Ski Club 2.0 Home
Snow Reports
FAQFAQ

Mail for help.Help!!

Log in to snowHeads to make it MUCH better!
Username:-
 Password:
Remember me:
durr, I forgot...
Or Register
(to be a proper snow-head, all official-like!)

Beginners guide to forecasting, GFS models?

 Poster: A snowHead
Poster: A snowHead
Seems there are some experienced weather peeps on here who understand the various models, so could anybody share some basic steps, ie where to go for some easy to understand graphs like the ones Brian occasionally posts, and how to interpret them?

Cheers,

Greg
snow conditions     
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
I'll chip in with a few sites I check, although my disclaimer is that I have little idea what I'm looking at so anything I suggest should be taken with a pinch of salt Wink

GFS data from the current run can be found here by including the relevant coordinates (eg Tignes is 45.49 lat, 6.92 long), selecting the GFS 0-180hr metogram, then typing in the password shown on the page. Key info this gives is precipitation and 850mb temperature (ie, temp at approximately 1500-1600m).

Data for succesive GFS forecasts can be found here. Select "Diagramme" in the lefthand menu, then select "Europa" in the middle menu, then select "Genf" (Geneva in German) and "Ensemble t850 und NDS" from the drop-down menus. This gives the same information as above (precipitation and 850mb temperature) for Geneva, but includes the last dozen GFS forecasts and an annual average. The average data is shown by the red line, the current forecast by the blue line, and the average of the last dozen forecasts by the white line. When all the different forecasts are in close alignment you can infer that there is some certainty in the GFS forecast data, but if the lines are all over the place it is difficult to say with any confidence what the weather will do.

This chart from Meteo France is a good summary of this season's snow depths at La Plagne.

This link is a precipitation radar covering most of Switzerland and some French and Italian resorts. No use for forecasting, but fun for nowcasting!

This site gives forecasts for Switzerland, but I think is based mostly on GFS data. This site does the same, but covers Europe without a focus on a particulr country.

Chamonix Meteo seems to be an actual "weatherman" rather than a simple mathematical model from the GFS data.

I'm sure other people who know what they're talking about will be able to add more useful sites.
snow report     
 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?
Good start Rob. I'm also no expert - but you can discern a lot from the 850mb charts at www.wetterzentrale.de. Click on "GFS ENS 12Z" at the top (ish) of the page, then on "GFS" in top left box, then on the "850 hpa temperatur" link and use the links to scroll forwards in time. As Rob says, the data used to produce the coloured charts comes from the thick green line on the ensembles run he described.

A bit more info on the ensembles - Rob wasn't entirely accurate about the runs being the last dozen or so forecasts - it's not quite like that.

Ensembles:
All the major numerical models (GFS, ECMWF, UKMO etc) run more than once each time they are issued. The reason for this is that although they each get fed a lot of data into them for each run, the data can never be considered perfect. To counteract this the model is run again with slight changes to the original data. Using the GFS as an example, the model is run 10 more times, 5 times with the data positively changed and 5 times with the data negatively changed.

By following the ensembles you can get an idea of the likelihood of the forecast given by the main (control) run of the model being correct - if a large number of the ensemble runs are very similar to it, it is a good indication that confidence is reasonably high of the outcome. If very few of the ensemble runs show that outcome, then it is quite probable that it may well change.

The ensembles are done four times a day and are known as "00z", "06z", "12z" and "18z". They are generally available on about a 6 hour delay, so you can see the 00z until around midday, when it will change to the 06z and so on... The data input into the forecasts changes on each run as follows:

00z - Weather buoy, satellite data, shipping data, country data, NOAA data
06Z - Weather buoy, satellite data, shipping data
12Z - Shipping data, Satellite data ONLY
18Z - Weather buoy, satellite data, shipping data, country data, NOAA data

Basically you want to see low temps in conjunction with precipitation in order to achieve the best snow possibilities!


