Met not fit for purpose! [Archive]

Ken Wells

2nd Sep 2008, 15:48

Why is it that since the MetOffice moved from Bracknell, they are consistently wrong with the weather forecast?

I have now seen pilots getting the latest forecast and due to the bad track record ignoring it, with the comment " don't worry it is always wrong"

Credibility is now a safety issue.:D

A few weeks ago we planned a trip to France forecast was poor so we cancelled our trip. On the day we should have gone the weather was fantastic and the next day as well.:ugh:
As we had rearranged we missed out on a great weekends flying due to accepting the guess work of the Met office.:{

With all the sat photo and high tech kit, you think they could get it right some of the time. Maybe if they got some seaweed and a pine cone, or just looked out of the god damn window they might just get it right!!!

Weather Wizards please respond!:confused:

172driver

2nd Sep 2008, 18:23

You are getting your info from the wrong source. Try this (http://euro.wx.propilots.net/) and this (http://www.eurometeo.com/english/gfs-wz), that'll sort you out :ok:

PS: there are many more sites, but together with the embedded info in homebriefing.com these ones have - so far - never let me down.

S-Works

2nd Sep 2008, 18:39

Interestingly enough I represent GA on the Met Office Users Working Group and the quality of the forecast as I have been shown is surprisingly accurate and delivered to very tight SLA's. Any pilot is able to file a complaint if they believe that a forecast is inaccurate and are encouraged to do so especially if they think safety is compromised.

The computer model they use is much more accurate than it was just a couple of years ago and planned upgrades coming will nearly double the modelling power over current levels.

The weak link might therefore be considered to be elsewhere and perhaps a little bit of self diagnosis on interpretation might be called for.....

However if you are adamant that the forecast for a period was wrong than I am happy to have it investigated on your behalf and post the results of the inquiry. So please let me have the dates in question, your interpretation of the forecast and the actual and I will get the MetOffice to do the same.

Lister Noble

2nd Sep 2008, 18:45

If I need an ordinary non flying forecast for the day,I look at Norwich TAF and find it the most accurate of all the forecasts available!
Give your local aviation forecast a try for everyday use,it's usually the most reliable.
Lister:)

Fright Level

2nd Sep 2008, 18:54

Norwich TAF

Why is the Norwich weather on VHF Volmet always "corrected"??

PS Bosex, thanks for the Euroweather link. I use Euro Propilots a lot (despite the pop ups) and Meteox, but the Euroweather is a new one on me. It's Google Ad feed is uncanningly accurate as well ..

Ads by Google : Cut down 9 lbs of stomach fat every 11 days by obeying this 1 rule. FatLoss4Idiots.com

Just how did it know :}

IO540

2nd Sep 2008, 19:33

There are two main weather models we pilots get access to:

The UK Met Office model - they release the MSLP charts, F214/215, and a little bit of other stuff. The rest of the info - notably any 3D information - is commercially restricted because they sell it to commercial weather repackagers who in turn sell it to various businesses which depend on weather.

The Global Forecasting System (GFS) model - this is AFAIK a U.S. run thing and is free to all. Most of the "pilot weather" websites run on GFS data. You can get everything out of GFS, including 3D profiles (tephigrams, for cloud tops etc) and the model can be run up to 384 hours ahead (for those who want an "idea" beyond the UKMO MSLP chart dates but appreciate the 300+hr accuracy will not be great).

Neither model is consistently better than the other, but the UKMO model is sometimes claimed to be more accurate for southern England because it is claimed to be optimised for the SW airflow which dominates the region.

I think it helps to understand some of the shorthand.

A PROB30 TEMPO is generally an ar*se covering by the UKMO, and since ICAO permits only PROB30 or PROB40, a PROB30 has to go in for anything from PROB01 to PROB30 :) So, PROB30 very rarely materialises yet this results in so many upsets over unnecessary cancellations - often these are caused by flying schools banning operations if there is a PROB30 +TSRA. One needs to validate the actual likelihood of nasty weather by looking at radar (Meteox (http://www.meteox.com) covers the UK and bits of Europe) and Sferics (http://www.blitzortung.org), and sure enough the majority of the time the stuff just doesn't exist where you want to go.

A PROB40 means the forecaster actually thinks it might come, so this one is more serious.

I find TAF forecasts of low cloud (warm front kind of stuff) fairly reliable, although one can see a warm front with one look at the MSLP chart.

The weather which caught out a lot of pilots was last weekend in the south east when the TAFs really were way out, right up until very close and even then they were way off, on the optimistic side. Sometimes, the weather is not possible to forecast because the picture is very complicated and unstable.

Perhaps the simplest thing would be to revise the TAFs more frequently, but this might breach ICAO regs.

Another thing is to look at where the weather is moving from, and look at METARs at airports upwind. If an airport 50nm upwind is reporting OVC002 and the wind is really blowing, then you are likely going to get OVC002 as well, an hour later.

Like (smart) accountants say "cash is king, everything else is conjecture", METARs, radar, sferics, sat images are king, and everything else is conjecture. Never bet against actual weather.

The best thing any PPL can do is get the IMC Rating or an IR, and while the weather doesn't care about your paperwork (and will still ice up your plane if you fly in the wrong place, etc), you get a lot more options which enables a flight to be planned as fully IFR even if launched "VFR", knowing that you can fly an ILS somewhere, stop for tea, and review matters.

I gave up on VFR after about the 3rd cancelled PPL lesson - with a random-future-date cancellation rate of at least 75% it's all but useless for preplanned flights. Some people will disagree with the 75% figure but they are probably instrument capable pilots who are happy to fly through anything in UK Class G having launched "VFR" into OVC005 (as I am). If you want to be 100% VFR legal, a plain PPL is hopeless for anything preplanned.

robin

2nd Sep 2008, 20:51

Bose X

We appreciate your point, but the comments on the lack of accuracy and failures in timings are so common now we cannot accept the Met Office's views as to their accuracy

I spoke to a rep at Air Expo, as I often operate within 2 miles of the Met Office. The Exeter TAF is often wildly out - I now understand why people describe weather orecasting as astrology with numbers.

The Met Office person told me that the measures of accuracy are to do with precipitation and visibility. Everything else is outside their measures so no wonder they can claim to be accurate.

