Weather Forecasts - the reality

I think that the Channel is extremely well provided with weather forecasts compared with other parts of Europe. In many harbours in the Baltic there is no internet access, and if you don't have a local bank account, getting a forecast via a mobile phone costs a fortune. There is very little on VHF. Local harbours may have a piece of paper pinned up with a page from an internet weather site. Navtex is little use, since the forecast for the day seems to arrive at about eleven o'clock in the morning.

And Frank is right: in the present state of the art weather forecasting is still very imprecise. And then you must allow for local factors: the wind round St Catharine's point may be very different to that off Cowes.

It would be very nice to have a forecast which says: 'As you leave the Solent via the Needles at 6a.m. the wind will be 12 knots from the SW; at midday in the middle of the Channel it will be ...', but you're not going to get it. GRIB files are all very well, but until we understand the physics of the atmosphere better, and develop powerful enough computers, then you're going to have to accept that what you get is a forecast, and that forecast may not match reality.
 
Robin said:
There are two different questions in this discussion in my view, firstly the accuracy of the forecasts and secondly the availability of them.
Click to expand...

I would agree that the availability of forecasts, shoreside vs afloat has become blurred in recent years. Internet forecasts, whether form the met office or other sources are readily available in phone (GPRS) range through various items (ask for one of these for Christmas, they're brilliant and cheap to run)

The typical UK sailor probably finds the inshore waters forecast of more relevance than the shipping forecast, for no other reason than the granularity issue raised above. In VHF range this is available at far more sociable hours than the R4 shipping forecast transmissions.

Sometimes, I find internet forecasts to be somewhat overmodelled. If the inshore forecast says "unsettled, with a brisk to strong SW flow" then that is better, or at least more honest and accurate than someone trying to predict exactly where and when a F5 changes to an F6-7. Having said that, I am interested that the professionals (Frank and Simon) seem to agree on the need for a broad base of sources, and that is isn't always transparent. Personally I find that apart from the ubiquitous grib (gfs), WW7 (via windguru) is more detailed and Theyr (Halo, as it used to be) seems to be derived from a different model altogether. I like web sites like weatheronline, because of the personal interpretation but a lot of the other websites seem to be little more than regurgitated, low resolution gfs data.

I think that as people requiring a professional and technical forecast, but with many of our number lacking the knowledge to understand it, we are never going to be a happy bunch of bunnies if the forecast is not for good weather.
 
Very interesting debate. Weather forecasters are on a hiding to nothing. Short term forecasts do not predict what we see out of the window because the modelling is not sufficiently detailed so scorn is poured upon them. In truth, in the near future it is unlikely to get much better until fundamental concepts - such as turbulence - that can affect weather on the local scale are understood and modelled. However, the forecasts are reasonably accurate given the variability of UK weather. Long term forecasts are another matter. If the weather is chaotic then the maths shows that accurate forecasting will not be possible no matter how much computing power is thrown at the problem. As a taxpayer I would prefer we didn't waste money on ever more expensive computing systems chasing the unattainable. Until proof is provided that weather is not a chaotic system we should just accept the current limits to forecasting and the Met. Office should stop chasing the dream of predicting barbecue summers.
 
It seems from this thread that the more means of accessing weather information you have on your boat the more you complain about it.

Personally when I'm sailing I get my weather information from the barograph onboard, the bits of paper stuck up in harbourmaster windows or from the VHF if and when I hear the announcement. And to be honest, I've always been happy with that. You've just got to accept the fact that any forecast is a guess by either a computer or a man who spends a lot of time looking at the sky and neither of those sources are infallible.

The fact the coastguard only give a two day forecast I think is fine. Any forecast for the third day is probably going to be wrong anyway, so what's the point in broadcasting it?

As for the shipping forecast on the BBC, you have to accept for every one person who wants to hear it, there are probably half a million people who don't. We're lucky they broadcast it at all.
 
Woodlouse said:
It seems from this thread that the more means of accessing weather information you have on your boat the more you complain about it.
Click to expand...

I think you have a point, up to a point. With tinternet, if you don't like the shipping forecast then it's always possible to find one you like :confused:. Then complain because the real weather didn't follow the forecast. With less information you are forced to look out of the window, listen to the wind and use a bit of common sense. But you will probably end up more cautious than the Met office.

