For help, advice and discussion about stuff not related to aviation. Play nice: no religion, no politics and no axe grinding please.
User avatar
By George
FLYER Club Member  FLYER Club Member
#1860946
Is it me, or have they got worse? As an example in South Wales yesterday we were told to expect rain all day and heavy thunderstorms all afternoon. The reality was a light shower at 08.00 then a totally dry day after that with light clouds.

I had cancelled a proposed flight on the strength of the forecast and although it was minima on visibility, it was wildly off the mark on what had been forecast.

With all the computing power available why do forecasts so often get it wrong?
#1860948
Suppose an estimate of wind speed of a forecast 10 hours ago was 5 mph out. Then the frontal position could be 50 miles out, which may be enough to make a planned flight either feasible or unfeasible.

I always try to back up forecasts with netweather radar to see where the rain has got to, and to visualise the speed at which it is moving.
#1860949
The forecast in South Wales was for a risk of thunderstorms. In my part of South East Wales that is exactly what we got. Thunder, and a period of the heaviest rain here for quite a while. Thunderstorms are often very localised, but this forecast was bang on, and I certainly wouldn't have wanted to be airborne.

The squall line showed up very clearly on the Met Office rainfall radar.
User avatar
By GrahamB
FLYER Club Member  FLYER Club Member
#1860950
The fact that you can look at forecasts based on different models, and each will say something slightly (or more) different, should tell you something.

Ultimately they are based on both measurements and a whole range of assumptions; different modelling assumptions will generate different results, and assumptions can be wrong.
johnm liked this
By johnm
FLYER Club Member  FLYER Club Member
#1860952
We had a BBQ for 50 people yesterday afternoon at a nearby farm. It is close to Fairford and we put up some large Gazebos just in case but Fairford's TAF was bang on :thumleft: and we saw just a few spits and spots.....
#1861001
I keep a notebook where I note down the things I've learned after, pretty much, every flight. One of the early notes says:

Never, ever cancel a flight on the basis of a forecast until you are at the airfield.


From Above the Law:

Everybody in aviation, at some point or another, if not quite continuously, moans about the ‘met man’ – the meteorological officer who reports on and forecasts the weather. The ‘met man’, who is quite often a met woman, works for the Met Office, which makes sense, I suppose.

The Met Office, which is based in Exeter in the south-west of England, has a super-computer which cost £97 million and is capable of doing 14,000 trillion calculations a second, which sounds like the kind of number I’d make up, just to try to be funny, but it isn’t. This computer is eighty per cent accurate, which sounds pretty good until you realise that the weather patterns in the UK tend to go in four-day runs, sometimes more.

For instance, as I write this we are in a weather pattern in the east of England which means that the weather we had yesterday – dull with the odd shower – is not only the weather we’ve got today but is the weather we’re likely to have for the next week. The temperature at midday is not likely to vary by more than a couple of degrees all week and even the wind speed and direction isn’t supposed to be very different tomorrow to what it was yesterday. That means that if you say ‘much the same as today’ you will be accurate in your forecast for the next seven days – that’s eighty-six per cent accurate and you’ll have saved £97 million. And, when I say ‘you’ll’, I do mean you, because that was your tax money.

That’s not to say that the Met Office has wasted money on its computer – oh no. Their massive computer can give them all sorts of interesting statistics – like the fact that the warmest place in the UK is the Scilly Isles and the sunniest is Bognor Regis. Interestingly enough, equidistant between these two is a city called Exeter. Coincidence?

https://www.amazon.co.uk/Above-Law-Adventures-police-helicopter/dp/1785632620/ref=tmm_pap_swatch_0?_encoding=UTF8&qid=1627208539&sr=8-2
#1861006
JAFO wrote: ... as I write this we are in a weather pattern in the east of England which means that the weather we had yesterday – dull with the odd shower – is not only the weather we’ve got today but is the weather we’re likely to have for the next week.


In fact the forecasts for today included the rain and thunderstorms that @George said were forecast for Wales yesterday.

Looking out the window it's a calm and settled day so far, with mediocre viz.

If I look at FR24 or ADSB Exchange, there isn't a GA aircraft to be seen in the bulgy bit of East Anglia. I think we all believed the forecast.

Of course, this could mean it's an 'Everyone switch off their Mode S' day

Rob P
JAFO liked this
#1861026
Ninety minutes later and there's a glider over Rattlesden and a Chippy over Essex.

