Long range weather forecasts

[The Hat]

So who do you trust?

For London ..... On Sat the wind direction and speed are supposed to be....

Weather.com SW 14mph
BBC Weather S 17mph
Met office S 23mph

They agree that the wind will be coming from a S/SW direction but to have such differences in the wind speed surprises me.

[dublinpilot]

Those differences shouldn't surprise you. The wind speed will vary by more than that amount during any given day.

Just take that the the wind speed is forecase to be 15-25mph, more or less ;)

dp

[airborne_artist]

Forecasting the wind speed four days out is almost impossible - in fact, forecasting almost anything four days out is hard work given that the UK gets most of its weather from the easterly movement of Atlantic depressions.

[Fuji Abound]

None of them and all of them!

If I were you (and forgive me if you have loads of met expereince) I would not pay too much attention to general weather forecasts. Inevitably the problem is they are forecasting for a lengthy period of time (the whole day) and unless the scenerio for the day is very stable, it will change.

I dont know how the Met office arrive at the wind strength in their general forecasts but I would presume it is some sort of average of what they expect to happen throughout the forecast period.

Saturday is a very good example. If you go back to basics and look at the Met office model charts for the day it becomes clear that the high pressure that has been dominate the last few days is being forced south by an active low pressure which will be centred to the north east of Iceland by mid day. The associated front will cross the country during the afternoon and Saturday night depending where you are. Not surprisingly, the wind will increase in advance of the front and I suspect will become very gusty in frontal activity. You can expect a sgnificant change in the wind during the day depending where you are in the country.

[Edited to add that I would guess there is quite a bit of uncertainty about what will happen over the weekend. The low is sufficiently North and the High sufficiently dominant that it may remain relatively stable over the south of England for longer than predicted. Based purely on expereince and NOT on any scientific interpretation it seems to me at this time of year there has been a trend for the lows to track further north than they use to which seems to me to account for the better weather we have had during recent Autumns.

Interestingly in a discussion I had recently with a chap who has been flying for ever we got talking about the TAFs - I commented that they seemed in general less reliable than when I started flying. He was quick to agree and felt it symtomatic of the trends on which forecasts have for so long been reliably based havign chnage din the last four or five years. I suppose the soath sayers would say - ah yes, that be global warming, that be

Ah well, back to feeling the sea weed for me!]

[The Hat]

Don't the BBC work with the Met office. Wouldn't it make sense for them to have the same forecasts.... Sorry I got carried away there, that would be far too simple.

[EvilKitty]

The BBC does use the Met Office for its weather information. (http://www.bbc.co.uk/weather/metoffice.shtml)

However, the two things no one here has considered are the location and the time - just where is London, and what time is the forecast for? Also, at what time was the forecast generated?

On the BBC site you can enter the postcode you are interested in and will get different numbers (even though they may all be "in London"). Also the forecast time and type may differ - one may be a midday report, a mid afternoon report, worst weather report, average conditions report. lastly reports are constantly updated, so you also need to check the time at which the feed used was generated.

[BackPacker]

Saturday is a very good example. If you go back to basics and look at the Met office model charts for the day it becomes clear that the high pressure that has been dominate the last few days is being forced south by an active low pressure which will be centred to the north east of Iceland by mid day. The associated front will cross the country during the afternoon and Saturday night depending where you are. Not surprisingly, the wind will increase in advance of the front and I suspect will become very gusty in frontal activity. You can expect a sgnificant change in the wind during the day depending where you are in the country.

What I find is that the European weather forecasts are generally very good at predicting where the high and low pressure systems go, and what the fronts are going to do in between plus the associated weather. However, over a longer period, it is usually the timing that's off. So if a front passage is predicted, four days in advance, at noon someday, then the front will pass that day, but it may happen anywhere between midnight and midnight.

www.wetterzentrale.de/topkarten gives you access to the more or less raw data from the European forecasts centers. I find it very handy, if I need to do my own long-term forecast, to compare the different outcomes of the different runs of the different models, to see what the general pattern but more importantly the timing is.

[scooter boy]

Talking about long range weather..
I was disappointed to the point of almost sending an e-mail into [email protected] to complain about the fact that the video TV forecast has shrunk from 3-4 days to barely 24hrs!!

They are really hedging their bets now and not wanting to risk being wrong about the forecast.

Just a general idea, that's all I want, but oh-no we'll put a wet glass sphere up or a picture of some autumn leaves instead rather than risk predicting another 24hrs forecast.

SB

[Fuji Abound]

Scooter

I am missing something:

http://www.bbc.co.uk/weather/ukweather/#no_url

seems to run right through to Sunday when I posted this.

[airborne_artist]

For a different input, try http://www.metcheck.com/V40/UK/HOME/ - the location search box accepts ICAO codes.

Looking at the Met Office pressure chart for Saturday, I'd be surprised at much more than 15kts of wind in the London area until the evening at least.

How about a challenge? The closest to the EGLL actuals for 1300 local on Sat :ok:

[scooter boy]

Thanks Fuji,
Nice to see the long range data is available online, shame that they don't tag it on to the TV forecast though any more, I am pretty sure that pilots are not the only group looking at the weather with baited breath.
SB

[IO540]

Long range forecasting is a contradiction in terms :)

My favourite site is NOAA (http://www.arl.noaa.gov/ready/cmet.html).

Enter the ICAO code.

Choose a forecast dataset e.g. 0-184hrs.

Under fields to plot select "choose from below".

In the fields, select

- mean sea level pressure
- temperature 2m agl
- accumulated precip
- total cloud cover
- wind flags
- wind speed

(or whatever turns you on, but you can't mix the 2D and 3D plots, etc)

and type in the anti-ripping graphic security code at the bottom

It reckons the wind will be about 8-10kt S/SW.

[gfunc]

Hi folks,

As a weather geek who has been studying this weather stuff at uni for nearly 10 years (I'm nearly finished!) and running forecast models I feel duty bound to point out that todays 5-day forecast is as good as a 2-day forecast was 20 years ago. I'll omit the details about how good the 2-day forecasts were 20 years ago.......

Anyways, I though I would list a couple o sites that might be useful for you all:

http://www.meteo.psu.edu/~gadomski/ewalleur.html (http://www.meteo.psu.edu/%7Egadomski/ewalleur.html) Penn State E-wall for Europe. My personal rule of thumb - always trust the ECMWF.

http://www.meteox.co.uk (http://www.meteox.co.uk/) - European weather radar.

I'll try and dig some more out, the European met agencies aren't particularly forthcoming with free data nowadays.

Cheers,

Gareth.