Met Office loses BBC weather forecasting contract after 93 years | Hacker News Reader

Met Office loses BBC weather forecasting contract after 93 years

theguardian.com

109 pointsedward83 comments

I've lived in the UK for the past 8 years now (having lived in various places in the East Coast USA for the rest of my life).

I learned quickly that English weather forecasts are useless past the current day. It is probably due to the geographical location. East Coast US was accurate to 3 to 5 days. I just use Dark Sky[0] to get weather for the next several hours, which is mostly accurate.

Given the situation with the contract I see lots of back and forth noting accuracy - but Met Office does just as good a job as anyone else, for probably a much higher fee.

[0] http://forecast.io

Yup - you are somewhere on a fairly narrow island with warm water flowing South to North and a cold wind at high altitude blowing West to East (or is it East to West? I always get that mixed up).

It is going to be slightly changeable.

I'm assuming the Met will still bid for the contract, and we shall see who gets it. Pity them the first time they get it badly wrong!

The met has bid for the contract, and lost the bid. The BBC and the met have both independantly confirmed tha. The met office isn't going to be providing weather data to the BBC.

Some of the staff (the presenters tended to be real meteorologists workig for the met office) might transfer over; severe weather stuff is still going to come from the met office; the Shipping Forecast is not a met office thing so should be unaffected.

> the Shipping Forecast is not a met office thing

Isn't it? It always starts with 'And now the shipping forecast issued by the Met Office, on behalf of the Maritime and Coastguard Agency, at...'

Ah, yes. Sorry.

The met office provide the data to the Maritime and Coastguard agency, who then provide that to the BBC. The presenters are BBC staff, not Met Office staff. So there's no change to the Shipping Forecast.

http://www.telegraph.co.uk/news/science/science-news/1181929...

I stand corrected, thanks.

http://www.telegraph.co.uk/news/bbc/11819164/Met-Office-lost...

Can this really be about an app and about including probability statements? Good heavens.

> The met has bid for the contract, and lost the bid. The BBC and the met have both independantly confirmed tha.

Where does it say this, please? All the articles I have found say that the contract will be put out to tender, but none actually say that the Met Office will be prevented from bidding.

http://blog.metoffice.gov.uk/2015/08/23/met-office-in-the-ne...

> You may have seen reports in the media today that the Met Office won’t be the BBCs main weather provider when the current contract ends.

> Obviously everyone at the Met Office is disappointed that we won’t be supplying weather presenters and graphics to the BBC in the future.

http://www.bbc.co.uk/news/uk-34031785

http://www.theguardian.com/uk-news/2015/aug/23/met-office-lo...

Someone from the Met Office was on a bunch of Radio Four news programmes (probably Today, World at One, PM) saying that they've lost the contract.

That's fine, but where has anyone said that the Met Office can't bid in the upcoming tender? The Met Office blog post that you linked (thanks, I hadn't seen that) implies that they will no longer be providing services, but is this because they will decline to bid or because they are not permitted to bid?

Agreed, many dynamics as far as UK weather concerned and probably the hardest country/area to do a forecast due to those dynamics.

Indeed even the Guardian reports this http://www.theguardian.com/uk/2011/sep/03/uk-weather-defies-...

> East Coast US was accurate to 3 to 5 days

Interestingly in the US, the East Coast forecast is more accurate than the West Coast.

This is because the trade winds in the US mostly move West to East, so most of the Weather moves in that direction as well.

There are more weather sensors on land then in the Sea, so there is more data available to make a forecast with.

And, up until the last year or so, there were parts of the west coast (basically, off the olympic peninsula) that didn't have any offshore radar coverage. So there were certain classes of dangerous storms that could sneak up and whack Seattle. (One of the storms that did in a floating bridge did this, basically, no one knew that there was a hurricane force storm until it hit the coast, and from then, it was an hour to Seattle)

I would be interested to see a global heatmap of weather predictability.

As a life-long east coaster, I never understood the frequent grouses one hears about weather reports being inaccurate. I consider them to be quite good given how much inherent chaos is in the system they're trying to predict.

It's not just about the number of sensors. The clouds form over the sea. Predicting when and where those clouds will form is much harder than predicting when and where those clouds will move.

Clouds form over land too.

