Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

The Met Office - not fit for purpose?

Tech Log The very best in practical technical discussion on the web

The Met Office - not fit for purpose?

Old 4th Sep 2009, 10:29
  #101 (permalink)  
 
Join Date: Aug 2003
Location: Not a million miles from EGTF
Age: 63
Posts: 1,571
Metman

Couldn't agree more with your very detailed and (in my view) accurate summary.

When I was gliding, the weather forecast was a vital tool in deciding whether or not it was going to be worthwhile rigging or not. For that I used synoptics for the next day or so and would monitor the BBC when they did use them properly.

On the day itself, what I chose to do was what the weather would allow me to do and I would change the task to suit the conditions.

Now I am a hobby power-pilot I see that many of my fellow pilots have spent days planning a specific flight for lunch at a particular airfield or to meet a slot-time at the Sywell Rally, for example.

When the weather prevents this happening because of early fog not burning off, or a delay in the passage of a front by 4 hours, their day is wrecked and they go home disappointed fuming at the Met Office.

I would really like the BBC, or someone, to do a debrief of the weather to explain why the weather changed from that forecast. It would help in understanding the ebbs and flows of the weather (not climate) that we fly in.

But now because of the target-setting culture and KPIs no-one dare risk being in the wrong, and that is a shame.
robin is offline  
Old 4th Sep 2009, 11:05
  #102 (permalink)  
 
Join Date: Aug 2001
Location: se england
Posts: 1,126
Robin

i think the last line of your post hits the nail squalrely in the head.
What place do targets and KPIs have in the met office.Essentially NONE just because it is an INexact science. Just as it is in certain arts of the NHS, Education or the Police. (Mind you after the banking fiasco perhaps they shouldnt be used anywhere.

If one thing more than any other has crippled the UK in the last 20 years it is management techniques being pushed into organsiations who do not have profit as a goal (or shouldnt).

The met office has a simple obvious objective which is to try and produce an accurate foreacst of the weather, Everyone knows its hard to be right all the time but when it s someones objective to be right X% of the time human nature, greed, spin (or lies as its the same thing) all start to play a part and corrupt the process.

Its not an entirely New Labout thing -plenty of Thatcherites pushed the 'accountability' li(n)e in the public sector and I certainly do not rembember them as less decitful or dishonest than todays lot -but NL certainly share the blame for pushing some issues even further and failing to correct errors from the past.

Rant over, the suns come back out
pax britanica is offline  
Old 4th Sep 2009, 23:49
  #103 (permalink)  
 
Join Date: Jun 2001
Location: somewhere in Western Canada
Posts: 199
Metman, that is absoluely spot on! (and so is Robin's last sentence)

I have to add that it seems to be the same over here (Canada).
CaptW5 is offline  
Old 5th Sep 2009, 09:56
  #104 (permalink)  
 
Join Date: Aug 2003
Location: Not a million miles from EGTF
Age: 63
Posts: 1,571
I'm sitting at the computer looking out the window at solid low cloud and heavy drizzle

The TAF for an airfield less than 3 miles away is

050803Z 0509/0518 18005KT 9999 SCT035 BECMG 0509/0511 27010KT

The give-away is the wind direction which is bound to bring damp conditions but that is local knowledge.

On no forecast I have seen is there a sign of this rain/drizzle having been forecast. So anyone flight planning into the local airstrip will find conditions rather less to their taste than they expect.
robin is offline  
Old 5th Sep 2009, 11:51
  #105 (permalink)  
 
Join Date: Jun 2002
Location: Germany
Posts: 25
Here we go again

I'm sitting at the computer looking out the window at solid low cloud and heavy drizzle

The TAF for an airfield less than 3 miles away is

050803Z 0509/0518 18005KT 9999 SCT035 BECMG 0509/0511 27010KT
Robin, if you don't say, at what time you looked out of the window at which precise location and for what airfield the TAF is valid for, your statement is coming without the slightest amount of substance behind it.

Judging from the handwriting I'd say your TAF is from Exeter, so I've taken the liberty to add some Exeter METARs so your audience on this thread have a chance to make up their own opinion. (source: the wonderful ogimet.com)

SA 05/09/2009 10:50-> METAR EGTE 051050Z 28012KT 9999 VCSH FEW012 SCT015 BKN030 15/12 Q1024=

SA 05/09/2009 10:20-> METAR EGTE 051020Z 22003KT 9999 FEW012 BKN015 15/12 Q1024=

SA 05/09/2009 09:50-> METAR EGTE 050950Z 23006KT 9999 4000E -RADZ SCT010 BKN015 14/12 Q1023=

SA 05/09/2009 09:20-> METAR EGTE 050920Z 22006KT 9999 VCSH BKN025 14/11 Q1023=

SA 05/09/2009 08:50-> METAR EGTE 050850Z 22004KT 9999 VCSH BKN025 14/11 Q1023=

SA 05/09/2009 08:20-> METAR EGTE 050820Z 20005KT 160V240 9999 SCT030 14/11 Q1023=

SA 05/09/2009 07:50-> METAR EGTE 050750Z 18005KT 9999 FEW030 12/11 Q1023=

There was indeed a brief spell of drizzle which brushed parts of the airfield, but checking the Exeter webcam now, visibility seems just fine to me. If they had planted a TEMPO group in the TAF to get it right for those who can't be satisfied, they would have gotten flak from the other direction. This all is so ridiculous !