We are required to ensure we are able to fly safely and must check the weather before flight. As the CAA has a contract with the Met Office, we need to indicate we have checked their information, as the main supplier. Like others I check elsewhere as well

However, if I had taken the view of the Met Office on Sunday I would have stayed in bed, yet the day was good enough for a visit to another airfield.

Similarly, weather warnings appear on the Met Office website, vanish, then reappear, then vanish again.

Given that the USA have evacuated a major city for a Cat 4 hurricane that turned out be be a Cat 1, the Met Office is not alone.

I agree with an earlier post - I tend to ignore a Prob 30 and listen to a Prob 40. But what happens if the Prob 30 appears? If I had an incident, the MO would say they had forecast the event, even though we all know the probability is very low, and most times never appears.

THe other area of MO failure is in the timings. Too often this year and last, we have watched the forecast arrival of a significant event drift back over successive TAFs, and these may fizzle out to nothing or arrive 12-15 hours late - losing a whole day's flying.

No. The forecasts are weak and getting weaker. I have flown a lot this year by turning up on day with a poor forecast, but where my own local knowledge indicates the forecast is likely to be pessimistic. Using the MO information and overlaying it with my own knowledge and experience, I've done quite well.

The problem is that too many pilots take the MO forecast and (even worse) the BBC forecast, and have cancelled. On the other hand others have flown in obviously poor conditions and had serious problems.

Both cases show that we, as pilots, have little faith in the MO forecasts which now lack any credibility, no matter what their KPIs may show....:yuk:

S-Works

2nd Sep 2008, 21:02

Robin,

Please provide me with the exact dates and the nature of the problem and I will have them investigated.

But like I have said a lot of the problems are down to the interpretation.

Personally I have not had a problem with the accuracy of forecasts and flying over 400hrs a year you would think I would be a prime candidate.

I am not defending the MetOffice but I have yet to see direct evidence that they are not doing the job. So if you can provide it I will gladly follow up and post the results. After all if there is an issue with the quality of the information being provided then it needs to be addressed.

As a point, I flew on Sunday and the weather was as forecast and at no time did my interpretation lead me to believe that I should stay in bed.

Tony Hirst

3rd Sep 2008, 00:00

Another thing is to look at where the weather is moving from, and look at METARs at airports upwind. If an airport 50nm upwind is reporting OVC002 and the wind is really blowing, then you are likely going to get OVC002 as well, an hour later.That is excellent advice, it is the best way to get a handle on how the forcasts are panning out. If you use the Met Office's Global TAF/METAR Search function then this is a very quick and simple process that adds minutes to the flight planning.

Don't put weight on such a technique to validate radiation, advective or convective weather such as BR and TS/CB forcasts though.

For me, the often heard glib comments that "it is usually wrong" are unquantifiable, unverifiable and wrong in my experience.

Ken Wells

3rd Sep 2008, 07:40

Thanks Robin, glad to know I am not the only one.

I say again "the forecasts are getting worse, NOT BETTER!!"

I now check non MetOffice sites mainly Sat Photo sites for more acurate info.

Seems that most Gov bodies now just defend, rather than admit problems.

Ken Wells

3rd Sep 2008, 07:42

Thanks 172 driver, makes the point that there is better info outthere!

Doesn't address the problem with UK office though!

S-Works

3rd Sep 2008, 08:13

Alternative source
Thanks 172 driver, makes the point that there is better info outthere!

Doesn't address the problem with UK office though!

Like I said Ken, provide the evidence and I promise it will be investigated and the results posted. I have yet to see any of the detractors provide evidence of their claims and my own experience has been exactly the opposite.

I sit on the working group which is made up of the industry to keep the Met Office honest. I would be delighted to take any complaints to them. But the have to be real complaints backed up with real data not just the venting of a few people on forums who possibly might not be interpreting the charts correctly.

Mariner9

3rd Sep 2008, 08:39

Bosey, you're forgetting the current culture that everything has to be somebody else's fault these days.

Pragmatic, accurate responses are not what's called for :=

Fuji Abound

3rd Sep 2008, 10:10

I fly a bit each year, and I cant grumble at any of the forecasts in recent times.

For example, my last three trips to the west country have all involved some element of IFR, the cloud base was exactly where it was predicted to be and the approach into St Mawgs also revealed the runway pretty much when expected - based on the TAFs not the METARs as it happened, although the METARS agreed with the TAFs.

In fact on one of the trips back I wanted to know three days in advance when the predicted front would clear the east coast because I didn’t want to divert and the base could well have been below minima. As it turned out it was below minima for most of the day. Remarkably (although it was after a chat with the met office) three days in advance they were correct within an hour or two of predicting when the front would clear through. Mind you I have to say the morning TAF was less accurate as the front cleared perhaps a couple of hours more slowly than the TAF predicted!

It has been such a rotten summer that I think we all want to find a time within the forecast when we think we can get airborne in VMC. Inevitably we are disappointed when it doesn’t necessarily work out precisely as predicted - perhaps we are inclined to blame the forecast when in fact the forecast always gave us a probability of less than 100% of the prediction.

I also have a pet theory that the weather pattern has been different this year than in past years for as long as I can remember. Perhaps if I am correct that is making the modelling the met office do more of a challenge with perhaps more prob 40s and greater variation in the weather over the span of a day.

I was told only last week by someone with more than 10,000 hours in GA that part of the problem was the increasing reliance on automated weather stations. Look at today he said, the base on the ATIS is being given as scattered 800, broken 1,000, overcast 1,200 but I will eat my hat if you are not in the soup at 700 - just look at it. 10 minutes later I was in the soup at 1,300, not a foot lower and the met prediction of the tops was 3,000 feet - I was on top at precisely 2,850 and skimming the bubbly bits at 3,000. Now that aint bad.

Must check if he ate his hat.

However, this is only my limited experience. Bose is right if you know otherwise I think in the “exact” science of weather forecasting you will need to demonstrate that rather more than half of the prob40s are considerably different than forecast!