When warm wet air meets cool air you get clouds. The goal is being able to measure the humidity and temperature of those colliding air masses.

One needs to account for relative humidity, which the ocean provides. Much of the US landmass between the pacific crestline and the rocky mountains is arid desert (from the canadian border to mexico). Second, there exist ocean-specific dynamics like el-nino, which don't really have an overland anolog.

> This is because the [westerlies] [...] move West to East

FTFY

It's probably worth noting that Forecast.io and Dark Sky use the Met Office's radar, prediction system, and severe weather warning system (in conjunction with other data sources) to create their forecasts.

Just yesterday I was at a garden party where they predicted rain from 11am to 8pm, BBC predicted it would stop at 3pm, and that's what happened. Very nice web app, but in my experience Forecast.io's predictions aren't very good.

My old boss was a South African expat who had also lived in the UK for a little while. Her comment on the two weather services: In the UK, they'll predict it will rain tomorrow at 3pm for 15 minutes, and it does. Whereas in Capetown, you can actually physically see the storm approaching over the sea, and they'll still be forecasting fine weather...

I've lived in the UK my whole life and was not even aware forecasts could be reliably accurate so far in advance!

I've found that Weather Underground[0] seems to be more accurate than most - not sure if this is actually the case or it's just a perception thing though (I believe they "crowd-source" the sensors somewhat, in that people can install relevant hardware and software at home to provide meteorological data, and also the app asks for feedback if the current weather data is correct or not)! I trust it enough to decide whether to bring a coat or not for that day - as others have mentioned, I had no idea the relative accuracy of forecasts in the UK is lower than elsewhere, interesting!

[0] http://www.wunderground.com/

I half-jokingly tell people that if they get their forecast from the BBC (via the Met), they should expect the inverse weather.

I wonder if the BBC has checked the quality of the forecasts based on what actually happens? I use AccuWeather and the Aix Weather Widget (for Android) because they're nearly always right (where I live, at least) - and gives me much more useful information.

For example, I can't translate the Met's '50% chance of rain' - for a whole day - to real life. To me, that means it could not rain, or rain heavily all day, or anything in between.

I actually learned that the rain percentage is not the percentage chance that it will rain but is in fact the percentage of the coverage area that they will expect to get rain. So 50% rain meaning no rain at all or rain for the whole day is entirely reasonable.

That's funny. I had long suspected that was the case but never knew for sure. It's also important to know where the weather station is for the forecast area too -- here they can cover relatively large areas with significant elevation change, which can greatly affect things like temperature and wind speed.

Reasonable unless you want your forecast to be at all useful!

How does that work when the percentage is supposed to cover a small area, like a town?

FWIW I've found that combining the iPhone weather app with a rain radar app gives me usefully accurate predictions of rain.

The Met Office is useful for coarser extreme weather warnings, but I don't pay much attention to its daily forecasts.

I've always interpreted it as the percentage of time it will rain, i.e. if it says 50% chance from 7 to 8, it will rain during about 30 minutes of that interval and not rain during the other 30 minutes.

(equivalently, you could say it's the % chance that it will be raining if you choose a given instant X in the interval).

I've never read an official interpretation but that seems to fit reality (where I live) quite well.

50% means that they ran the simulation say 10 times and in 5 of those simulations there was rain in the area. How to translate that to real life - you should probably carry an umbrella if you don't like getting wet. Also interesting to note that weather presenters usually bump up the probability of rain by 10-20%, because people complain about false negatives (rained when it was predicted not to) rather than false positives (didn't rain when it was predicted to).

If you're interested in a good intro to this topic I recommend the chapter on weather prediction from The Signal and the Noise by Nate Silver. Its a fascinating read.

Yes, it's the probability of rain.

Not quite:

http://www.srh.noaa.gov/ffc/?n=pop

Interesting that it only measures the chance of "measurable rainfall" and does not attempt to predict the duration of the rain.

http://www.metoffice.gov.uk/news/in-depth/science-behind-pro...

First ghit. First duckduckgo hit.

And this matches what I thought it meant. It doesn't make it useful to me in real life, though.

The accuracy?

http://www.metoffice.gov.uk/about-us/who/accuracy/forecasts

There are local variations which mean for example, that if you live in the welsh valleys, someone on one side can be having sun, while on the other side (and in view) there is rain.