Last edited by Well, the person's real but it's just a made up name, see? on Mon 25-02-08 18:11; edited 1 time in total
snow report     
 You need to Login to know who's really who.
You need to Login to know who's really who.
carled, many thanks for that correction, and the helpful explanation of the different forecast runs Smile
snow report     
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
Talking of gfs models... the current 00z run for Geneva is an absolute belter! Look at that temp drop about to hit and the precipitation lump right underneath! That could be a big dump of snow... snowHead snowHead snowHead
ski holidays     
 You'll need to Register first of course.
You'll need to Register first of course.
Sweet! Good timing for me next week Smile
snow report     
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
just hope the forecast rise in temp doesn't happen the next week!!

oh, and how do you get to that chart?? I followed your links above to http://www.wetterzentrale.de. but I get a map of Europe with temperature colours on it by following your instructions.

regards,

Greg
ski holidays     
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
gregh, follow Rob's instruction starting from "Data for succesive GFS forecasts can be found..."
ski holidays     
 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.
Quote:

just hope the forecast rise in temp doesn't happen the next week!!


To be honest Greg, it's not too bad. The lines are well scattered meaning there's little agreement at present, but there's a fair spread under the line. I'd settle for heavy snowfall in the next few days then gradually warming temps from when I arrive on Sunday! Especially bearing in mind we're fast approaching mid-April!
latest report     
 Ski the Net with snowHeads
Ski the Net with snowHeads
Excellent, thanks alot!

How quickly does the 6hr update appear, ie I guess they did a run at 6am, but the graph still shows the 00 run????

regards,

Greg
snow report     
 snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
gregh, read my post again - third in this thread - the answer to your question lies within... Very Happy
snow report     
 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.
Doh, thanks, missed that one!!!
ski holidays     
 So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
carled, good man. I must've done that explanation about half a dozen times on here Laughing

Edit: or good woman ? carled isn't gender specific and I don't want to stereotype weather geekery as a male preserve wink
latest report     
 You know it makes sense.
You know it makes sense.
any chance a moderator can pin this topic to the top as a sticky or somesuch?? Also it appears the link to the graph carled posted means the image auto updates Very Happy
ski holidays     
 Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
The 06z run shows a colder forecast for the first half of next week than the 00z run; hope this trend continues!
snow conditions     
 Poster: A snowHead
Poster: A snowHead
brian, right first time... definitely a man... "carled" is just a concatenation of first name and surname with "Carl" being the first part!

I'd say the latest update (6z) is showing most runs having pretty good agreement throughout the rest of this week, with only a couple of (hopefully) random runs appearing to (hopefully wrongly!) go for warmer temps... snowHead
snow conditions     
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
Brian - being more of an expert than many of us on here on the subject... have you any idea why the blue control run occasionally (a classic example above) goes for an out-lying run? I've been told that the further apart the median (white line) and control (blue line) are, the more inaccurate the forecast is likely to be. In the 6z run, you can clearly see that, starting on the 8th, then particularly around the 10th April, the control run suddenly skews right off into never-never land...

Surely they don't randomly choose which run becomes the control run, do they?
ski holidays     
 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?
carled, no, my understanding is what you said above, ie. that the control run is the input data unchanged and then the other ensemble runs have the data varied to reflect its known accuracy.

However, I have noticed that from time to time the operational will be a marked outlier and then sometimes the others will follow it on subsequent runs. Sometimes not though. Confused
latest report     
 You need to Login to know who's really who.
You need to Login to know who's really who.
So more questions on this mornings model if your not all sick of my questions Wink

If the current forecast is blue, it shows a steady rise in temps until 14th April, then a nice drop off until the chart runs out.

There is a huge disagreement in the models here, with one run showing -5 whilst the blue shows +11 and everything inbetween!

Do I read this as saying "the forecast really has no clue at this particular moment in time?"

