UKBouldering.com

Topic split: Stick to the Met Office (Read 4347 times)

shark

Online
  • *****
  • Administrator
  • forum hero
  • Posts: 8697
  • Karma: +625/-17
  • insect overlord #1
Topic split: Stick to the Met Office
December 08, 2020, 10:53:49 am
Drizzly and dank looking at Burbage not as forecast. Again. Driving on to the Tor instead

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#1 Re: Topic split: Stick to the Met Office
December 08, 2020, 12:45:26 pm
What forecast are you using, that's exactly what was predicted?

shark

Online
  • *****
  • Administrator
  • forum hero
  • Posts: 8697
  • Karma: +625/-17
  • insect overlord #1
#2 Re: Topic split: Stick to the Met Office
December 08, 2020, 12:57:10 pm
What forecast are you using, that's exactly what was predicted?

BBC. No rain forecast

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#3 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:05:57 pm
Stick to the Met Office.

T_B

Offline
  • *****
  • forum hero
  • Posts: 3078
  • Karma: +149/-5
#4 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:39:42 pm
Agreed. Met office has been pretty consistent for the Peak this past couple of months, apart from the thurs/Friday a couple of weeks ago. I don’t bother looking at others any more.

Stuart Anderson

Offline
  • *
  • regular
  • Posts: 42
  • Karma: +2/-0
#5 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:42:54 pm
Stick to the Met Office.

If I could add, not just the Met Office but their mountain forecast plus the pressure chart and rain radar are really useful.

The Norwegian site is decent too https://www.yr.no/en

cheque

Offline
  • *****
  • forum hero
  • Posts: 3389
  • Karma: +522/-2
    • Cheque Pictures
#6 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:44:47 pm
The Met Office smartphone app is really good- lets you save locations so you can just scroll down comparing what the weather will be like at different places at the same time rather than repeatedly searching.

T_B

Offline
  • *****
  • forum hero
  • Posts: 3078
  • Karma: +149/-5
#7 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:47:16 pm
Stick to the Met Office.

If I could add, not just the Met Office but their mountain forecast plus the pressure chart and rain radar are really useful.

The Norwegian site is decent too https://www.yr.no/en

Whilst I use yr.no for other mountain areas I think it’s overly optimistic for the Peak.

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#8 Re: Topic split: Stick to the Met Office
December 08, 2020, 01:59:36 pm
Whilst I use yr.no for other mountain areas I think it’s overly optimistic for the Peak.

Ditto for Yorks grit areas, also never seems to pick up dankness as well as the ominous met office 20% grey cloud forecast does.

tomtom

Offline
  • *****
  • forum hero
  • Posts: 20282
  • Karma: +641/-11
#9 Re: Topic split: Stick to the Met Office
December 08, 2020, 02:06:27 pm
I find a whatsapp network of climbers based around all major locations works far better than anything else :')

Nothing like someone looking out of the window to give an accurate account of the weather.

That said - I have on multiple occasions ignored their advice and been duly disappointed...

lagerstarfish

Offline
  • *****
  • forum hero
  • Weapon Of Mass
  • Posts: 8810
  • Karma: +812/-10
  • "There's no cure for being a c#nt"
#10 Re: Topic split: Stick to the Met Office
December 08, 2020, 06:35:32 pm
Drizzly and dank looking at Burbage not as forecast. Again. Driving on to the Tor instead

I too fell for more of the BBC's lies this morning.

I'll not be moving to less planet destroying transport to the crag until I get better conditions predictions.

(Or electric bikes get a lot cheaper)

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#11 Re: Topic split: Stick to the Met Office
December 08, 2020, 06:58:14 pm
Whilst I use yr.no for other mountain areas I think it’s overly optimistic for the Peak.

Ditto for Yorks grit areas, also never seems to pick up dankness as well as the ominous met office 20% grey cloud forecast does.


It was a not dissimilar classic winter high pressure gloom about fifteen years back that made me delete all the euro computer based forecasts from my bookmarks. Worthless.

Met Office, occasionally MWIS, but also keep an eye on the synoptics so you've got an idea whether the forecasting is straightforward or not and how it isn't likely to vary from the prediction.