The met office does forecasts for a broad area, if you want hyperlocal, wunderground is where I go. The met office could also run forecasts for each london borough, but I doubt you or anyone else would sit through a hyperlocal forecast on the news.

I've found the opposite. Android/Yahoo/Google weather are generally telling me the wrong thing. I've come to rely BBC weather.

Rags like the Daily Mail are FUDding the hell out of this story. The Guardian's take is at least a bit more balanced.

Personally, I think it's a good thing. I like the Met Office (and in the interests of disclosure have worked with it on a couple projects) but it shouldn't have an entitlement to providing weather data for the BBC.

I'd like to see something like MeteoGroup get the contract. Would be really interesting to see what it could bring to the table.

True, although they fail to mention the quoted MP, Ben Bradshaw, is the 'Minister for the South West' and MP for Exeter where the Met Office is based (and employ 700-800 people)

I've been using the Met Office Android app for some time. The app could use polish, but the data is superb. If you want to know if it's going to rain in the micro-region where you are now, there's nothing like it.

Edit: It's this one: http://www.metoffice.gov.uk/services/mobile-digital-services...

I'd argue Weather Pro by MeteoGroup is far, far superior. The only thing I find the Met Office app useful for is additional detail on severe weather warnings.

Thanks for the tip. Not that it makes a big difference, but I'll just note that the MeteoGroup app is a couple of quid, but the Met Office app is free.

Yeah, to be fair that is worth noting. But I find that the increased accuracy, access to data and functionality makes it worth it.

Last few versions on Android they've totally broken from a usability perspective IMO:

* Yellow location search button is now in the way

* Moving the menu to the top means you can't use it one-handed now

* The most useful thing in the app for me (rain radar) now takes two presses to get to, instead of just one

I'm not convinced the accuracy is any better than the Met Office app - the only thing for me which made it more useful was the rain radar images which at least give you some heads up as to rain...

To be fair, the Met Office app has usability issues too. Basic stuff like zooming / panning the map is hit and miss.

As a photographer, I try to follow up weather when I'm out and about. As many people have reported, the met office is next to useless, even on the short term!

One I've found works /wonder/ is the Norwegian weather service at http://www.yr.no/ . I'm not kidding, these guys not only give pretty accurate forecast, but they over england and they provide it in neat meteograms that are quite precise and easily read.

Now, how to explain that yr.no can predict weather in the UK and the met office can't... Met office has FUDed the country with "Oh but it's difficult in the UK, we're and island blah blah" but clearly, that's bollocks.

Why does the met office get international recognition for accurate forecasting if it's so useless?

http://www.metoffice.gov.uk/about-us/who/accuracy/forecasts

I don't know, but I second the recommendation for yr.no. As a Brit, I find the Met Office to be more inaccurate for 3-5 days ahead than yr.no. On the day itself the Met Office isn't bad, but I look at http://raintoday.co.uk instead to time my trips out between showers. Perhaps it comes to timeframe, or vagueness.

My anecdata is the exact opposite. I find the met office very accurate.

One thing I have noticed is that the idiot glyphs they use don't tie into my expectations. I see a picture of the sun behind a cloud with a rain drop on it and I think "light showers". It actually means "40% chance of rain". A cloud with a rain drop and no sun still only means that there's a 50/50 chance of rain.

Next up, Privatisation of the Met Office then.

Shame as while the BBC websites presentation of the data is pretty poor the MET provide open data feeds (hard to see that continuing if they are privatised) of 3 hourly forecasts out to 5 days for 5000 locations that are considerably more accurate and detailed.

http://i.imgur.com/vuCklmn.png is their example usage of the data.

That sounds like a slippery slope argument. You might as well worry about the privatization of the NHS next. With this type of approach, arguments degenerate into political partisan rhetoric and achieve nothing.

Shrugs, in theory you could apply slippery slope argument to just about any conjecture on what will happen next.

However this Government has a track record of (badly) privatizing things and a fairly clear ideological bent.

> You might as well worry about the privatization of the NHS next.

There's quite a lot of private provision already in the NHS.

Forensic Mental Health Units tend to be privately run hospitals, especially the medium and high secure units. There's a few privately run in-patient units treating eating disorder (especially for children). Drug and alcohol rehab is often provided by 3rd sector groups.