(say at least ten times - I really must stop keep agreeing with Bose :confused:)

airborne_artist

3rd Sep 2008, 12:40

This summer's weather 'harder to predict', leading weather forecaster admits - Telegraph (http://www.telegraph.co.uk/earth/main.jhtml?xml=/earth/2008/09/03/eaweather103.xml)

Clearly he'd seen this thread and thought it was time to 'fess up :ok:

Fuji Abound

3rd Sep 2008, 14:06

This summer's weather 'harder to predict', leading weather forecaster admits - Telegraph

Clearly he'd seen this thread and thought it was time to 'fess up

and

I also have a pet theory that the weather pattern has been different this year than in past years for as long as I can remember. Perhaps if I am correct that is making the modelling the met office do more of a challenge with perhaps more prob 40s and greater variation in the weather over the span of a day.

I promise I had not read that article!

Just getting together my cv as a weather girl!! :ok:

aviate1138

3rd Sep 2008, 14:07

Clearly it's the 1/10,000th part of total additional atmospheric CO2 added by mankind since 1750 that has caused the problem!

Where's that seaweed?

Why do they present the forecast as a given and not say " It might be that but might also be this."

Then they have really covered their ar*es!

Seriously though, if was so predictable everyone would do it. Black art afaiac.

Ken Wells

4th Sep 2008, 17:48

Well today was wrong again. How many examples do you want!!

Tell you what lets see if PPRuners can predict Saturday's weather

Few locations

Liverpool

White Waltham

Glasgow

Norwich
:D

S-Works

4th Sep 2008, 17:50

Come on then Ken, tell us exactly why it was wrong? Give us the forecast and actual and explain why it was wrong.

I have just checked the forecast v actual for today and they appear to me as they were forecast.

And for the record, you have not actually given ANY examples to date......

flies floats farts

4th Sep 2008, 17:58

In our op we micro analyse the weather hourly and daily all year long - the forecasts are, and have been for some time (years), totally useless. Thank god for radar (and no thanks for the opportunity to pay more for quicker updates) and webcams.

The health and safety approach of - tempo dust devils, prob30 plague of locusts, prob 30 tempo snow has gone too far. In addition, most of the tempos now last the entire forecast period - just in case !!

We often get a forecast something like : 10K+ SCT 1800 BKN 2500 PROB 30 TEMPO 4000m SCT 1000. What actually happens is the complete opposite: 4000m SCT 400' OVC 800 PROB 30 TEMPO 10K SCT 1800 etc (and like most prob 30s sometimes we never see the good bit of the forecast)

Having said all that, according to an airline friend of mine who flies long haul, he is absolutely amazed that the met office is able to very accurately and consistently forecast the wind and temperature at almost all of the waypoints on his 5000 mile flight plan. !!!!

Sorry Mr Met man but it's true - and as for examples you can audit yourself - check the forecast against the actuals.

S-Works

4th Sep 2008, 18:21

If the comment was aimed at me I am not mr met man and have no vested interest in protecting the MO. In fact quite the opposite, I sit on the working group meant to keep them honest. Therefore am happy to take genuine complaints to them and have them reviewed and if there is a genuine case make sure it is fixed.

Personally have never seen forecasts and actuals like you are claiming as a regular occurrence. But if you would like to give me the ICAO code for the location you are claiming this happens I will have the forecasts, actuals, after-casts etc pulled and reviewed in accordance with the MO SLA.

Also for the record if you really are seeing those type of discrepancies then they are a reportable incident and I can advise you of the process to do so.

At the end of the day the MO are contracted by the CAA to provide a service to aviation, if they are failing to do so we need to ensure they put it right rather than just complaining about it on forums.

robin

4th Sep 2008, 18:59

Bose

As I said earlier on, they have an SLA with the CAA and are measured on two indicators only. These do not involve the timings of fronts or even wind speed ad direction - they are (I believe, visibility and cloud amount)

They can certainly claim to be meeting those, as it would be difficult not to. I was told by the Met person at Air Expo that if we wanted more accuracy in the forecast to meet, say fronts, precipitation etc, then someone will have to pay for it.

As for an example, the EGTE TAFs were wrong (sort of) on Tuesday and, arguably, on Wednesday. The a*rse-covering prob 30 may give them a get-out though.

And is it true that there has never been a forecast which has not had 'cloud on hills' in it, even when the forecast is for CAVOK?

S-Works

4th Sep 2008, 19:09

Robin,

I am not arguing with you or anyone else. I need specific cases and I will have them investigated.

If you have a specific example of a forecast and actual being miles apart give me the ICAO code and time and date.

How far from accurate are you saying that the front timings are off?

I am not sure I understand your point about clouds on hills can you explain more? My understanding is that even on CAVOK days you are likely to get cloud on hills due to orthographic lifting and wave effect.

robin

4th Sep 2008, 19:45

Bose

I'll check my logbook for the date, but the TAFs on one day in July for EGTE started off with frontal rain due through around 10.00Z. The 1019 TAF put it back to around 1300Z and and by the 1622 TAF to 20.00Z. The rain arrrived around 22.00Z.

I can appreciate that these indicate a slowing of the weather systems, but the day was perfectly good for flying and (on the basis of the morning TAFs and BBC weather forecast) a number of students phoned up to cancel.

Similarly on Saturday last week at EGTE, the forecast showed poorish visibility, but not the low cloud we had around the airfield.

We are in a credit crunch affecting a lot of people, and they are not going to waste money on petrol getting to the airfield if there is a risk they won't be flying.

Our local schools are tearing their hair out trying to keep going, but they have lost a number of possible days of flying because students and hirers believe the forecasts.

Oh, and the cloud on hills bit is in the Airmet part of the Met Office site. It is just so strange to see CAVOK or SKC throughout the region, but with Cloud on Hills in every one I've ever seen.

S-Works

4th Sep 2008, 20:26

Ok so give me the EXACT dates and I will have them looked at.

Interestingly though I am seeing the possibility of localised weather conditions and a lack of correct interpretation rather than incorrect data.

As for the frontal comments, fronts do slow and the metoffice corrects the forecasts as it happens.

Nibbler

4th Sep 2008, 21:37

Interesting thread.

I work at a flying school and I ferry aircraft about in the South and Midlands when I'm not at 'work' so I find myself taking a daily view on the METARs and TAFs. Exeter, Bristol, Bournemouth, Brize, Boscombe, Heathrow and Gatwick are the sites I take most interest in. The reports and forecasts very accurate and have yet to let me down but I have noticed a couple of strange 'local affects'.