Although the Met Office have not been properly privatised the promotion of 'competition' saw them lose their 93 year long BBC contract, which was awarded to the French and unsurprisingly the BBC forecasts are not as good. Not because they're French, because they haven't got 150 years experience of our fickle island weather.

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#12 Re: Topic split: Stick to the Met Office
December 08, 2020, 07:09:16 pm
What page do you use to look at the Synoptics? I know you can see the fronts on the met office vids but sure there must be a better resource.

Also in my simple mind winter highs were good news conditions wise, do they have to have moved in from the north or east to have the dry air?

tomtom

Offline
  • *****
  • forum hero
  • Posts: 20282
  • Karma: +641/-11
#13 Re: Topic split: Stick to the Met Office
December 08, 2020, 07:18:06 pm
The high is best positioned off to one side so it brings in a bit of northerly or easterly. The last coupe we’ve had have seemed to sit over us and just lead to clag.

Personally the most useful data is rain radar - and I use an app that gives it in 5 min increments with a 5-8 min delay from real time. Scrolling back and forth to see what’s coming and from what direction is most useful (I’ve found).

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#14 Re: Topic split: Stick to the Met Office
December 08, 2020, 07:36:10 pm
The met office rain radar is decent too and that has high def detail for the previous 6 hrs which is super useful like you say.

tomtom

Offline
  • *****
  • forum hero
  • Posts: 20282
  • Karma: +641/-11
#15 Re: Topic split: Stick to the Met Office
December 08, 2020, 07:39:44 pm
It’s all the same radar (there is only one network) just the number of time slices and how close to real time it is depends on how much you or your app maker pays!

I’ve a met office rant brewing so I’d better stop posting 😄


GazM

Offline
  • ****
  • forum abuser
  • Posts: 533
  • Karma: +29/-0
    • Highland ramblings
#16 Re: Topic split: Stick to the Met Office
December 08, 2020, 07:59:53 pm
What page do you use to look at the Synoptics? I know you can see the fronts on the met office vids but sure there must be a better resource.

I use these. Scroll down to the '9 panel GFS' (not sure what the GFS bit means, but it's a handy view of synoptic charts.
https://www.wetterzentrale.de/


teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#17 Re: Topic split: Stick to the Met Office
December 08, 2020, 08:04:02 pm
Cheers Gaz

Sorry for starting to get you wound up TT! Re radar just meant the past six hour data is higher def (smaller square size) than the forecast data. Like you say seeing which way systems are moving and when they have cleared certain areas is the most useful thing.

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#18 Re: Topic split: Stick to the Met Office
December 09, 2020, 09:51:47 am
I view synoptics at https://www.metoffice.gov.uk/weather/maps-and-charts/surface-pressure/#?tab=surfacePressureColour

The 'proper' ones inc geostrophic winds are under the 'black and white' tab. Like Gaz, I used to use wetterzentrale to view these until the Met did a user survey and everybody complained they were having to use a german site to view them.

For radar I use the Home and Dry app, which is brilliant. This is probably my most used resource along with the met app, although the forecasting/ time projection needs a pinch of salt, especially re orographic. Or raintoday.co.uk on desktop.

Plus the sat view on https://en.sat24.com/HD/en/gb/visual - they also have an app and you can switch to infrared for night.

When my Dad was gliding a lot he used to give the overall forecast a quick glance before taking a deep dive into tephigrams aka SkewT plots - atmospheric soundings from balloon releases. They graph dewpoint with altitude and illustrate things like inversions very clearly. Not the most intuitive things but might be worth getting your head around if you were projecting somewhere particularly condensey. This met factsheet PDF is probably the best start but google is your friend.

highrepute

Offline
  • *****
  • forum hero
  • Posts: 1288
  • Karma: +109/-0
  • Blah
#19 Re: Topic split: Stick to the Met Office
December 09, 2020, 10:55:18 am
I feel like I'm missing something. But MetOffice has been as crap as everything else of late.

For example. Right now in Sheffield it says 10% chance of rain and has a white cloud.

I look out the window and it's raining. Both BBC and google have predicted rain showers.

This seems to be happening every weekend at the moment, I'm at the crag it's raining and metoffice is telling me it's 5 to 10% chance of rain.

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#20 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:10:01 am
Rain radar on met office showing isolated showers in Sheff area currently. I’ve always taken the 10% as it being likely to rain for 10% of that period, rather than a 10% chance of it raining.