And please note how nicely the windgroup in the TAF worked out. Thus far. For those who need to keep their fishing rod straight.


Please allow me to remind all you TAF slayers out there :
Civilian TAFs are primarily designed for IFR guidance and nothing else. With regard to VFR flying the TAF contains only a YES/NO information for the type of airspace the TAF is intended to be used in.

There are threshold values of weather parameters which, if crossed to the better or worse, make an amendment or a subdivision in a TAF mandatory. These are defined by ICAO in Annex 3 and in some countries augmented by national legislation with regard to national airspace rules.

TAFs may be additional guidance for a VFR flight and are frequently (ab)used as such, but TAFs alone are by far not sufficient to get an accurate impression of weather conditions on an intended VFR flight. TAFs are not meant to be.

They are meant to be a concise planning aid for commercial aviation.

(edited for spelling)

Last edited by SIGMET nil; 5th Sep 2009 at 12:05.
SIGMET nil is offline  
Old 5th Sep 2009, 14:39
  #106 (permalink)  
 
Join Date: Jun 2002
Location: Germany
Posts: 25
Drizzle showers

With regard to Robin's TAF I just had a look at the nearby Camborne vertical sounding of today (sep 05 2009) 12 utc.

It shows an unstable moist layer below 5000 ft amsl. For a normal rain shower, the moist unstable layer where dew point and temperature are equal would need to reach up at least to the -10 isotherm. To create snow, which can melt to rain.However, this, our moist unstable layer is entirely in the positive temperature region.

This is the typical setting for drizzle showers, especially with some warm advection going on at 5000 ft amsl, as today. The subtle lifting caused by this warm advection is right where its needed, level with the moist unstable layer. Also there is a little thermal lifting going on over land. Liquid drizzle droplets form in the moderately low cloud layer. Low level warm advection and low level clouds are the main ingredients. Due to the unstable stratification (cloud layer follows moist adiabates) the drizzle falls out like little showers.

At least that's my view of it. If somebody has a better one, please let me know. My textbooks and my instructors said, drizzle showers don't exist, but according to all my experience - even deep inland - they are a regular occurence.
SIGMET nil is offline  
Old 5th Sep 2009, 17:28
  #107 (permalink)  
 
Join Date: Apr 2009
Location: now in Zomerset
Age: 57
Posts: 124
As a relatively low time pilot I was interested in the comment that TAFs are meant for IFR traffic and not for VFR.

My instructors tell me always to use the Tafs along the route.

Metars should only be used with caution as they are only a snapshot.

But looking at Robins post and assuming it was Exeter it does look like the TAF underplayed the actuals and the later TAF showed a more realistic forecast matching the Metars.

An innocent question though. If VFR traffic is not being served by the TAFs then where do we get our information.
peter272 is offline  
Old 5th Sep 2009, 21:39
  #108 (permalink)  
 
Join Date: Jun 2002
Location: Germany
Posts: 25
As a relatively low time pilot I was interested in the comment that TAFs are meant for IFR traffic and not for VFR.

My instructors tell me always to use the Tafs along the route.
Thanks, Peter, for your interest.

Using TAFs along the route is ok. You will catch a lot of the more significant weather. You just might not find out with them if visibility is 5 or 50 miles or ceiling 1600 or 4000 ft amsl and other likewise subtleties. Anything above 8000 m (in Germany 5000 m - national differences !) and 1500 ft agl is not really of interest in a TAF.

Low clouds (with or without drizzle) tend to linger preferrably on the upwind side of hillridges. So the same clouds the TAF discarded as irrelevant for flat airport terrain can be a much bigger nuisance on cross country VFR flights.

Like Robin already indicated, the low level wind can be a great guidance to identify trouble zones, which will be normally the upwind sides of geographical obstacles, where you should put particular emphasis on getting accurate reports or other information (e.g. trustworthy webcams). Same goes for coastal areas - you need to know, if the wind will blow from the sea or from land.

So like Metman I would recommend to get familiar with weather charts and do some of your own forecast. Use various sources to catch errors.

