[Jerry Morefat]

What's view on metcheck Mr Morefat? I've always assumed they'd rely on metoffice data so it's just a rehash. Any knowledge about that? Thanks.

I could be wrong, but I think both the raw observations and models used by Metcheck and the Metoffice are different. I know very little about what Metcheck do under the hood, but a quick google search suggest they use the Global Forecast System (GFS). I think observational data from satellites run by the US's National Oceanic and Atmospheric Administration is fed into the GFS model. Whilst, I think observational data from EUMETSAT (a European organisation) satellites are used in the Met Office's Unified Model (UM). According to Wikipedia, the horizontal resolution of GFS is 35-70 Km, whilst the version of the UM used by the Met Office for the UK has a horizontal resolution of between 1.5 and 4 km. Based on this alone, I would expect the Met office forecasts to outperform Metcheck's.

[mrjonathanr]

Thanks for your considered reply. I'll be sticking with metoffice I think.

[Johnny Brown]

It's become a bit of a bug bear of mine, but these UK forecasts produced by random countries aren't more accurate

What Jerry said. The metoffice is one of, if not the best in the world. Just because yr.no and metguess produce some nice ways to present the data it doesn't mean they are worth looking at. My take has always been that these are big computers producing forecast data for massive areas. I don't think there is any human input - the classic example was when metguess continued to produce 'scorchio' signs just because we had high pressure in January. The reality was a week of gunmetal skies and freezing fog.

For anyone interested in weather the best thing you can do is to learn to interpret a basic synoptic chart. The big problem with modern forecasts is they are desperate to avoid this, which means they then have to commit to saying, eg 'it will rain in Manchester in 12 hours'. If you know its actually a cold front moving in overnight you can check its progress over breakfast live on the radar and satellite, and make your own assessment.

Forecasts are actually amazingly accurate nowadays, its just that trying to pin down exact times and locations like they do on the news or on yr/ metguess is false precision.

[dave]

Does the Norwegian met office come recommended then?

No. The Met Office forecasts will be more accurate for the UK. The met Office provides the BBC with their forecasts and so they are identical, if presented in a different way.

I was told a few years ago by the met office that the bbc get the same raw data from met office but they do their own analysis on it. I certainly can remember seeing bbc and met office giving wildly different forecasts in some instances.

[mrjonathanr]

For anyone interested in weather the best thing you can do is to learn to interpret a basic synoptic chart.

I can, they just don't seem to be displayed nowadays.  It's like some weird satnav mentality has overpowered all the meteorologists.

This is all I can find. Maybe I'm missing something.

[tomtom]

I use an IPad/Phone app by MeteoGroup which gives the best nowcasts/6-12 hour predictions I've seen.... The rain radar data is more up to date (1/2 hour delay) and has precip type..

I've a bit of a soapbox to clamber on to here - as the Met office is uber stingy with its data (it now releases 'some' for free) when its paid for by the tax payer.. Outrageous. We pay twice.

[Jerry Morefat]

I was told a few years ago by the met office that the bbc get the same raw data from met office but they do their own analysis on it. I certainly can remember seeing bbc and met office giving wildly different forecasts in some instances.

Are you sure? Which part of the Met Office did you talk to? What did you say exactly? According to a Met Office Spokesman from January 2010, " We have always been in the strongest position to provide the BBC with accurate and detailed weather forecasts and warnings for the UK." This suggest to me that the met Office did provide the BBC with their forecasts, up until Jan 2010 at least. In fact, I think the weather presenters on national TV are all Met Office employees.

I've a bit of a soapbox to clamber on to here - as the Met office is uber stingy with its data (it now releases 'some' for free) when its paid for by the tax payer.. Outrageous. We pay twice.

I don't think this is quite true. As I understand it, the Met Office is a trading funding and as such doesn't receive any money directly from the Treasury. All their money comes from contracts with their 'customers'. Admittedly some of these customers, such as the BBC and MoD are funded directly with public money but other customers are private companies.

  I realise this is all very  .

[Johnny Brown]

I can, they just don't seem to be displayed nowadays. 
This is all I can find. Maybe I'm missing something.

Hmm, you've lost me anyway - those are the proper synoptics, added to the new metoffice site after whinging (including my own) about having to get them via german sites. The old site is still available with the colour versions, though you lose the geostrophic wind: here.

I use an IPad/Phone app by MeteoGroup which gives the best nowcasts/6-12 hour predictions I've seen.... The rain radar data is more up to date (1/2 hour delay) and has precip type

I use raintoday.co.uk for radar, and sat24.com for satellites (though the daily shots from NASA's modis are always worth a look, frequently amazing). Not sure who operates the radar network but the satellites are not owned by the UK obviously, wouldn't be surprised if the radar data is similarly bought in by the met office.

Be good to split this topic off mods - a  weather resources thread would be useful to me at least.

[SA Chris]

Hmm, you've lost me anyway - those are the proper synoptics, added to the new metoffice site after whinging (including my own) about having to get them via german sites. The old site is still available with the colour versions, though you lose the geostrophic wind: here.