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#21 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:15:53 am
Quote from: Met Office
For example, a 70% chance of rain represents a 7 in 10 chance that precipitation will fall at some point during that period.

spidermonkey09

Offline
  • *****
  • forum hero
  • Posts: 2814
  • Karma: +159/-4
#22 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:17:54 am

This seems to be happening every weekend at the moment, I'm at the crag it's raining and metoffice is telling me it's 5 to 10% chance of rain.

Story of my life climbing in the Peak the last few years. Burbage is particularly prone to it I think!

 
I’ve always taken the 10% as it being likely to rain for 10% of that period, rather than a 10% chance of it raining.

I don't think this is right, even in the summer the rain likelihood is often 10% for much of the day, even on days where its abundantly clear no rain is going to fall. I  *think* this is backed up by the clarification here but my brain is mush today so cant be sure! https://www.metoffice.gov.uk/weather/guides/what-does-this-forecast-mean

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#23 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:25:24 am
I’m learning so much about the forecasts 😄 So their forecast is just shit then and doesn’t match the satellite data, or Highrepute is just unlucky!

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#24 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:28:54 am
I think the trick is not to take a football supporters approach to forecasts. If you look at a couple and they're different, rather than trying to pick a winner take the message that there is significant uncertainty. If we're a day or so ahead I'd then look at the synoptics, if it's now I don't bother with forecasts much at all and just look at radar and satellite data.

jshaw

Offline
  • **
  • addict
  • Posts: 138
  • Karma: +4/-0
#25 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:36:24 am
Haven't read the whole thread but I read somewhere that weather forecasting is more inaccurate than usual at the moment as the agencies get some of their data from sensors on aeroplanes. As there are fewer flights at the moment, especially trans-Atlantic, it's increasing the inaccuracy of an already unpredictable science.

(I might have mis-remembered some details. Apologies if this has already been discussed above)

spidermonkey09

Offline
  • *****
  • forum hero
  • Posts: 2814
  • Karma: +159/-4
#26 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:55:18 am
I think the trick is not to take a football supporters approach to forecasts. If you look at a couple and they're different, rather than trying to pick a winner take the message that there is significant uncertainty. If we're a day or so ahead I'd then look at the synoptics, if it's now I don't bother with forecasts much at all and just look at radar and satellite data.

I would really like some sort of 'percentage uncertainty' metric on a forecast as currently I don't think there is a way to know this without being a meteorologist.

So their forecast is just shit then and doesn’t match the satellite data, or Highrepute is just unlucky!

I increasingly think most of the forecasts are shit. Maybe due to jshaw's info, or climate change or whatever, but the consistent shitness/uncertainty of the forecast in what is fundamentally a pretty flat country is amazing. Maybe its because I only care about rural/hilly areas so dont notice if the forecast is spot on all week for urban conurbations, which tbf is probably what most people would prioritise.

teestub

Offline
  • *****
  • forum hero
  • Posts: 2574
  • Karma: +166/-4
  • Cyber Wanker
#27 Re: Topic split: Stick to the Met Office
December 09, 2020, 11:57:49 am
Met office abs MWIS will often say in their written reports that there’s uncertainty, particularly when there’s storm systems fucking shit up.

I think also as climbers what we want out of a forecast is a lot more than most people. For example it’s not raining up here today but there’s zero chance of anything drying out. Most people going out for a walk would be happy with the ‘not raining’ bit and maybe look at cloud level, wind and visibility. We want to know how long it will take Earl crag to dry out when it’s raining.

spidermonkey09

Offline
  • *****
  • forum hero
  • Posts: 2814
  • Karma: +159/-4
#28 Re: Topic split: Stick to the Met Office
December 09, 2020, 12:13:33 pm
Met office abs MWIS will often say in their written reports that there’s uncertainty, particularly when there’s storm systems fucking shit up.

I think also as climbers what we want out of a forecast is a lot more than most people. For example it’s not raining up here today but there’s zero chance of anything drying out. Most people going out for a walk would be happy with the ‘not raining’ bit and maybe look at cloud level, wind and visibility. We want to know how long it will take Earl crag to dry out when it’s raining.