Similiar question for the snow in the bottom of the graph, there appear occasional peaks ie 10th, 13th, 15th, 19th and 21st. Most are only one or two lines, can I read anything into this, ie "some" precipitation is likely but the forecasts aren't sure when and how much?

cheers,

greg
ski holidays     
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
gregh wrote:
Do I read this as saying "the forecast really has no clue at this particular moment in time?"
...
"some" precipitation is likely but the forecasts aren't sure when and how much?


Yes and yes, but please note the disclaimer in my first post Wink
snow report     
 You'll need to Register first of course.
You'll need to Register first of course.
looking ok for next week on the models, no silly temps, and maybe a chance of some fresh snow!

5 sleeps to go!

Greg
ski holidays     
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
should I (and the EoSB folk) get excited about that huge spike on the 22nd, or is it just noise seeing as it's only one colour, and the other "runs" don't agree?

cheers,

greg
ski holidays     
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
gregh, no cause for excitement. If other members start to pick it up on the next run then you can start to take an interest.
snow conditions     
 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.
doh, cheers Brian! How many do you need to get agreeing for it to be more likely?

Looking like a sunny week in Trois Vallees next week on the whole then, no freshies likely.......
ski holidays     
 Ski the Net with snowHeads
Ski the Net with snowHeads
bumping this up, can a nice moderator make this a sticky please???
ski holidays     
 snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
Meteo It is one of my big interests.
My english isn't good enough to explain this.
I will do it in Spanish:

-Tenemos dos líneas en los meteogramas; la superior és la línea de temperaturas a 850 Hpa.

Como podéis ver hay muchas líneas; cada una de ellas nos indica una de las posibilidades del mismo modelo (en este caso el GFS). Cuanto más juntas están las líneas mayor certeza tiene la predicción.
La línea gruesa de color rojo es la media de los últimos 30 años.
La línea gruesa azul es la salida más probable (la que se ve en los mapas)
La línea gris es la media de todas las salidas (la que se ve en los maps del modelo ENS.

-Las líneas de abajo es la precipitación. Sigue la misma dinámica que las temperaturas. La precipitación total en un lugar seria el area que delimita cada línea.
Decir que cuanto más lejos en el tiempo, mas improbable es que se cumpla la predicción.

Para calcular la cota de nieve:
- El nivel de 850Hpa está en unos 1450 metros de altura; aunque varía mucho en función de la presión. En los mapas del GFS de temperatura 850 Hpa; hay una líneas con unos numeritos como por ejemplo 140; por ejemplo mirad este mapa:

Estos lineas se llaman isohipsas. Nos dan la altura del nivel de los 850 Hpa. Es decir en el caso de que ponga 140; quiere decir que la temperatura que marca el mapa se encuentra a 1400m (hay que añadir 1 zero al número que marca el mapa).
Mediante esta información se puede calcular la cota de nieve partiendo que cada 100m se baja 0,6ºC; y que la nieve aparece a partir de 2-3ºC.

-También hay que tener en cuenta el nivel de 500Hpa, si hay una bolsa de aire frío a 500Hpa de por ejemplo -30 la cota de nieve será de 500m aunque a 850 Hpa haya 0ºC. Debido a que en el momento de precipitar el aire frío en altura se desploma hacia abajo rompiendo la estratificación de las capas de aire. Así pues puede empezar a nevar con temepraturas de +8 o +9ºC (yo lo he visto en mi ciudad el año pasado); y luego bajar la temperatura en picado (en 5 minutos) hasta los 0ºC.

I hope that you understand it
ski holidays     
 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.
I've made this thread sticky - you might collectively want to consider a page in the wiki, for details see here:

http://snowheads.com/ski-forum/viewtopic.php?t=18919
snow conditions     
 So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
Can someone proficient at Spanish and English try and translate metalhead132's effort Smile?
snow conditions     
 You know it makes sense.
You know it makes sense.
metalhead132 wrote:
Meteo It is one of my big interests.
My english isn't good enough to explain this.
I will do it in Spanish:

- We have two sets of lines in these forecast graphs, the top set is represents temperatures at 850 Hpa.