Local airfield weather can be very different from the reporting site just a few miles away. Old Sarum and Boscombe is an example where the TAF and METAR can read 9999 and at Old Sarum you can be sat in mist all day, yet the sites are just 2 miles away from each other.

Students and Hirers will read the TAFs and METARs then call the ops desk for our thoughts. Many a time I've had to give a view, taken from the METAR and TAF, significantly different from the information the caller has derived from the same reports. Some read the reports with hope, others with doom and a few incorrectly.

I have noticed a great number who don't use the 215, visible sat or rainfall radars which are also provided by the MO. These are tools which, taken with the TAFs and METARs, should allow us to 'take a view' on the weather and how it may affect the flight.

Forecasters do seem to take an over gloomy view for the TAF and I think it's fair and right to do so. You can plot the weather quite easily using the tools from the MO to decide if you agree with the forecaster or not. I'm not trying to teach anyone to suck eggs here but I've seen so few do it, seemingly stuck using just the TAF and METAR reports.

With all due respect to the people who have made complaints about the failings of the MO - Bose has provided a direct route to the heart of the organisation and you should provide the information and prove your complaint to be true and not just some pointless winge, which in reality proves nothing more than your own inability or lack of willing to read the weather correctly for yourself. From where I'm sitting it looks like the latter, as every TAF forecaster and METAR report downwind of your planned flight had to have been wrong, all day, including the 215 forecast and the weather radars. How likely is that then?

I await with a large humble pie at the ready.

:ok:

IO540

4th Sep 2008, 21:48

Maybe a big part of the problem is the PPL training establishment which takes a very cautious view.

Pre-PPL, you are flying on the instructor's license (so to speak) and he will not take risks. So a prob30 tempo, no matter how obviously totally wrong, has to be taken as a prob100.

On the self fly hire scene, so many renters are on marginal currency as well as knowledge, that the same applies.

Pilots are also taught to stay VFR (legally they have to) and again this leads to strict interpretations.

If however the UKMO said "OK here is the TAF but you should also look at this and this and this [weather radar, satellite, sferics, etc] " that would be different. But only self-taught IFR pilots discover these other sources - necessarily because the level of weather provision to sub-FL250-capability IFR pilots is pretty crap here in Europe.

If I was running a flying school I would ignore all prob30 stuff for purposes of all dual training/trial lessons and just use the MSLP charts for "will tomorrow be ok" purposes. Ignore cold fronts unless occluded or severely curved *at* the relevant place. Warm fronts will nearly always be crap for VFR...

007helicopter

6th Sep 2008, 06:24

Ken, as you started the thread and bose has offered to get an official response to the question it would be really interesting if you or anyone else can give some specifics ?

Lasiorhinus

6th Sep 2008, 07:15

A weather forecast is just that - a forecast. It is not, and never claims to be, a definitive statement about what the weather will do.

If you have a way of knowing with 100% accuracy what the weather will do in the future, please go and start your own forecasting company. With that sort of accuracy you will be very rich.

A Comfy Chair

6th Sep 2008, 08:39

bose-x ... I don't think the thread starter will, but I would like to thank you for volunteering to look into something for a ppruner... its guys like you that make this website worthwile :ok:

I have no experience of the met reports for the smaller airports around the UK, but I have operated into Heathrow for a few years now. (And Stansted, Gatwick and Amsterdam a few times when the Met Office was being pesimistic ;))

I'd have to say that the TAF's are, by world standards, absolutely first rate.

Ken Wells, you have to think about what other purposes the met forecasts have other than for just local flying clubs. Weather reports have major impact on passenger flights, changing alternate requirements, to the level of even preventing approaches. They do get them wrong occasionally, and they'll usually err towards the pesimistic side, given the strife an aircraft can get in if the weather turns worse. I'd like to know that the weather on the TAF is the worst that i'm going to encounter, and anything better is a positive.

You've seen yourself that the weather is often better than what you read the night before on the TAF... so why not wait until the day to decide? Same goes for students... why cancel the day before, when you could wait until the day, and find out what the weather actually is. This is the UK we're talking about after all :}

Metars are usually much more accurate anyway...

robin

6th Sep 2008, 08:55

Perhaps that is the real problem.

Remember that we are required to take the most up to date information to ensure that the flight can be undertaken safely - Met included.

Using the morning TAF/METRS/Airmet/F215/F214/length of seaweed etc, you can launch off into conditions that a later forecast determines to be unsafe.

And vice versa, but the vice versa seems to be more common.

In the type of flying I normally do, there is often not an opportunity to get an up-to-date forecast for the return trip, say 2 hours duration, so the morning forecast is almost invariably been replaced by a more current one.

My way around this is to call the home base for the latest TAFs at a couple of local airfields and current conditions at the site.

What I don't get with that are the general conditions en route.

At one event on the 12th July, there was a particularly nasty area of Prob 30 showers that merged into a band of continuous heavy rain and cloud on the deck - remember RIAT?

Either side of that area the weather was pretty good, but it was certainly 'exciting' finding a way through. Thank goodness someone on the ground had a smartphone where he could monitor the rainfall radar which showed where the worst bits had been and the clearer areas. Perhaps I ought to get myself an iPhone and use it as an on-board weather radar...

Actually, I never ignore Prob 30s but always take them into account and look to see whether the forecast fits.

In my gliding days, we had a bl**dy good metman who brought the subject to life. He taught us to use forecasts and to plan our tasks within the limits of the weather.

On the other hand, certainly at my local flying club, many power pilots plan a specific trip for a purpose, and look to see if the weather is ok for that trip. If not they will cancel. They don't even bother to see if there is a perfectly acceptable alternative flight.

NeuterCane

6th Sep 2008, 19:27

It would be interesting to see the TAFs/METARs for the locations/times that Ken refers to.

Poor Met Man/Woman not only has the difficulty of predicting the future but they also have to work within the rules laid down by ICAO and CAA (UK AIP, 3.5 gen, 3.5.4 c)). They cannot work outside these rules, and so often quite large changes (in cloud base/cover particularly - see the range 1500 FT to 4900 FT?) can occur over the period of a TAF without those changes being 'notifiable' - the forecaster has to consider how best to 'summarise' them in the forecast, and they no doubt more often than not go for the poorest set of conditions within that range - safety first. They can only put TEMPOs and BECMGs in if they expect conditions to breach the thresholds given for TAFs in UK AIP.