Unacceptable that the Met can't even tell us when The Ice World will be in good condition  :lol:


cheque

Offline
  • *****
  • forum hero
  • Posts: 3389
  • Karma: +522/-2
    • Cheque Pictures
#29 Re: Topic split: Stick to the Met Office
December 09, 2020, 12:23:22 pm
When people say that forecasts are shit are they not just judging them against a utopian ideal that’s never actually existed? They’re surely better now than they’ve ever been.

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#30 Re: Topic split: Stick to the Met Office
December 09, 2020, 12:25:16 pm
Quote
the consistent shitness/uncertainty of the forecast in what is fundamentally a pretty flat country is amazing.

It's really not, we're on a west coast island which is slap bang at the wavering latitude of the polar front, with a climate that should be much colder but is kept anomalously warm by a ocean current. It's one of the least predictable places in the world.

tomtom

Offline
  • *****
  • forum hero
  • Posts: 20282
  • Karma: +641/-11
#31 Re: Topic split: Stick to the Met Office
December 09, 2020, 12:32:32 pm
I've been involved in research looking at the effects of different rainfall products on predictions of flooding and river behavior etc.. and from what I remember...

So for Met office rainfall products there are:

For past data: At 5 min time slices, Rain Radar at 500m or 1000m grid cells (pixels) depending on how much you want to spend...

For the future: 'Nowcasting' that basically predicts where the rainfall will move and how it will change that goes forward up to 6 hours.

Then there is coarser resolution (4km I think) products from the High Res European model that goes forward 54 hours in one hour steps.

On some apps (I like snowradar for example) you can see the switch between the past (and near present) radar - to the nowcast (it seems to extrapolate where it moves to..) to the longer term model.

Rainfall associated with fronts passing is I believe more robustly predicted than showers - that may bubble up or form at more random locations - though the models often get the general area where the showers form right, eg orographic precip when air hits hills (NWales, W. Pennines good examples).

There are a multitude of ways of predicting where rainfall will go - and how much will fall. Many of which are based on some fairly heavy physics - and many of which are AI/Machine learning based.

spidermonkey09

Offline
  • *****
  • forum hero
  • Posts: 2814
  • Karma: +159/-4
#32 Re: Topic split: Stick to the Met Office
December 09, 2020, 12:47:15 pm
When people say that forecasts are shit are they not just judging them against a utopian ideal that’s never actually existed? They’re surely better now than they’ve ever been.

Yeah probably, recency bias might explain my feeling that they are more unpredictable now than a few years ago.


It's really not, we're on a west coast island which is slap bang at the wavering latitude of the polar front, with a climate that should be much colder but is kept anomalously warm by a ocean current. It's one of the least predictable places in the world.

I'll take your word for it, although my experience with Tasmania's forecast was amazingly reliable given the crazy and unpredictable weather they get down there (snow in summer on Mt Wellington!).  Although there are probably meteorological reasons for this.

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#33 Re: Topic split: Stick to the Met Office
December 09, 2020, 01:10:05 pm
Not having been I don't know the details, but it's worth noting Tas is on the same latitude as northern Spain. And with the Southern Ocean circulating a near circular polar continent (with Patagonia being the only significant land) the circulation of the south polar front is somewhat less chaotic than the north.

In my experience a lot of first world has a climate like Spain/ California and people regard any other weather like Tas, BC etc as terrible and unpredictable. Patagonia aside they've all been picnics compared to Scotland.

shark

Online
  • *****
  • Administrator
  • forum hero
  • Posts: 8697
  • Karma: +625/-17
  • insect overlord #1
#34 Re: Topic split: Stick to the Met Office
December 09, 2020, 01:12:44 pm
Just taken a look at Burbage. Everything wet including Remergence.

Thanks for tip JB/Cheque. Have downloaded the Met Office App. 🇬🇧

(🇫🇷  :rtfm: )

Johnny Brown

Offline
  • *****
  • forum hero
  • Posts: 11437
  • Karma: +690/-22
#35 Re: Topic split: Stick to the Met Office
December 09, 2020, 01:43:39 pm
Yeah, was hoping to get out too. Might have to make my annual trip indoors...

Nutty

Offline
  • ***
  • obsessive maniac
  • Posts: 359
  • Karma: +17/-0
#36 Re: Topic split: Stick to the Met Office
December 09, 2020, 02:32:40 pm
I would really like some sort of 'percentage uncertainty' metric on a forecast as currently I don't think there is a way to know this without being a meteorologist.

