakening / greenland melt cooling.
The NAO can also be argued to have opposing effects, eg in the positive phase stronger westerly flow allows for greater northward heat transport from the gulf, but at the same time the stormier weather mixes out the warmer surface with cooler water below....so it can be argued both ways.
Another interesting conflict is whether the northern arm of the jet stream is trending further north or further south. I've mostly read about the impact of the expanding Azores high on milder european winters, forcing the jet to the NE, and this is easy to see with re-analysis tools over recent decades (as well as just looking out the window).
But a recent study is gaining attention arguing that the melting ice will drive the northern arm of the jet about 300 miles further south in time. Theory being as the arctic area is warming faster, and the sub polar north atlantic is cooling, the atmospheric temperature gradient line (which drives the jet strength) will move southwards.
That's as far as i got....and it's topical now as we have an unusually southerly jet for September into NW europe. Whatever the reason it's a pattern we want to see continue and shift further east. There will be plenty of dry spells as always...but early patterns are useful to watch as we know they can become entrenched (eg Azores-Iberia low last few seasons).
Modelling beyond this week shows the scandi high will finally ease up and allow these low pressure systems to move east into the north sea area. This should keep a lid on temps for most of europe, but as usual it will depend on where any ridges develop behind.
Here's the most wintery chart I've seen so far this autumn....very unlikely to happen at t+300 hrs, but the general trend east is showing now on all models, and it's good to see that these potential colder patterns are at least entering the fantasy world of possibilities this early. Other models btw are much flatter, keeping any cold further north.
Ensembles are looking very tight for the next 6 days, meaning you can have high confidence in what they are showing.....slight cooling, then milder 15th for NW alps. Beyond the mid point though the scatter is large...I'd guess there will be cooling at the eastern end of the alps but pure speculation for now.
Seasonal wise still waiting on the main models update tomorrow, but ECM does come out a few days earlier than the others every month....so I had a look at the Oct-Nov-Dec forecast compared to last month. August update on the left.
 
Continuing with todays theme I think you could argue both ways about which update was better.
Negatives....Sep update looks flatter, northerly jet, higher pressure / milder over alps
Positives....peak atlantic high is more central with higher anoms over Azores-Iberia, allowing NW flow at times, more like late last autumn.
In any case, it's a 3 month anomaly from one model....see what the others say
|
|
|
|
|
|
You'll need to Register first of course.
You'll need to Register first of course.
|
|
Thanks polo - appreciate your useful insight
|
|
|
|
|
|
|
|
@Chris Brookes, there's always something interesting going on, even if now it's incredibly monotonous 
Looks like a stream of atlantic lows heading E / NE across the north sea, probably 3 seperate systems over the next week or so. September already looking above average ppn north of the alps. A few blustery days in there, nothing extreme but fairly remarkable change for Ire-UK after a long settled dry period....the switch has flipped early. Likely to warm up in general after 15th as high pressure gradually builds in following the troughs SW-NE.
Personally, even though it's largely dry and mild here, cool nights, warm afternoons, it's my favourite season in terms of how the season change feels.....apart from any excitement of winter weather at some point, I won't miss the flies, wasps, mosquitoes, 25c nights etc....and I've had my fill of courgette-tomato-basil harvest recepies....bring on the swiss fondue, cosy fires and thermal underwear.
The Sep multi-model update was uninspiring....just a bland weak signal across every month for slightly above average pressure over central europe. You could say c'est pas terrible, but at the same time there was scant formidableness.
On to the next one.
Hurricane seasons have been my focus this week....looked at data back to 1850, but aware that from 1970 on is more accurate due to satellites. Am dubious as to how much impact it has on europe anyway, given that recent years have been above average ACE (a measure of tropical storm strength and duration). However this year is currently well below average...we are at the mean peak date now around Sep 10th yet the tropical atlantic is so far very quiet. Still a month to go for catch up of course but either way we already have warm seas to the north etc, so maybe we'd don't need any more heat transport. And so far so good for all those living in the gulf of Mexico / America 
Not much to add....north-south split ongoing, very flat, probably turn milder 3rd week....good a time as any to start chopping wood.
|
|
|
|
|
|
|
|
Noticed the UK is still running well above average temps even in September, showing +2c up to the 12th, now dropped to +1.6c. The Met office use quite an old baseline though of 1961-90, while most anomaly charts are vs 1979-2009, or even 1990-2020. While it might not have felt super warm, apparently it's the high nighttime minimums propping up the average temperature each day, possibly due to warm seas to the north and east. On the other hand parts of Scotland (Kinbrace, Hatchery 103m asl) saw their first sub zero's of the season yesterday.
Somewhere else that may get a cold blast is the southern hemisphere, looking at a rare stratospheric warming developing over the antartic.
Our own polar vortex is up and running now and off to a slow start. A lot of hope this season that the Nina / strongly -QBO combination will keep it spinning at below average speeds so it can release more arctic air into mid latittudes.