Understanding TAFs (and any forecast) properly requires the user to know what the change/amendment criteria are for that particular forecast type and ask themselves 'what would have to happen to the predicted conditions to a) cause the forecaster to include changes, or b) cause an amendment?'. Only then do you get an idea of what the possibilities are. Don't like the leeway? Lobby ICAO/CAA. RAF criteria is different, so don't apply UK AIP regs to RAF TAFs/Forecasts (though still done by Met Office forecasters)

The magnitude of the problem should not be underestimated - let's not forget that aviation forecasters are predicting the future and aiming to put numbers to cloud base, amount, visibilty, wind speed, direction etc hours into the future and working within the constraints of ICAO and CAA.

All developed nations (USA, Germany, France, Japanese, Chinese, Australians, combined European states in ECMWF etc etc) invest millions in supercomputers and forecaster training to 'solve' the future weather - maybe its not that easy.

I don't think the forecasters or the MO would claim to be perfect - but allow for the aspects that are beyond their control.

Oops Edited Civil TAFs to RAF TAFs, last sentence third para - ;-)

VFE

6th Sep 2008, 20:51

Might I suggest we all keep details of occasions when the TAF's don't turn out accurately and send these details to Bose-X? Lawd knows he's said it's ok to do so enough times now!

For the record I've seen more changeable WX in the UK over the last coupla years and a Met Office who perhaps struggle to forecast accurately as often as we'd like, which is of course 100% of the time!

Is this down to a dependance on computer forecasting software models?

My view on WX forecasting is that it's a bit of an artform and requires an experienced and adaptable human meteorologist' hand analysing the situation, given the myriad of different factors involved, rather than a reliance on software but hey, I appreciate this is 2008 and money is God...

VFE.

GOLF_BRAVO_ZULU

6th Sep 2008, 21:11

Let’s be fair lads, the Met wallahs are on a hiding to sod all. Britain doesn’t have a climate; we borrow other people’s. We call it temperate maritime but it’s so narrow North and South. as to be meaningless. Whether or not it’s the famous climate change, the WX is noticeably more extreme and erratic than it was, say, 20 years ago. If the MO does err on the cautious side (ie forecast worse than actual), in this sue-the-arse-off-you-at-the-drop-of-a-hat mentality we’ve acquired, who could blame them? The answer is, learn MET and read the synoptic chart. There’s more too it, though, than passing a multi-choice exam.

I, personally, don’t believe it’s worse now than in the ’70; long before computer models were invented. I’m reminded of my QXC in NOV 71; forecast 3,000 ft wind 30 deg short and over 5 Kt too light. I learned a lot about flying, as they say: over 15 mins late landing at Ipswich and long after Ipswich Homer had gone for his dins.

There used to be a Forecaster at Manch (Polish I think) who always gave a Night/Morning forecast of "possible hill fog", just in case.

Spamcan defender

6th Sep 2008, 21:25

Gotta feel sorry for Bose-x. All the guy has done is offer to investigate folks met issues and has asked for evidence yet folks refuse to provide it and give the guy nothing but hassle :ugh::ugh::ugh:.

Feel for you mate, I really do. :ok:

Spamcan

VFE

6th Sep 2008, 21:33

I wouldn't worry Spamcan, he's got thick skin! :}

All joking aside, sadly it does show that some posters here are all mouth and no trousers if they can't take a guy up on a decent offer like that.

VFE.

IO540

6th Sep 2008, 21:45

Does the UKMO actually have any civil or criminal liability whatsoever?

I think this is a myth.

If there was any liability, or even duty of care, they would have been sued for billions back in 1987.

More like, they get fed up with 10,000 old ladies phoning them up if they forecast a sunny day and the said ladies get their tomato plants washed out by a downpour.

LH2

6th Sep 2008, 22:19

Just to say, it looks very silly when people start venting their opinions on a subject they are in no way qualified to comment(1).

If you have any doubts regarding your use of weather information, by all means take up Bose's offer to have it looked into, or contact the Met Office directly with hard evidence of your observations, but our opinions on the subject are less than worthless. As has been pointed out already, just because you don't know how to use weather forecasts doesn't mean there's anything wrong with them.

(1) The first three posters to insinuate that passing a 40 question multiple choice exam makes you any wiser will be permanently relegated to my ignore list.

robin

6th Sep 2008, 22:51

Well that's told us...

So if I get this right:

If the weather turns out to be different from the forecast, then

a) we read it wrong in the first place
b) probably weren't qualified to understand it properly
c) didn't look at the right bit of the forecast
d) local effects got in the way of an accurate forecast
e) we misunderstood what the forecaster said.
f) because of UK conditions, we have to accept that the weather is 'mobile' and difficult to forecast. So changes are inevitable
g) Under no circumstances was the forecast at fault, and those of us who thought the forecast was wrong, are a bunch of whingers who aren't fit to lace the boots of those fine gentlemen and ladies of the Met Office....

:cool:

But I take Bose_X point and in future will be making a report of any issues with the forecast directly with the Met Office.

FlyingForFun

7th Sep 2008, 09:01

Bose-x,

I have a pet theory, and after skim-reading this thread I don't think anyone else has mentioned it yet, but it could explain why so many people perceive the TAFs to be inaccurate, but the Met Office stats show they are not.

I am talking about the PROB30s, PROB40s and TEMPOs. IO540 touched on this quite a way back, but I think there is a very specific problem, and that is that these elements of TAFs are too general.

Let's take today's TAF for Bournemouth, for example:

EGHH 070602Z 070716 29013KT 9999 FEW031 BKN048 TEMPO 0716 28015G25KT PROB30 TEMPO 0716 8000 SHRA

Many pilots, especially novices, would look at the PROB30 TEMPO 0716 8000 SHRA, and interpret that as meaning that there could be rain, and associated low visibility, at any point during the day, so they'd stay on the ground.

However, what I would do is look at the rainfall radar, which at the time of writing (just before 0900Z) clearly shows a band of rain to the north and east of Bournemouth. This band, which is associated with the occluded front shown on the Metform 215, is moving south-east, and does not appear to be going anywhere near Bournemouth today.