It's all free at the point of delivery, and paid for by the NHS, but there's plenty of private provision within the NHS already and it's only going to grow.

I doubt that, it's part of the British Army

It has historical links to the british military, but it's part of BIS (Department for Business, Innovation and Skills).

http://www.metoffice.gov.uk/about-us/who/management

Here's my prediction:

30% chance of more than 1mm of rain tomorrow.

Just like every other day.

https://en.wikipedia.org/wiki/Climate_of_the_United_Kingdom

    Month 	Jan 	Feb 	Mar 	Apr 	May 	Jun 	Jul 	Aug 	Sep 	Oct 	Nov 	Dec 	
    Rainfall ≥ 1 mm 
    days 	13.2 	10.4 	11.5 	10.4 	9.9 	9.6 	9.5 	9.9 	9.9 	12.6 	13.1 	12.7

I'm a techie and I've been using the Met Office service more recently as part of obtaining my private pilots license.

Before I get to their ability to forecast correctly I need to say that their tech appears disorganised to say the least. Check out this garbage: http://www.metoffice.gov.uk/aviation/ga. Specifically the text under General Aviation services. They attempted to release a new version of the General Aviation interface, sending all the pilots at my local club in to a tizz. They fluffed the launch of this new interface so badly they were forced to hit reverse and reopen the old app. Looks like you can still register for the new app, but they're now supporting and maintaining the old app too. Oh and to use the new app you need to create a new account...I've no idea why.

If this is at all representative of the overall state of the Met Office websites and systems I'm not surprised they're not the first choice for the BBC.

Ok so that's not great. What about the weather predictions? Well, accepting that we live in a somewhat changeable part of the world here in the UK, and given that I'm not a meteorologist take this next bit with a pinch of salt. I was due to fly at 1pm today. I shit you not the Met Office website stated severe weather warnings with heavy rain due to hit at 3pm. I checked the forecast all morning and rain was pretty much guaranteed at 3pm. The forecast was so bad I warned my other half about it as she was due to drive long distance. 1pm rolls around and it starts raining lightly. Okay - this must be the start of the heavy rain, so I write off my flight. 3pm comes around and I shit you not, bright sunshine and it's stayed bright and sunny ALL afternoon. Literally the opposite of what was forecast. Apparently the heavy rain is now due tomorrow.

Obviously the Met Office site now shows sunny with clouds. Thanks Met Office, how very kind of you to update the weather forecast to match what you can see out of your windows.

I know nothing about the Met Office, but sometimes inaccurate weather like that happens because the bad weather moved elsewhere, or because it was in one part of the forecast area, and you were in another.

Next time it happens find a site like http://weatherspark.com/ and check the entire area and see if the bad weather actually is there, just not where you are.

Have you tried accu weather? I typically find them to be most accurate, and don't bother with the bbc's.

No I haven't I'll definitely give them a try. Thanks :)

This is a fab site to use in conjunction with predictions, as it provides an animated radar rainfall map of the past few hours: http://www.raintoday.co.uk/

Interesting that New Zealand's MetService/Metra is in the running. New Zealand is an island like the UK, so weather forecasting is difficult here too. MetService currently supplies the BBC's weather graphics.

http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&o...

I couldn't find anywhere that said the BBC were dropping the Met Office. Is this FUD / standard media hype? From what I gathered the BBC has decided to put forecasting out to tender. To which, the Met Office is very welcome to bid. It may end up meaning that the Met Office is gone, but not necessarily.

So while the Guardian headline is strictly true, some bits of the report make it sound like it is a done deal.

This may be of help - http://www.bbc.co.uk/news/uk-34031785

Will the main requirement of the tender be being at least X% more accurate than the current provider?

Or will it just be a matter of who will offer the cheapest service?

If weather predictions are basically fiction (which they seem to be), then we may as well go with the cheapest fiction.

Weather predictions are not fiction. Some businesses pay millions for a specific weather forecast (one that could dramatically influence the profitability of a venture) in order to enable better planning. A good example of the importance of weather and climate forecasting would be the Climate Corp sale to Monsanto for ~$1B.

