Met Office UK forecasts - hedging their bets?

[Beadle]

To be fair in needs to be read in conjunction with the general forecast, theres a bit more going on than just a change in wind direction, look at a pressure chart and the forecast might make sense.

Assuming you were reading from Berwick on Tweed to Whitby forecast

Yes quite right

I thought the point was to give us a good idea of what to expect without having to do that depth of analysis.

I could calculate/guess that something unpleasant may be in the offing but I would hope for something a little bit better than I can do.

 

[dunedin]

I suspect the Met Office figure was a typo, if you still have the hard evidence do us all a favour and send it to the Met Office, they will investigate your complaint and reply with an explanation.

I would be surprised if it was a typo - as I assume the forecasts are all automated. I didn't bother to save that version. A later version had narrowed to 2mph - 26mph, which is the sort of thing often seen recently.
Agree dont shoot the messenger for bad weather. And "F3-4, possibly F5-7" could be reasonable for a 24 hour inshore waters forecast, as wind does change during the day. But F1 - F9 equivalent is a bit wide range for a forecst covering a 3 hour period and a narrow locality

 

[bilbobaggins]

'A short study of forecast performance'

Just over a decade ago, a conversation with a prominent Chartered Meteorologist at a boring RYA drinkies party uncovered the fact that the Met Office published no data on how accurate their forecasts were. It was presumed that they generated such stuff internally, for their senior officers were paid considerable sums for achieving certain performance targets ( I seem to remember a certain Chief Executive being required by the National Audit Office to repay chunks of his bonus when it was discovered he'd been 'revising' the figures in his favour ), but nothing was made available to the public.

The upshot was that said met-man collaborated on a survey of the accuracy of the Shipping Forecast. The historical data was winkled out of the Met Office Library, and he'd determine what was a 'hit' or a 'miss', in terms of what use we yotties might make of the Friday night SHIPPING FORECASTS for sea area Portland, and the subsequent ACTUALS reports from the Channel Light Vessel Automatic. The same was done for the Saturday night forecasts, and the following Sundays' reports.

The results of analysis were published in Practical Boat Owner 355 of 1996 as ‘A Short Study Of Forecast Performance’, with the objective of providing we leisure yachtsmen with an indication of likely accuracy, which might aid our decision-making. There was then a minor storm - a 'flurry of forecasters' - for the Met Office's view was that any analysis of THEIR data was impudent, intrusive and certainly wrong. The fact that WE pay for their stuff seemed irrelevant. PBO was forced to print a 'qualifier' statement provided by said Met Office, which effectively 'poo-poo'd the survey as incompetent and misleading.

The article was later printed in the magazine of the Royal Meteorological Society, ‘Weather’, where it was rather more warmly received.

I'm told by the quite-prominent Chartered Meterologist that he was approached some years later by the Met Office's then Most Senior Forecaster, who apologised for the unprofessional treatment of my friend and his survey, which was done not by a forecaster, but by a marketing wonk employed by the Met Office.

Perhaps it's time for us to revisit this survey, to see what changes in performance have occurred……

 

[Blue5]

Cannot comment on the set up for marine forecasts from the Met Office but was involved in a working group dealing with aviation forecasts. Basically the Met Office were paid by CAA to provide a forecast that provide certain criteria and the amount they are paid dictates how long the forecaster is allowed to spend preparing the forecast. On a yearly basis we had a meeting with Met Office to discuss information included and possible alterations (wish list) but the overiding factor was that Met Office were following their remit to CAA.

Met Office is a business and the bottom line is you can have as much information as you want so long as you pay for it, hence, phone the forecaster and the birth of independant forecasting companies like Simon Keeling.

One thing to remember before shooting all the forecasters is that we sit in an area where the weather is very fluid and very difficult to pin point to the accuracy demanded, a bit different to sitting in the middle of a big high pressure for several weeks.

 

[smartcom]

Shipping forecast sources

As I understand it, the MCA commission trhe Met Office to provide the shipping and inshore waters forecasts.
The RYA, and other interested bodies, feed in to the MCA what the user requirements are for the weather services.

 

[smartcom]

Something like this was done recently - a paper was presented at the Royal Inst. of Navigation's Weather and Sailing seminars a couple of years ago.

I can't remember the exact details but I think it was done somewhere around Exeter, and the locations chosen for measuring the actual weather were poor choices, subject to very local meteorological effects, so no meaningful results were produced.

Interestingly, that year the event was held at the Met Office in Exeter. Only a couple of Met Office people deigned to attend the event, and they had a junior PR person do a tour of the establishment and the poor girl got so much flak about things she didn't really know about, she just knew what she had been told to say in her training.

By the way, those events are well worth attending if you want to keep up to date with weather and its uses in sailing/boating. From memory I think the next one is early next year - have a look at the RIN web site for information.

 

[Cantata]

Thanks for confirming that. Still collecting 'evidence' before going into battle!

 

[richardbrennan]

The current five day forecast for Cowes gives the wind on Tuesday as 5mph gusting 30mph!

 

[neale]

We need to know what the conditions will be for the next 30 miles, a typical day’s run.

We don't all do 5 knots you know

I would certainly need to know the weather for the next 100 - 150 miles.

 

[SmileyGiley]

The possible gusting strength, as opposed to the forecast average strenght seems to be a newer phenomenon. I think it's all down to statistical modelling on those electri-histrionic slide rule thingies.

For several years I was heavily (oh, how heavily the time passed) involved in positioning of seismic streamers. The basic idea is that you make a load of measurements (gps marker buoys, acousic ranging systems, compasses strapped to the cable) and feed it into a whizz bang computer. It makes a model, and then sees how accurate that model is 10 seconds later when it does all the measurements agai. We used to get stats very like the weather forecast, except a little more believable. However one set of figures was the 95% likely hood position. That was generally pretty tight and pretty believable. However, there was also a maximum statistical probability of incorrectenss figure ... which basically said.... it's here... within a few metres... BUT i could be 50 m out. These were simply an analysis of the robustness of the model. The possible error figures really coudn't be believed, because the obvious impossibility of x being that far from y. SO.... I think these are just the possible maxima to the model. The weather is WAY more complex than the stuff we were measuring, so although they can say forecast 3-4 (of stuff it, lets say 3-5 to be on the safe side) they can also see a theoretical error in theri model up to 150 kts. OK... well we better bung that in too.

Believe me... no nor do I really, but thought I'd have a rant. The alternative is work...