Rob P
By rogerb
FLYER Club Member  FLYER Club Member
#1861075
Perhaps there is a tendency to warn of the worst that could happen just in case.
And not wanting to appear incautious we react as though it will.
Bit irritating though when a glorious flight would have been possible.
User avatar
By PeteSpencer
FLYER Club Member  FLYER Club Member
#1861090
It's amazingly patchy here in E Anglia:
I set off this arvo in bright sunshine to take a load of stuff to the airstrip 14 miles northeastish.

Halfway there it was apparent that there was some serious weather in the direction of the strip and sure enough two miles later I ran into a mother of a storm: the wipers couldn't clear the rain and I slowed to a walking pace.

I decided to carry on but two miles down the country lane I suddenly encountered a deep puddle halfway across the road sending water over the top of the car. The car immediately spluttered , all manner of warning lights came on and I thought the engine was going to stop.

I kept the engine going and it was clearly mis-firing: Decision time: I decided to carry on as I was less than a mile from the strip but rounding the next bend was a deep brown flood fed by torrents of muddy water coming from the fields.

As I was pondering the next move a 4 x 4 came in the opposite direction and the water was up to his axle.

I came home....................

Home was bone dry , no trace of rain and for the last three hours thunder has been rumbling all round the house , but nary a drop of rain.

I'll let you know if the car starts in the morning.... :roll:


Edit. Well it started and the engingne management caption disappeared so no obvs harm done by its soaking ..,,,, :x
Last edited by PeteSpencer on Mon Jul 26, 2021 6:20 pm, edited 1 time in total.
User avatar
By mick w
FLYER Club Member  FLYER Club Member
#1861092
PeteSpencer wrote:It's amazingly patchy here in E Anglia:
I set off this arvo in bright sunshine to take a load of stuff to the airstrip 14 miles northeastish.

Halfway there it was apparent that there was some serious weather in the direction of the strip and sure enough two miles later I ran into a mother of a storm: the wipers couldn't clear the rain and I slowed to a walking pace.

I decided to carry on but two miles down the country lane I suddenly encountered a deep puddle halfway across the road sending water over the top of the car. The car immediately spluttered , all manner of warning lights came on and I thought the engine was going to stop.

I kept the engine going and it was clearly mis-firing: Decision time: I decided to carry on as I was less than a mile from the strip but rounding the next bend was a deep brown flood fed by torrents of muddy water coming from the fields.

As I was pondering the next move a 4 x 4 came in the opposite direction and the water was up to his axle.

I came home....................

Home was bone dry , no trace of rain and for the last three hours thunder has been rumbling all round the house , but nary a drop of rain.

I'll let you know if the car starts in the morning.... :roll:


Just don't use the Choke , it's already flooded :wink:
#1861096
Thunderstorms form when the air is unstable. A disturbance develops into the parcels of air rearranging themselves and result in a thunderstorm. All the forecast is telling you is that the air is unstable. It can't actually tell you exactly when or where a thunderstorm might form; only that the air is unstable and that it might. Whether a thunderstorm will or won't develop seems to be the least reliable part of weather forecasting; all that is happening is that they're telling you that the conditions are right for them.

I find that it helps to better understand where and how the forecasters are getting their information. Then you can "level up" your insight by understanding the big picture, and the likelihood of different weather outcomes - instead of just the most likely outcome. This helps hugely with planning.
kanga, townleyc liked this
#1861155
If you want an accurate forecast, you really have to do it yourself using the online tools which are available. This is because the generic forecasts like those from the Met Office or the BBC are trying to give snapshots for a particular location in a simple to understand way, so they don't give the information about trends and direction of travel which you want for flying.

I'm just back from a camping weekend at a gliding club and made forecasts for Sat and Sun, mainly using the RASP gliding weather tools - http://rasp.stratus.org.uk/index.php/rasptable-desktop. Our main concern was rain (gliding was a secondary activity for my group), and I could use the tools to see where bands or patches of rain were due to develop, which directions they were moving in, etc. As a result I correctly predicted no rain during daytime either day, while the BBC forecast appeared to show showers al day - but the BBC was giving around 30% chance of showers each hour, whilst I was tracking the more granular data and forecasting only for one specific place. In the 20 mile circle around the gliding site there were, indeed, showers in some places, so the BBC wasn't inaccurate, just not as granular as I wanted.

The learning curve to use these tools is quite steep (you have to cross-check different sources and use experience of past forecasts to decide what to predict), but with practice in 20 mins or so you can produce a very accurate picture.
kanga, Paul_Sengupta liked this