Predicting the future is really really hard. Over the last few decades we've made a lot of progress and we're about to get a lot better, too.

I'm sure the £4.5M bonus pool paid for 'improved performance' really stuck in the craw.

http://www.independent.co.uk/news/uk/home-news/met-office-st...

Indeed and also no details upon who will now run the contract is concerning in ommision and somewhat telling. So fun times ahead.

There is another subtext: the current government is on a privatization drive. Met Office is a prime candidate for being spun off, since it's already trading. Making them lose one of their biggest customers will force them to find new ones (readying MO for full privatization) or failing that, put their finances in an unsustainable position (readying MO for full privatization or even shutdown).

The BBC has many ongoing fights with the current government, so this was probably a relatively easy wish to grant in order to build back some favour. Among other things, it could end up reducing expenses for them, so it's a win-win as far as the BBC is concerned.

The met office is already a "trading fund" - which means they're required to run on a commercial basis.

http://www.metoffice.gov.uk/about-us/who/management

The BBC was not a huge amount of money for the met office. They have plenty of other customers. (Sky, ITV, Channel 4, are other media companies.)

Being a trading fund only means that trading receipts must account for 50% of revenues. It says nothing about where those revenues must come from, nor where continuous capital investment must come from.

Their revenues are still 85% from government contracts (bid and no-bid) according to 2013 accounts [1]. From a cursory read, I'd say it cost the government £16m in 2013 and produced £12.3m in profit, with more than £170m of revenue coming straight from governmental contracts and £30m from commercial ones.

I can see people looking forward to cutting £16m from the yearly budget and force all internal contracts to be bids, lowering that £170m bill. It makes an awful lot of sense, from a certain perspective.

[1] http://www.metoffice.gov.uk/media/pdf/n/e/Annual_Report-web....

Not sure if available worldwide, but there was a very interesting Radio 4 program recently, "What's the Point of The Met Office?", asking all sorts of awkward questions of this venerable institution.

http://www.bbc.co.uk/programmes/b06418l5

Of particular interest is the Met Office's constant demand for bigger more expensive computers, and its recent moves into the more politically contentious territory of climate change policy.

>Of particular interest is the Met Office's constant demand for bigger more expensive computers, and its recent moves into the more politically contentious territory of climate change policy.

It doesn't seem that interesting to me. One of the reasons supercomputers are built in the first place is for studying weather and climate [0]. And "politically contentious" or not, climatology is a reasonable thing for a meteorological office to study. For example, the National Weather Service in the United States does both.

Also, the Met Office has been involved in climate research [1] for a long time, so it's hardly "recent."

[0]: For example: http://news.nationalgeographic.com/news/2005/08/0829_050829_...

[1]: http://www.metoffice.gov.uk/climate-change/resources/hadleyc...

I take it you didn't listen to the programme.

No. What is it about how the Met Office studies climate change and requests new computers that is interesting? They both seem self-explanatory, so it didn't seem like listening to a 28-minute program would be worthwhile.

I did. It was mostly horseshit. If there's anything you think is particularly important you're probably better off pulling put those points and presenting them independantly. There were so many innaccuracies in the program that usin it discredits any point you want to make.

> Of particular interest is the Met Office's constant demand for bigger more expensive computers, and its recent moves into the more politically contentious territory of climate change policy.

I am confused at why this is contentious, this is exactly what I would expect them to do?

Its not contentious. The program was by an anti science climate change denier and was a total pile of shite.

> anti science climate change denier

Not everything is quite so black and white. You shouldn't be so quick to divide people into "goodies" and "baddies". You can be absolutely against spending public money on climate change research, whilst not being 'anti science'.

Quentin Letts is a climate change denier.

http://www.theguardian.com/environment/2015/aug/05/whats-the...

Czech MET lost contract with major televisions long time ago. Not much has changed, except science is now sponsored by state, not TVs.

Gotta love this:

1/Extremely alarmed by rumours BBC to drop UK Met Office in favour of foreign weather forecaster. Vital 90 year old strategic relationship.

2/Not only weather forecasting. Shipping forecast, extreme weather/crisis events/national defence. Madness if true. Govt must intervene

No, government doesn't have to intervene. Silly man, competition drives quality. We don't need a governmental lid on quality.

Reply on news.ycombinator.com