As you can see there are quite a few lines, each one tells us one possibility of the model (in this case GFS). The closer together they are, the more certain the prediction.
The thick red line is the 30-year average.
The thick blue line is the most probable model run (and is the one seen in maps)
The grey line is the average of all the model runs (and is the one seen on ENS model maps)

- The lower set of lines represents precipitation, following the same arrangement as the temperatures. The total precipitation at a given location is proportional to the area bounded by each line. One can say that, the further off in time, the less probable a prediction is.

To calculate snowfall:
- The 850mb pressure level is at about 1450 meters altitude, with significant variation as a function of pressure (highs and lows). On GFS model temperature maps there are numbered lines, say 140, like on this map:


The lines are called isohips. They give us the altitude of the 850mb pressure level. In the case of the marked 140, the temperature shown on the map is at 1400m (one adds a zero to the map number).

From this information one can calculate snowfall knowing that each 100m gain is a 0,6ºC temperature drop and that snow begins to appear at 2-3ºC.

- One has to keep in mind the 500mb pressure level, if there is a cold air pocket at 500mb of -30, say, snowfall might start at 500m although it is only 0ºC at 850mb. This is due to the fact that, at the onset of precipitation, cold air aloft descends and breaks the statification of the air layers. Thus it is possible for it to start snowing at +8 o +9ºC on the ground (I have seen this in my city in the past year); and then the temperature drops fast (in 5 minutes) towards 0ºC.



I hope that I have understood it.
latest report     
 Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
OK, great translation Comprex.
Thank you. Very Happy Very Happy Very Happy Very Happy
snow report     
 Poster: A snowHead
Poster: A snowHead
Heres and cut n paste of some info Brian and Carled posted on another thread..

Thought this might be handy for weather watchers ....

I've found that Wetterzentrale has updated ensemble charts for locations across Europe where latitude and longitude are whole numbers,

so http://85.214.49.20/wz/pics/MS_745_ens.png , is lat 45.00 degrees, long 7.00 degrees, which turns out to just to the NE of Sestriere. I've mapped a few other locations.

46 - 7 is on the Swiss French border above Argentiere
45 - 6 is on the edge of the Ecrins park a few km West of L2A
46 - 8 is on the Swiss Italian border near Saas Fee

57 - -5 is just across the Great Glen from Nevis above Loch Arkaig. If only Aonach Mor was 1000m higher (deep sigh) ...

All of these are "-ish" for locations, btw! the "whole degrees" thing makes it a large area being covered...

http://85.214.49.20/wz/pics/MS_1147_ens.png - Garmisch Partenkirchen/Seefeld/Ehrwald (self interest for Feb 2007)
http://85.214.49.20/wz/pics/MS_1047_ens.png - Lech/Zurs/St. Anton/Ischgl
http://85.214.49.20/wz/pics/MS_1347_ens.png - Kitzbuhel/Bad Gastein/Skiwelt
http://85.214.49.20/wz/pics/MS_1146_ens.png - Madonna di Campiglio
http://85.214.49.20/wz/pics/MS_846_ens.png - Saas Fe/Monte Rosa/Zermatt/Cervinia
http://85.214.49.20/wz/pics/MS_744_ens.png - Isola 2000/Auron
http://85.214.49.20/wz/pics/MS_746_ens.png - Morzine / Les Gets / Avoriaz
snow report     
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
A good site is USAFE (the US Air Force site). Although it is only every 24 hours (from 48 hours out) it gives precipitation rates - and whether it is solid/liquid. You can get a fair idea whether its going to be heavy/light/rain/snow, though of course mountain forecasting is never precise. My personal anecdotal experience is that it is as reliable, if not more reliable, as the Met Office for UK and Alpine forecasts.