Based on this, I would bet that there won't be any rain over or to the west of Bournemouth for at least the next 3 hours, and if I was planning a trip to the west, so long as I was happy with the forecast strong winds and the possibility of clouds at 1500 as per the Exeter forecast, I would be ready to launch off right now. On this basis, those posters on this website who are claiming that Met Office TAFs are not accurate probably have reasonable grounds.

When the Met Office look at their stats, they will no doubt say that it was PROB30 TEMPO over a long period. The fact that the rain did not materialise for at least the first part of that period would not mark that forecast out as having been wrong. But if I can, with no more training in meteorology than what is required to scrape through the ATPL theory exams (and, to be honest, I've forgotten most of that anyway!), look at the publicly available Met Office data and forecast that there won't be any rain in Bournemouth for at least a few hours, why can't the Met Office do the same, and either cut out the PROB30 TEMPO, or change its time period to something a bit more specific?

FFF
--------------

IO540

7th Sep 2008, 09:09

FFF - you are quite right, but since when was this new fangled thing called the INTERNET a part of PPL syllabus?

:)

A PROB needs to be supplemented by radar or similar data but only if people know about it. The availability is only a recent development - Meteox (http://www.meteox.com) has been around only a year or two.

And, once outside one's home or club environment, how many airfields have a public PC with internet on it? Mine certainly doesn't. And the vast majority of pilots have no mobile internet and therefore no means of getting supporting info on weather once away from home.

I doubt anybody saying the forecast is duff will be able to support it with data - not because they are wrong but because getting the evidence (other than one for one's local area) is never easy. And anytime there is a PROB in there, you are wasting your time totally.

Steve N

7th Sep 2008, 10:09

The fact that Met office don't think there is a problem is the problem. No doubt they would prefer people send them their data individually. It can then be ignored. Forums like this are the closest we can get to "Watchdog" like embarrassment of large monoliths so keep plugging away at this. Making our points here as many above have done is the best way to bring about a change. As has been said before Met office can and I'm sure do check what happens against what they predict so they DO know. Forget the ins and outs of TAFS etc. That is a red herring. It's all starts from the synoptic forecasts.

Save of copy of this NOW: http://www.meteonet.nl/aktueel/brackall.htm. Then as the week develops see what actually happens.

Like many other pilots I do this all the time in the hope that I can get to that weekend fly-in etc. IMHO there appears to be a fundamental problem with the weather models used in that they always predict lows will track farther east than they end up doing. Take this weekend. According to the above it was predicted to track up the English channel into France. In fact as we saw it tracke across the mid west of the mainland. When you look at what they actually do they invariably turn north as they get to the UK. This happens over and over again and in fact we all learned about it during our PPL (Coriolis effect). It is common sense that If you can't predict where the centre of a pressure will end up is then anything forecast based on that will be wrong. My 2p Steve

Fuji Abound

7th Sep 2008, 10:11

The local TAF gives a prob40 of the cloud base being 1,000 with viz of 4K, less in showers. The area forecast supports the terminal forecast. The majority of the time the base is 1,800 with viz of 6K. As a PPL with a hundred hours are you happy to be flying in the prob40 bit or relying on the weather being better?

So when there are parts of the forecast that are on your margins - do you not fly because of those parts, and if those parts turn out to be better than forecast are you upset at the loss of a days flying in consequence.

My point. It is the forecast with a component on your margin that are the ones you don’t like - the stay on the ground, go scenario. You don’t like them because of the marginal component. If the marginal component wasn’t there you would be happy, until you were pi**ed in the air when it turned out it should have been in the forecast.

I am not making excuses for the (in)accuracy of forecasts.

I am simply observing that we all have are own personal limits but the trouble is forecasts will often straddle our limits, so we like the good bits, not the bad. The reality is whenever you are near your margin it is very unlikely you will see just the good bits if you are flying any sort of distance.

I can think of a couple examples. On a recent flight the passenger wanted to enjoy the scenery. The forecast predicted scud running would work fine for most of the route and the forecast proved correct. However the margin between low level flying and remaining legal was small particularly over the route which included some higher ground. On a few occasions the decision had to be made whether to work around the low cloud or give the scenery bit up as a bad job and go IFR. With 100 hours I know I would not have been happy working my way around the low bits, not being sure how far left or right of track I would end up.

I had an occasion last year when I knew my passenger was not that comfortable a flyer. We ideally wanted to get back. The forecast was for strong winds but I anticipated the landing would be just within my limits. Never the less I planned a diversion. As it turned out the wind ended up being over 40 knots with the whole lot being cross wind and to make matters worse, very gusty. The whole flight was thoroughly unpleasant with light to moderate turbulence. In so far as the wind was concerned the forecast was wrong, but in hind sight probably by only around 20% in terms of the strength and 20% in direction. The turbulence was to be expected but perhaps not as bad as it was.

I don’t know, but say I don’t like 10% of the forecasts if I need to go somewhere. That means 90% of forecasts even if they prove a little inaccurate I don’t really think about. For example if the cross wind is 5 knots more than predicted I suspect it doesn’t really register. However when I didn’t like 40% of the forecast there were more occasions when I thought the forecaster had got it wrong because smaller changes in the actual weather put me closer to my personal limits.

And so, my point is, when the weather is on your margin you don’t go flying because you know marginal weather usually contains a forecast of bits that are below your personal limits and because marginal weather is rarely uniform (low bases are rarely exactly 1,200 feet over the entire area) there are almost always bits below your limits. Having decided not to go you watch the progress of the weather and note that none of the marginal bits turned up. Of course you understand that the forecast predicted this was likely but we still like to convince ourselves that had we gone flying we wouldn’t have run into any of the marginal bits either.

In short I am not saying you didn’t understand the forecast, or weren’t qualified to understand but at least on some occasions you did the sensible thing but of course wished you had gone aviating!

NeuterCane

7th Sep 2008, 10:50

If issues are raised with CAA (or BOSE), then the Met Office cannot ignore comments.

Regarding use of the longer range charts for planning booking of aircraft/planning trips, it is a simple fact that the further out you look the less accurate it is going to be, but again actually think about the magnitude of the problem. Even for a front or low that is moving steadily with no changes, if the rate of movement is only 1 KT/hr in error, then in 120 hours the front/low will be incorrectly placed by 120 miles. That says nothing for developing systems that accelerate/decelerate or features that dont even exist on the analysis.