Meteoblue have a 'predictability' percentage for each day's forecast. I also like that they (and the Windy app) show the predictions of multiple models, helpful for quickly seeing if there's general agreement or not.

highrepute

Offline
  • *****
  • forum hero
  • Posts: 1288
  • Karma: +109/-0
  • Blah
#37 Re: Topic split: Stick to the Met Office
December 09, 2020, 02:38:36 pm
Meteoblue have a 'predictability' percentage for each day's forecast. I also like that they (and the Windy app) show the predictions of multiple models, helpful for quickly seeing if there's general agreement or not.

Meteoblue looks good.

duncan

Offline
  • *****
  • Global Moderator
  • forum hero
  • Posts: 2952
  • Karma: +332/-2
#38 Re: Topic split: Stick to the Met Office
December 10, 2020, 05:10:08 pm
A possible thread spit? "Come Rain or Come Shine" (1000s of other song titles are available)

Weather forecasts are hugely better than in the 80s or 90s. The UK, as JB says, is a fairly tricky place to forecast: exactly where that shower falls is critical for us but it might only cover a mile or two. Despite this, UK weather forecasts are much better than in California or the mid-west USA in my experience. The met office are one of the best in the business and surely have greatest expertise for forecasting in the UK. If I was only going to use one forecaster it would be the met office every time. However the correct answer is to look at all forecasts available and go with the most optimistic.


Jerry Morefat

Offline
  • **
  • addict
  • Posts: 140
  • Karma: +7/-0
#39 Re: Topic split: Stick to the Met Office
December 11, 2020, 09:58:37 am
There was a recent UKC thread on weather forecasts with some really interesting contributions from a Met Office forecaster but alas I can't seem to find it. From what I remember, UK weather prediction is a lot harder than most other places (as many users have pointed out) and the recent reduction in flights which collect observations has had some impact on forecast accuracy.

Although old, this article gives a good idea of how much forecasts have improved with time and suggest the ECMWF, not the Met office, provided the most accurate forecasts.

As an aside, the ECMWF is due to relocate from Reading to Bonn after Brexit. Another Brexit bonus  >:(

Fultonius

Offline
  • *****
  • forum hero
  • Posts: 4315
  • Karma: +138/-3
  • Was strong but crap, now weaker but better.
    • Photos
#40 Re: Topic split: Stick to the Met Office
December 11, 2020, 01:10:59 pm
Surprised nobody has mentioned Windy.com /  app yet.

While the Met Office arguably has a better model for the UK, it is based on the ECMWF model with a little of its own "special sauce" added in.

The nice thing about Windy is that you can hit "compare" and it will, side by side, show the predictions of the main 4 models:

ECMWF
GFS
ICON
NEMS (Meteoblue)

If the models are similar, you have higher confidence, if they're all over the place you can assume things are changeable. It's got a -1hr rain radar and can even do relative humidity plots, freezing levels etc.

And it's got a really nice interface.

tomtom

Offline
  • *****
  • forum hero
  • Posts: 20282
  • Karma: +641/-11
#41 Re: Topic split: Stick to the Met Office
December 11, 2020, 01:23:46 pm
Shark - can you insert a “it” before the “to” in the title please :)

AKA Mr Vegas

Offline
  • *
  • newbie
  • Posts: 1
  • Karma: +0/-0
#42 Re: Topic split: Stick to the Met Office
December 11, 2020, 10:19:54 pm
I can recommend xcweather.co.uk which usually seems pretty accurate. I remember laughing at 2:50 because it said that it would be raining by 3pm and there wasn't a cloud in the sky. Ten minutes later it was pissing down.
Also, raintoday.co.uk is a really good indicator of where it's raining and which way it's going.
Metcheck.com is a good site for percentage possibilities of rain too.

Paul B

Offline
  • *****
  • forum hero
  • Posts: 9626
  • Karma: +264/-4
#43 Re: Topic split: Stick to the Met Office
December 12, 2020, 05:40:06 pm
I like Ventusky for wind. Possibly because it's mesmerising to watch.

 

SimplePortal 2.3.7 © 2008-2024, SimplePortal