http://ows.public.sembach.af.mil/wxcharts/wxcharts.htm

It does seem to suggest a dump at high altitudes at least at New Year, and I'd expect the snowline to fall to below 1500m as the cold front goes through.
latest report     
 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?
The same charts are available at Wetterzentrale: http://www.wetterzentrale.de/topkarten/fsfaxsem.html
latest report     
 You need to Login to know who's really who.
You need to Login to know who's really who.
do we know how accurate each of the individual predictions in the ensemble has been to date? Would be interestuing to see given the variety of thoughts each one seems to have...?
latest report     
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
Anyone know when the temps will drop and stay at the 'normal' winter level in the Alps going by long range forecasts Puzzled
latest report     
 You'll need to Register first of course.
You'll need to Register first of course.
Chris Brookes, AFAICS no prolonged spell of it in sight,...but that could change quite quickly.
latest report     
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
I thought it'd be useful to store this here too (NB 8E 46N):

skanky wrote:
You might find this useful, too:

http://85.214.49.20/wz/pics/MS_080460_g05.png

The reason behind the URL should be obvious from carled & brian's posts.


Edits: relevant resorts wink

Murren Kaprun LDA Courchevel 1850 Valdesqui

http://www.wetterzentrale.de/pics/MS_746_ens.png Megeve Les Gets Flaine
http://www.wetterzentrale.de/pics/MS_065460_g05.png Megeve Les Gets Flaine

http://www.wetterzentrale.de/pics/MS_746_ens.png Les Houches Chamonix Avoriaz
http://www.wetterzentrale.de/pics/MS_070460_g05.png Les Houches Chamonix Avoriaz

http://www.wetterzentrale.de/pics/MS_746_ens.png Les Coches La Plagne
http://www.wetterzentrale.de/pics/MS_646_ens.png Les Arcs
http://www.wetterzentrale.de/pics/MS_065455_g05.png Les Coches La Plagne Les Arcs

http://www.wetterzentrale.de/pics/MS_746_ens.png Tignes Val d'Isere
http://www.wetterzentrale.de/pics/MS_070455_g05.png Tignes Val d'Isere

http://www.wetterzentrale.de/pics/MS_746_ens.png Nendaz Verbier
http://www.wetterzentrale.de/pics/MS_075460_g05.png Nendaz Verbier

http://www.wetterzentrale.de/pics/MS_846_ens.png Zermatt Cervinia
http://www.wetterzentrale.de/pics/MS_075460_g05.png Zermatt Cervinia

http://www.wetterzentrale.de/pics/MS_1147_ens.png Hintertux
http://www.wetterzentrale.de/pics/MS_115470_g05.png Hintertux

http://www.wetterzentrale.de/pics/MS_1347_ens.png Obertauern Schladming
http://www.wetterzentrale.de/pics/MS_135470_g05.png Obertauern
http://www.wetterzentrale.de/pics/MS_135475_g05.png Schladming


Last edited by Then you can post your own questions or snow reports... on Mon 14-11-11 21:22; edited 9 times in total
ski holidays     
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
One thing to mention is that snow will generally experience the wet bulb or dewpoint (taupunkt) temperature, not the dry bulb temperature as it is effectively a wet surface. This *may* help reduce melting (depending on values, obviously) unless in strong, direct sunlight.
ski holidays     
 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.
some more nuggets of goodness from Brian and Carled

Can anyone tell me how I get the nice GFS graph for a specific location, namely Engelberg: Latitude : 46N Longitude : 8E , or from Wiki Coordinates 46°49′N 8°24′E

Just whack "846" in the url Greg, e.g. http://85.214.49.20/wz/pics/MS_846_ens.png

Like this ...

http://85.214.49.20/wz/pics/MS_846_ens.png

... but since it's almost dead on 46.5 degs North, you'd probably be better looking at ...

http://85.214.49.20/wz/pics/MS_847_ens.png

... to put yourself on the same side of the Bernese Oberland as Engelberg.
snow report     
 Ski the Net with snowHeads
Ski the Net with snowHeads
kitenski, haven't you already posted this info before? (see your post, 8 posts up tagged Tue Dec 05, 06 13:24) Puzzled
latest report     



Derived from phpBB © 2001, 2002 phpBB Group & 2004-2013 snowHeads.com Terms and conditions  Privacy Policy