The UK Met office surface charts might be a good starting point. Then you get a wealth of US model output also for the UK. The GFS model on which this is based, is quite reliable in Europe, but like everything it errs occasionally and you have to be alert to catch these events. A wonderful website - but in German - for such stuff is Wetterzentrale. In the submenu fax/Bracknell you also find the MET Office surface charts. The GFS (/M-Europa) submenu also offers a lot. There are interfaces in English for this site in existence by third parties.

Drizzle you normally can't see on radar, but else RADAR is a wonderful means of getting information, what goes on between METARs and other reports.
Your remark about METARs being snapshots can't be emphasized highly enough in convective weather.

Try to get satellite pictures. There are wonderful resources available, such as Uni Bern (when working) or Dundee Receiving Station.

To have a look in vertical soundings is highly advisable to get familiar with the airmass that is just moving into your area. Select continent, time and the diagram type of your choice.

All these sources are good for trip planning and creating a mental picture of the weather situation. They are not intended for operational flight planning, as they may be unreliable and false. However, knowing a little more is never wrong.

Finally I just had a look at the offerings of the subscription service from the UK MET office. Maybe you can have a closer look at it with somebody who has subscribed. The 3 day text outlooks and all the satellite and radar stuff are looking nice.

In Germany, a flight weather subscription service has been a great success since its introduction a long time ago, because it gave pilots a huge degree of independence in choosing their preferred weather resources.

Studying high resolution radar and high quality satellite pictures frequently with the surface analysis in mind will give you a lot of "local experience" countrywide in relatively short time. But that is only my opinion.
SIGMET nil is offline  
Old 5th Sep 2009, 22:23
  #109 (permalink)  
Dit
 
Join Date: Apr 2006
Location: Europe
Posts: 108
Personally I believe the Met Office don't do a bad job of forecasting, but you have to put some work in to ascertain a more relevant forecast, due to the almost anti-liabalist TAFs sometimes produced.

Firstly, I look at the radar pictures and wind direction actuals and forecasts to get the general wind direction and speed. Then cross check this with the METARs and TAFs for airports upwind of your destination and alternate/s. Obviously terrain and local factors affect how the weather progresses across your chosen overview, and this has to be learnt by gaining local knowledge through experience.

Takes a little time, effort and experience though.
Dit is offline  
Old 6th Sep 2009, 07:49
  #110 (permalink)  
Thread Starter
 
Join Date: Dec 2006
Location: UK
Posts: 59
BBC forecast last night said Northern Ireland would have a sunny morning - front not further North than Dublin by late morning, expecting it early evening in Glasgow (7-8pm) ........ 0745 GMT rain in Belfast - we reckon the weather will be in Glasgow by lunch time ! Let's see what happens.
Just wondering is offline  
Old 6th Sep 2009, 09:53
  #111 (permalink)  
 
Join Date: Sep 2007
Location: devon
Age: 59
Posts: 9
Its all very well copying a list of actual weather reports for a station to try to justify the forecast(TAF). What one needs to remember is that significant changes in the current weather conditions i.e special reports (if they are being done as required!) are not broadcast outside of the aerodrome and on bulletins. The fact that 4000m was experienced on one METAR does not mean that these conditions didnt materialize frequently or infrequently between routine observations!
fireballxl5 is offline  
Old 6th Sep 2009, 13:33
  #112 (permalink)  
 
Join Date: Jun 2002
Location: Germany
Posts: 25
The fact that 4000m was experienced on one METAR does not mean that these conditions didnt materialize frequently or infrequently between routine observations!
Absolutely correct.

For the purpose of this thread I needed to accept this shortcoming. A lot of significant weather is hidden in SPECIs, which never leave the airport. METARs are only helpful if seen in context with the general weather situation.

For this reason I always check a lot of reports from the area in question, check a couple webcams I trust, have a look on RADAR, have a look on vertical soundings, maybe near-infrared satellite imagery or visible imagery... This way I try to get a feeling how frequent especially the minor nuisances may occur. Hundert percent certainty will never arrive.

I wanted to illustrate, that it is impossible to write the "perfect" TAF. Wether you perceive a TAF as helpful or not depends to some extent on your intentions and other personal variables.

And some TAFs just turn out to be not helpful at all. I'm honestly sorry about that.
SIGMET nil is offline  
Old 6th Sep 2009, 16:15
  #113 (permalink)  
 
Join Date: Sep 2007
Location: devon
Age: 59
Posts: 9
Could not agree more! For all the information at the disposal of the forecaster, the accuracy of the TAF depends first and foremost on an accurate METAR. Crap METAR in = Crap TAF out!! Simple really!Come on you guys!
fireballxl5 is offline  
Old 9th Sep 2009, 09:13
  #114 (permalink)  
 
Join Date: Nov 2003
Location: in the south
Posts: 80
Ok so we can talk all day about TAFs & Metars specifics but the fact remains the met office is getting it basically wrong too often.
eg;
As a broad picture yesterday they indicated a fine day in the south east for today. This morning, a whole 18 hrs later they now have rain showers spreading across the region later today !