Be good to split this topic off mods - a  weather resources thread would be useful to me at least.

Not sure how much longer the old site will still be available though; I suspect the plug will get pulled eventually.

I like the pressure chart on MSW

http://magicseaweed.com/North-Atlantic-Surf-Chart/2/#?chartType=PRATE&_suid=1351682118665028579421836669066

but not sure where their data is sourced from or if it is beter or worse than MO.

Agree this needs splitting off.

[dave]

I was told a few years ago by the met office that the bbc get the same raw data from met office but they do their own analysis on it. I certainly can remember seeing bbc and met office giving wildly different forecasts in some instances.

Are you sure? Which part of the Met Office did you talk to? What did you say exactly?

I emailed them about it years back, probably 2003 ish. No idea about the exact content, it as on my old work email. Basically I was wondering if they were providing the BBC forecasts then why the BBC said we were in for 5 days of solid rain in Sheffield, but metoffice said we were in for 5 days of solid sunshine. The impressions they gave was that the BBC's analysis was less sophisticated to theirs. Of course this could well have changed in the last ten years.

[tomtom]

They only recently set up as a trading fund... most of their $$ came/comes from UKPLC... You cant fund those swanky supercomputers from iphone apps

[andy_e]

I'm up for learning Johnny. What does the predicted occluded front over the UK on Saturday mean? Does it mean dry air is coming in? Will I climb 7C?

[tomtom]

7C+ if you touch rock between 11:15 and 11:23, with a pine cone in your back pocket - and three cows are lying down in the field next door (in the NE corner of course)

[SA Chris]

Crookrise then.

[Johnny Brown]

I'm up for learning Johnny. What does the predicted occluded front over the UK on Saturday mean?

Judging by the fact you know the terminology, and you have a geology degree (so presumably geography A-level), it means you are taking the piss.

[andy_e]

No, I've forgotten my A-level geography. I can remember the terminology but not what it actually means...

[slackline]

If you know its actually a cold front moving in overnight you can check its progress over breakfast live on the radar and satellite, and make your own assessment.

I find raintoday.co.uk useful for tracking the movement of rain as its based on the last few hours of (radar) observed rainfall.

[Duma]

I use the one that says it'll be dry.

Actually the ones I like:
raintoday - for next few hours
metoffice - forecasts beyond that
msw - pressure and wind charts

I also check bbc, yr.no, and metcheck, just on the basis that the divergence between the different models gives me some idea of how uncertain the forecasts are likely to be.

[fried]

On a Font note, I've found.

Accuweather (very pretty android widget) to be completely inaccurate.
France.Meteo is very good but with an awful display.
Yr.no can tell me when it's going to rain in the next hour to the minute, it's always wrong (when i've used it)

I'd love to find a widget that can display the france.meteo raw data in some form that doesn't hurt my eyes.

[Bonjoy]

I mostly rely on metoffice (always make sure you look at the update times, they can vary a lot between different sections of the site which can explain contradictions).

The only unmentioned site so far I rely on a lot is buxtonweather.co.uk. Partly for the webcams but also for the rainfall records (for Buxton obviously) which give a good idea how much has landed on the Peak in the last few days/weeks. This is especially useful for gauging how bad seepage is likely to be.

[mrjonathanr]

I rely on a lot is buxtonweather.co.uk. Partly for the webcams but also for the rainfall ... This is especially useful for gauging how bad seepage is likely to be.

Previously unknown to me. Quite handy some of that detail

[rich d]

The most accurate predictor I find is my diary. If it's a day off work and I have no family/social commitments it will be raining in the peak. This over the last 2 years has had a 97% accuracy of forecasting, which I believe is higher than the met office. I can link my diary for free=rainy days to the UKB Wiki if the mods like so you too can plan when to go to the works/depot.

[tomtom]

The most accurate predictor I find is my diary. If it's a day off work and I have no family/social commitments it will be raining in the peak. This over the last 2 years has had a 97% accuracy of forecasting, which I believe is higher than the met office. I can link my diary for free=rainy days to the UKB Wiki if the mods like so you too can plan when to go to the works/depot.

:D

[nai]

The most accurate predictor I find is my diary. If it's a day off work and I have no family/social commitments it will be raining in the peak.

That's pretty much how it's worked for me this year too 

[Fultonius]

YR.NO is corrected by humans, but it's not clear if that's just for Norway!

I find it is fairly accurate, but often check others to see how well the models match. If they're the same then I feel more confident in them, if they differ then I know it's a low confidence day. I wish they would state their confidence... (one thing I like about chamonixmeteo.com...)

Weather forecasts on yr.no are based on data provided by Norwegian Meteorological Institute and its international partner institutions such as European Centre for Medium-Range Weather Forecasts (ECMWF) and European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT).

Not entirely clear on their resolution - I reckon it's 10-15km as the consortium still seem to be "developing" a 2.5km model.