Strat PV wind forecast at 10hpa
NAO has been largely negative since about August 16th, leading to the jet stream dipping more south than usual. While AO is hovering around neutral but possible weakening into last week of Sep.
Nicked this graph off GavsWeatherVids.....looks like solar cycle 25 peaked early (Sep 24), you'd normally expect it to peak this year (being 5.5years since the last minimum in Dec '19 within an 11 year cycle), but it has been unusually weak so far in 2025 in terms of sun spots so that can't be a bad thing either.
NW alps...nice warm dry spell out to the 21st before turning cooler again....hint of a brief northerly flow still from around 22nd with an atlantic-iceland ridge appearing
The AIFS gif here shows temp anomalies at 1500m for rest of Sep,. starting now with the mini heat wave this week and then the potential northerly blast into the last week....obviously the latter frames likely to change, as other models have the cold air further west for now
|
|
|
|
|
|
You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
|
|
Signs of Snow over the Scottish Highlands this weekend , also hasn’t felt anywhere near above average so far this Sept in North Eseex
|
|
|
|
|
|
|
|
|
|
|
snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
|
Freezing level dropping and over 50cm of snow forecast at altitude in Saas Fee Only six weeks until we get there to try out our shiny new Magic Passes.
|
|
|
|
|
|
And love to help out and answer questions and of course, read each other's snow reports.
And love to help out and answer questions and of course, read each other's snow reports.
|
|
|
|
|
|
@mountainaddict, you seen WePowder magic number machine? There are 3 Zermatts and 2 Saas Fees currently in the top 10 this week, but Saas Fee/Staffal actually lead the mid mountain totals at 90+cm.
https://wepowder.com/en/forecast#snow-cumulative:144
I'd always divide by 2 being a half empty type.
GFS 06z has a tiny pool of colder air floating around on thursday....would be nice to see snow down to 1500-1600m locally, think the earliest I've ever seen it that low was on Sep 26th a few years ago, so an outside chance of smashing that record for much earlier snow.
|
|
|
|
|
|
You know it makes sense.
|
|
Thanks for that @polo. Any snow at all will be a good start!
|
|
|
|
|
|
Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
|
|
|
|
Poster: A snowHead
|
|
|
|
Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
|
|
|
|
Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
|
|
|
|
You need to Login to know who's really who.
You need to Login to know who's really who.
|
Storm Amy as it's now called will mostly pass just to the north of Scotland and then east into the north sea, but there are some unusual differences in the track at just t+48. Nearly all of the models suggest very little impact for the alps, maybe 5-10cm above 2000m, depending on whether most of the ppn arrives early or a few hours later as the temps drop over the weekend.
It's a very brief dent in the high pressure, so another short window of oppportunity before quickly going back to dry and mild all next week.
The most interesting difference across the models for me is that the most accurate one, AIFS, is stubbornly projecting the low to be a little further west. It's been an outlier for several days that you'd normally expect to have aligned by now. The slightly lower pressure over the west and northern alps mean it's showing temps 2-3 degrees colder than ECM/GFS/UKMO etc and with a longer spell of snow.
It's been performing really well this year, often leading the way as others catch on....so this will be a good test to see if it really is on a different level.
Looking at temps on Sunday, here's UKMO vs AIFS. Even allowing for slight time stamp differences, the AI model is still about 3 degrees colder.
And if AIFS has got this right the snowline will obviously be lower than others are currently showing, with maybe 25cm in the east. here's what it shows out to tuesday.
WePowder (GFS/ECM) for comparison
|
|
|
|
|
|
Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
|
|
|
|
You'll need to Register first of course.
You'll need to Register first of course.
|
@8611, has there ever been a Nordstau that wasn't "temporary"? Assume a translation issue, probably meant brief, as it's not much of a traffic jam....but yeah Austria getting the first decent snowfalls as often happens, maybe more to come towards mid month.
AI model once again leading the way.
|
|
|
|
|
|
|
|
| polo wrote: |
| AI model once again leading the way. |
Do you know in what way the AI model differs from the more established models like GFs, etc?
|
|
|
|
|
|
|
|
@rob@rar, yeah I know the basics, but given recent out performance am keen to dig deeper.
Traditional models rely on constant real world data collection, from ships, planes, weather stations etc, and every 6 hours or so they run this live data through their physics based models, using all the scientific theory (eg thermal dynamics) they've coded to produce a main Operational forecast, and then run all the ensembles / iterations based on slight tweaks to both data and feedback loops and so on.
AI models don't use any number crunching as such, they have just learnt how previous patterns evolved from the same starting data. No theoretical physics equations, just re-analysis of the past, typically the last 30-40 years.