Look where all the lows are on the charts SN referred to. How much surface data is there out there in the Atlantic? Satellite fills in the gaps, but there are many unknowns.

Jetstreams drive the weather, and if they are wrong there will be downstream consequences in days to follow. Airliners crossing the Atlantic report winds but western bound ones will avoid the strongest headwinds (so denying data that would be very useful to fine tune models). Even those heading east and riding the wind may well avoid the strongest winds to avoid the associated turbulence.

There are plenty of other models out there to consult and compare, they will suffer the same issues.

I don't think the MO are out there to deliberately spoil your flying or you barbeques. When I see that many lows floating around to the west over the next few days perhaps the 'hint' is that perhaps conditions will be too changeable to plan anything for certain - paint the front room instead ;-)

S-Works

7th Sep 2008, 14:27

Well just back from the Guernsey Air Rally where the weather was as forecast and just catching up on this.

I am not sure where the comments about the MO ignoring complaints are coming from. Every time I have asked for a complaint to be investigated it has been done so with good humour and accurate responses. The MO are VERY keen to provide a first class service and do not ignore any feedback in my experience to date.

But moaning about forecasts on forums without providing any FACTS does not constitute making a complaint to the MO which is then 'ignored'. Like I have said a few times already, provide hard facts and let me have them or complain direct to the MO with an MOR and they WILL be acted on.

EGHH 070602Z 070716 29013KT 9999 FEW031 BKN048 TEMPO 0716 28015G25KT PROB30 TEMPO 0716 8000 SHRA

Many pilots, especially novices, would look at the PROB30 TEMPO 0716 8000 SHRA, and interpret that as meaning that there could be rain, and associated low visibility, at any point during the day, so they'd stay on the ground.

Which would be the correct interpretation, but it is the pilot making the decision not to fly not a problem with the forecast. 8000m is legal VFR and I know my aircraft does not stop working when it gets wet. So if a pilot chooses to misinterpret the forecast how does it make the forecast wrong?

robin

7th Sep 2008, 14:32

Which would be the correct interpretation, but it is the pilot making the decision not to fly not a problem with the forecast. 8000m is legal VFR and I know my aircraft does not stop working when it gets wet. So if a pilot chooses to misinterpret the forecast how does it make the forecast wrong?

Interesting that you would use the term 'misinterpret the forecast' here. Perhaps they interpreted as being outside their personal limits.

One of my group aircraft did fly through a Prob 30 shower and damn near wrecked the wooden prop. That was more down to airmanship than weather interpretation, but subsequent to that we all got very nervous flying the new prop in any sort of rain - for a bit, anyway.

FlyingForFun

7th Sep 2008, 14:36

Bose-x,

Glad you had a good day out in Guernsey, and welcome back to the debate. I think you have missed my point, though.

I agree that the forecast weather is VMC, but it may well be below what many pilots would be happy flying in VFR (as well as robin's very valid point about wooden props, or perhaps you might not want to fly in rain in your open-cockpit aircraft because it's not fun), so I think that debate can be put to one side for now. Let's asume personal limits at the planning stage of the flight, for whatever reason, of 10k viz and no precipitation.

The issue is with the interpretation of the weather information. You said that my interpretation of the TAF was correct. However, my issue is that, just by looking at the Metform 215 and the rainfall radar, I was able to produce a better forecast than the TAF. The TAF said might rain, I said it won't rain. I've been sitting in my living room, 4nm from the airfield, for the whole of the period of the TAF.... and guess what? It hasn't rained. So my question is, if I knew it wasn't going to rain, why didn't the Met Office?

FFF
-----------------

S-Works

7th Sep 2008, 15:22

So my question is, if I knew it wasn't going to rain, why didn't the Met Office?

Or perhaps you thought it was not going to rain and it didn't. The MO thought it might rain so gave a probability that it might happen but at Prob30 it was a low probability?

Now if they had not put the likelihood of rain in and it did rain we would complain the forecast was wrong......
;)

NeuterCane

7th Sep 2008, 15:29

Taking a look at the radar imager for 1100 Local (1000 UTC) on the Met Office public web site (sorry, haven't figured out how to insert images), shows rain/showers just north and west of Southampton - and not very far from Bournmouth at all. FFF I take "won't" as meaning 0% possible, but really with rain so close (and also some further NW) I don't really believe the possibitlity was 0%. Seems to me they knew showers would develop in the area, but cannot predict the track of every one.

Southampton reported -DZ and -SHRA at 1020 UTC And 1050 UTC.

Seems to me Met O were alerting to a realistic alternative, without going overboard and forecasting truly awful vis and drastically lowered cloud base. In fact the cloud cover was very variable at Bournmouth, and actually lower than the 0716 TAF predicted, but all within the 'leeway' allowed for in ICAO's and CAA's rules.

EGHH 071350Z 28013KT 9999 FEW010 BKN021 16/12 Q1008=
EGHH 071320Z 29011KT 9999 FEW010 BKN021 16/12 Q1008=
EGHH 071250Z 32013KT 9999 FEW018 BKN022 16/12 Q1008=
EGHH 071220Z 31011KT 9999 BKN021 16/12 Q1007=
EGHH 071150Z 31010KT 9999 SCT022 BKN028 16/12 Q1007=
EGHH 071120Z 31012KT 9999 SCT021 BKN022 16/12 Q1007=
EGHH 071050Z 31012KT 9999 SCT021 BKN028 16/12 Q1007=
EGHH 071020Z 31014KT 9999 FEW009 BKN022 16/12 Q1006=
EGHH 070950Z 30011KT 9999 FEW009 BKN019 15/12 Q1006=
EGHH 070920Z 30013KT 9999 FEW010 BKN019 15/11 Q1006=
EGHH 070850Z 29014KT 9999 SCT024 15/12 Q1005=
EGHH 070820Z 30013KT 9999 SCT030 15/12 Q1005=

TAF EGHI 071159Z 071322 26008KT 9999 BKN025=
TAF EGHI 070906Z 071019 25008KT 9999 BKN025 PROB30 TEMPO 1016 8000 -SHRA=
TAF EGHI 070638Z 070716 29013KT 9999 SCT020 TEMPO 0714 28015G25KT PROB30 TEMPO 0716 8000 SHRA=

EGHI 071150Z 26006KT 230V300 9999 FEW020 BKN024 14/11 Q1006=
EGHI 071120Z 26006KT 230V300 9999 FEW020 BKN024 14/11 Q1006=
EGHI 071050Z 26006KT 230V300 9999 -SHRA FEW020 BKN024 14/11 Q1006=
EGHI 071020Z 25006KT 220V290 9999 -DZ FEW018 BKN022 14/10 Q1006=
EGHI 070950Z 25006KT 220V290 9999 FEW018 BKN022 14/10 Q1005=

VFE

7th Sep 2008, 16:02

As I hinted at before, it's all about (artistic?!) interpretation and as FFF unwittingly highlighted, maybe too high a reliance on TAF's as opposed to the bigger picture in some quarters? I would suspect a failing during a pilot's practical training. How many times does an airfield receive telephone calls from potential visitors and club members asking about the weather?