No wonder most people ( general public ) seem to think the metoffice is doing a poor job compared to a few years ago.

nuf said
fivegreenlight is offline  
Old 9th Sep 2009, 09:27
  #115 (permalink)  
 
Join Date: Dec 1999
Location: Berks, UK
Posts: 51
where did the broad picture forecast come from? Not commenting on your claim - might be quite accurate, but where did you get your information?
Metman is offline  
Old 9th Sep 2009, 18:29
  #116 (permalink)  
 
Join Date: Nov 2003
Location: in the south
Posts: 80
From the Metoffice !

Met Office: London & South East England: forecast weather

Of course they dont allow you to see what they were forecasting!
fivegreenlight is offline  
Old 9th Sep 2009, 19:14
  #117 (permalink)  
 
Join Date: Dec 1999
Location: Berks, UK
Posts: 51
You are of course right! They don't display the old forecasts, and its a pain in the arse!

I personally don't trust the web forecasts - either Met Office or BBC. The text invariably contradicts the pictures, and the overview pictures invariably contradict any animations! Oh, and the BBC site is often different to the Met Office site, despite them both having the same source data. I don't know if there is any forecaster intervention into what goes on the website, or if it goes straight from the computer (which is my suspicion). I also think they're just too general. I admit that I tend to use the BBC weather site out of sheer laziness, and I've not been impressed with the accuracy for a long time.

One suggestion would be to try the new area of the website:

Met Office : Invent - New Weather Visualisation

Not operational as such - still under trial, but certainly allows for more information to be displayed in hopefully a much more user friendly format.

However as I said before, the pressure charts do still seem to be accurate I think, which is why I've never really understood why other things can be so variable!

I do also think there are potential improvements on the way, but watch this space... I'm not convinced they have ever really completely got over the move from up here to the south west yet.

BUT if you are in a position to talk to a forecaster or read your own charts, I'd go for that any day
Metman is offline  
Old 9th Sep 2009, 19:37
  #118 (permalink)  
 
Join Date: Nov 2003
Location: in the south
Posts: 80
thanks for the link.
I had a suspicion that things deteriorated after the move southwest !
fivegreenlight is offline  
Old 11th Sep 2009, 08:39
  #119 (permalink)  
 
Join Date: Aug 2007
Location: Heartfordshire
Age: 62
Posts: 4
Forecasting is impossible

Just in case anyone is interested, weather forecasting by computer modelling is impossible. The reason for this is that the equations which describe how atmospheric components interact are “Chaos” or “Sensitive Dependence on initial conditions” equations.

Very roughly, the way the modelling works is that you plug all your current weather data in to the Equations, do the sums and the answer is a new set of weather data. This data is used as input to the next iteration. This is repeated until you have rolled forward to the required point in time.

Implications of Chaos

The behaviour of Chaos equations varies depending on the input data. For some input values the equations will be stable. That means that if you run it multiple times with slightly different starting data, the results will be roughly the same. Running the equations with a different set of data will result in answers which significantly diverge.
The only way round this is for the input data to be 100% accurate for an infinite number of points on the Earth’s surface. This is impossible.

This means that if you run the model several times with different data and the answers are similar, you can be confident of you forecast. If the results differ, all you know is that you don’t know.

In conclusion

If you look synoptic charts in the run up to the period you want to forecast and they are dominated by a single weather system, you can believe the forecast. If there are lots of different systems interacting, then you may as well revert to seaweed.

Regards

Techno Freak
TechnoFreak is offline  
Old 11th Sep 2009, 16:54
  #120 (permalink)  
 
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,510
"BBC forecast last night said Northern Ireland would have a sunny morning - front not further North than Dublin by late morning, expecting it early evening in Glasgow (7-8pm) "

Surely you are not saying, are you, that jo-public gets notice about the position and movement and speed of fronts, but that has all been removed from the sig wx. chart of us professionals. Regarding the confidence one could have in forecasts, seeing the position of fronts, and their likely direction of travel, gave me much relevant info, especially when on long-haul. However, given that a European average 4 sector day is landing back 11 hours after you've started, it is still relvant info, especially if TS/CB's are forecast. The question about their removal was asked months ago, but I never saw a definitive answer about it. Perhaps Metman & Techno Freak can help.
RAT 5 is offline  

Thread Tools
Search this Thread

Contact Us Archive Advertising Cookie Policy Privacy Statement Terms of Service

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.