The big difference is really about speed and efficiency, they are far less resource intensive, yet surprisingly accurate given how seemingly different the climate is today compared to previous decades. I guess you could infer that even if some areas are now warmer, drier etc, the way the patterns themselves evolve doesn't change as much as we might expect. The curious part is more to do with feedback loops, eg the arctic warming has downstream consequences, or the expanding Azores high etc, which a lot of experts use as a reason to dismiss past analogs.
|
|
|
|
|
|
You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
|
|
@polo, thanks for that, much obliged.
|
|
|
|
|
|
|
|
|
@polo, interesting as ever thanks! Any recommended links for AI forecast websites at all??
|
|
|
|
|
|
snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
|
|
|
|
And love to help out and answer questions and of course, read each other's snow reports.
And love to help out and answer questions and of course, read each other's snow reports.
|
How incredibly dull. This dirty high still has another week to go, very little sunshine, just cloudy every day. The massive Omega shaped block over France / UK is one of my least favourite patterns, and often takes weeks to shift. At least it's bringing colder temps to the eastern end this week.
Today
A change is on the cards, but again not ideal. While AO and NAO are heading negative for the rest of the month, another least favourite pattern is emerging with a west based -NAO, meaning high pressure over greenland-iceland and low pressure in the atlantic, but much too far west to benefit the alps. Further out a few models such as GFS send some low pressure towards the alps early next week (20-21), which would lower temps a bit towards a 2000m FL.
But the better models like ECM and AIFS show a flatter pattern for now (yellow shading = high pressure remaining in charge over alps), so would be a FL above 2500m, maybe 2800m.
 
Looking at Oct week 4, there are signs things are moving into a more favourable pattern with a scandi low to compliment the greenland high, potential for cooler NW-SE flow.
More hopecast than forecast at that range. At least it's going to rain again (AIFS ens for NW).
 
|
|
|
|
|
|
|
|
|
@polo, please keep the forecasts coming to suggest there is a light at the end of our tunnel of dullness
|
|
|
|
|
|
You know it makes sense.
|
I think this will be the fifth day of wall to wall sunshine in the north of the Uk. Check out the Glencoe Yad moss and Lakeland webcams and photos. You just need to go upwards a short distance to get into the sun 
https://www.winterhighland.info/cams/glencoe/
|
|
|
|
|
|
Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
|
|
@Nodeamus, it’s definitely going to change, we might miss the calm and dry spell if we get 10 days of rain….but yeah it’s pretty rare here in France to not see sunshine for so long, my solar batteries need plugging in. All starts Monday…..no consensus on FL, all over the place
|
|
|
|
|
|
Poster: A snowHead
|
|
@Peter S, I flew back from Bergammo into Belfast on Sunday evening and could see the Lake District peaks above the cloud, also the mountains of Arran and the Paps of Jura. Incredible!
|
|
|
|
|
|
Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
|
|
We even have some sunshine here at ground level in the Lakes today. Last few days its been necessary to climb to around 700m+ to poke your head out of the clouds.
|
|
|
|
|
|
Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
|
|
|
|
You need to Login to know who's really who.
You need to Login to know who's really who.
|
Suppose it's good to get these least favourite patterns out of the way before the season starts, but at the same time not something we want to see become a regular feature again.
Large amount of ppn showing up on the ensembles next week with initial FL around 2400m or lower, but probably rising to 2800m by wed-thurs, with a lot of switching airmasses, ie sometimes colder on the north, then the south, back to the north etc.
The west based -NAO, highs over greenland and lows over azores looks like persisting into fri-sat (25th). Still plenty of room for changes of course but quite a few runs showing temps of 20c (at 1500m) over the med. But current modelling also showing the pattern eventually moving east, which would make a huge difference to temps into month end. That final week pattern of greenland high becoming more of a mid atlantic ridge up to Iceland, with scandi - north sea low would be heavy snow to low altitutdes if it comes off. FI for now.
Here's the little trough-that-shall-remain-nameless off Portugal end of next week (only for those that are age verified as it is not suitable viewing for all)
|
|
|
|
|
|
Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
|
No change out to tuesday, FL rises a little 22-23rd, but then you see some interesting scenarios. AIFS again really different to most of the other models, bringing in the colder air pretty quickly from 23-24th on, while GFS, ECM, GEM, JMA etc have the low diving too far west initiallly, meaning some delay to the cold or some runs even continuing with the high snowline.
Most models do have a colder shot of air from the NW, but AIFS is by far the snowiest and most sustained outlook on recent runs.
Selection of the better anomaly charts for 21st, 24th (AIFS), 27th showing the potential.
     
|
|
|
|
|
|
You'll need to Register first of course.
You'll need to Register first of course.
|
|
oh well not this time, AIFS backpedaling quickly to match other models with brief NW flow into next weekend and the snowline likely staying quite high....typical for time of year
|
|
|
|
|
|
|
|
|
ECMWF forecasting cold temperatures day and night from next weekend. Below freezing or just above at village level for example at Arosa, Obergurgl and Zurs. A great help to snowmaking, already underway.
|
|
|
|
|
|
|
|
|
@Snow&skifan, they're making snow already? Wow.
|
|
|
|
|
|
You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
|
|
|
|
|
|
|
@Snow&skifan, surely just testing the guns vs actively making snow?
|
|
|
|
|
|
|
|
FAQ