The TAF is an aid, one of the many available (including simply looking out the bloody window but don't get me started on that!) for a pilot to determine if the WX conditions appear favourable for their particular journey. The TAF in no way absolves a pilot from responsibility does it? So why do some place their entire decision making process for a flight on it? As FFF stated, it is but one peice of information in the bigger picture and for what it's worth is more right than wrong, even in todays ever more unpredictable climate, when I check it five times a week for my patch.

My advice would be to continue the interest in Met following the written exam and past simply a fixation with TAF's. I flew for a whole year without ever looking at a TAF and avoided bad weather quite easily! lol

VFE.

PPRuNe Radar

8th Sep 2008, 08:53

Ken Wells

Any progress in identifying precise dates/times where the Met Office made a Horlicks of it, so that bose-x can get them to have a look ??

Ken Wells

10th Sep 2008, 17:15

will do, away till next week.

robin

11th Sep 2008, 20:53

OK - how about the sequence of the synoptics for, say, tomorrow.

During the week I have been monitoring the synoptics (not the BBC) with a view to a cross-channel trip. The forecast for tomorrow has showed a ridge building and a possibly good day, but in a number of the charts there were troughs indicated on the French side.

This morning, the chart showed the ridge stronger than before, but without the troughs. I spent the day planning and arranging the flight but late this afternoon we find the troughs back and an occluded front.

I'll be watching to see how the overnight charts develop as my optimism has started to plummet.

S-Works

11th Sep 2008, 21:09

If you are asking for advice on whether you can do your flight tomorrow, then looking at the forecast I would go.

The trough is way SW in France and it looks like during the day there will be a nice gap in the weather.

Where you going, LFAT? Want some company? :)

robin

11th Sep 2008, 21:46

No Cherbourg - and the trough is sitting nicely in that area......

VFE

12th Sep 2008, 09:03

then looking at the forecast I would go.

The sort of silly comment that as an 'instructor' you should not be coming out with, especially on a website like this where you don't have all the facts about destination, aircraft equipment, pilot qualifications, etc etc...

Tut tut!

VFE.

Stratus Fractus

12th Sep 2008, 10:11

Good Discussion, and some GREAT new websites for my favorites folder. I can see my morning weather routine being somewhat longer now. Some of the satellite images on www.sat24.com (http://www.sat24.com) are stunning (linked from meteox).

One website I use a lot and very few powered pilots seem to be aware of:

www.xcweather.co.uk (http://www.xcweather.co.uk)

Make sure you go into preferences menu and turn raw data on, then you will see that each station provides the current METAR. This gives a very quick overview of the regional situation upwind or downroute or whatever you like.

Cheers SF

S-Works

12th Sep 2008, 11:02

Quote:
then looking at the forecast I would go.
The sort of silly comment that as an 'instructor' you should not be coming out with, especially on a website like this where you don't have all the facts about destination, aircraft equipment, pilot qualifications, etc etc...

Tut tut!

VFE.

What???? For a VFR flight the aircraft type is irrelevant. One would assume that the pilot holds a PPL. The question was answered with the assumption that it was a basic PPL doing a VFR flight. Therefore anything else does not come into the equation.

The conditions are VFR, the current METARS support VFR as do the TAFS for that region.

You really do have a bug up your ass with me at the moment don't you.

Fuji Abound

12th Sep 2008, 11:17

Bose

As it turned out looking at the weather radar plenty of heavy showers along the north French coast this morning although the Cherborough penisula, rather than LFAT, was pretty clear.

VFE

12th Sep 2008, 12:17

Assumption = the mother of all cock ups.

And yes I do have a bug with you at the moment.

VFE.

bookworm

12th Sep 2008, 13:54

Let me first say that I think the forecasters at the Met Office generally do a very good job. Weather is not deterministic in practical terms, and all you can do is come up with a best estimate. Sometimes, it's going to be wrong. Better forecasts require much more money, which is just not justified by the value added to the forecasts.

That said, Sep 12 2008 around Cambridge was not one of their finest days.

TAF EGSC 120624Z 120716 26005KT 9999 SCT025=
TAF EGSC 120855Z 121019 26006KT 9999 SCT025 PROB40 TEMPO 1319 8000 SHRA=
TAF EGSC 121157Z 121320 VRB05KT 9999 FEW012 SCT030 PROB30 TEMPO 1420 6000 SHRA=

METAR EGSC 121150Z 00000KT 9999 FEW012 14/12 Q1013 =
METAR EGSC 121220Z 24002KT 9999 SCT008 SCT014 14/12 Q1013 =
METAR EGSC 121250Z 20004KT 160V260 9999 -RA SCT007 BKN015 14/12 Q1013 =
METAR EGSC 121320Z 19004KT 150V250 5000 -RA FEW005 SCT014 14/13 Q1013 =

A look at the radar suggests that they were caught out by how far west the rain associated with a slow moving front in the N Sea has penetrated.

This is not the PROB40/30 convection, which is still waiting in the wings to the west. What we have is something far more like the Norwich forecast, closer to the front:

TAF EGSH 121157Z 121322 29005KT 8000 -RA SCT012 BKN030 TEMPO 1318 4500 RADZ BKN009 PROB40 TEMPO 1316 BKN003 BECMG 1820 9999 NSW BECMG 2022 3000 BR BKN005=

Operationally, the only consequence for me is that I wish I'd picked up my coat when I left the house today. :(