The commonest complaint I hear raised by experienced yotties is when the forecast says it's going to blow a hoolie and it doesn't, so they miss a weekend's sailing. This is often blamed on the Met Office erring on the side of caution too much.

There is a good point made by several posters about issuing the percentage probability of an occurrence. Forinstance, saying "there is a 65% chance of it blowing a hoolie" would give us a choice, which most would decline, but which we would feel less inclined to blame someone else on.

For the weekend sailor (probably the biggest market) some sort of modifier for wind strength and sea state vs distance offshore, maybe as part of the internet version of the inshore waters forecast, would be useful as well. Even though experience tends to tell you what it will be, it takes time to accumulate and I still have to sell it to my crew, SWMBO. Not everyone appreciates that a NE6 is so much less of a problem when leaving our harbour, compared with a SW6.
 
Thanks for coming on with these views, Frank.
I'm one of the whingers though and I still would contend that the Inshore Waters forecast is not fit for purpose from my point of view.
I mostly sail in 'Gib Point to N.Foreland', a very long area north to south in which the difference in wind speeds, especially with a low crossing the N of the UK, can be very significant for a yachtsman. Yet the forecasts are too generalised and not updated frequently enough (and I mean on the CG MSI broadcasts, that's what I rely on) and are disturbingly frequently far removed from the reality of what is actually happening. I sail a lot, and I see this a lot.
I don't know what the answer is because I am (was) an IT specialist, not a meteorologist. But I repeat, that as far as I am concerned the Inshore Forecast system/service is not fit for purpose, it is not written with typical cruising people in mind.
Take for instnace what I heard several times this year, "winds F5-7". Right. F5, a great sail, but F7? No thanks. But if it said "winds mainly F5 with a 50% chance of F7 in squalls" that would be useful info upon which I can make a judgement. My point is that to say "F5-7" is no use for the people the forecast is aimed at. More thought could go into it.
 
F5-7

Take for instnace what I heard several times this year, "winds F5-7". Right. F5, a great sail, but F7? No thanks. But if it said "winds mainly F5 with a 50% chance of F7 in squalls" that would be useful info upon which I can make a judgement. My point is that to say "F5-7" is no use for the people the forecast is aimed at. More thought could go into it.
Click to expand...
Well, we've been sailing quite a bit in F5-7 this year and last year and that is pretty much what you get. There's been a lot of weather where it is blowing less than 20 knots between the showers, but in the showers or under the big black clouds it has been blowing 30+ knots for prolonged periods, five or ten minutes or longer, not a quick ease the main jobbie. If it was given out as F5 there would be plenty of people caught out and complaining, and if they say F5 gusting to 7 you just get a load of people on here saying there is no such thing as a beaufort force for gusts.

There has been a lot of F5-7 forecast this year and it's been pretty accurate most of the time up here in Argyll. The weather is changing and the forecast reflects that. I know what it means, and I don't find it difficult to make decisions based on such a forecast. (Mostly the decision is that we only go downwind on these days).

- W
 
Hopefully final thoughts - but who can tell?

Robin is correct in that there are several issues that all too easily get conflated.

In my very first post, I was trying to draw attention to the realities of what can be forecast rather than what some claim to be able to forecast and what some do forecast. A forecast that covers up to 5 or so days can only be in general terms. Where the highs and lows are, what is the general wind direction and force. This is so whether it is the Met Office Marinecall, Météo France, Theyr.com, PredictWind, Accuweather etc etc, The reasons are set out in my very first post. The size scale is around 100 miles.

When it comes to short period, detailed forecasts, operational meso-scale models can do no better than deal with a scale of around 25 to 30 miles generally and, maybe 10 miles near the coast. Again see my first post. British and French forecasters may add a little embroidery such as of detail “locally force 7 near headlands”. Météo France, quite sensibly, sometimes opts out and simply says “Effets des brises”. The Greek and Italian Met services opt out altogether and do not even try to give coastal forecasts.

In principle, over and near land models can come down to very small scales; I have seen output from an experimental small area 300 metre grid length model. The problem here and that some users and providers do not recognise, and are unwilling to acknowledge, is the practical limitations of meso-scale models, see my first post again. One of these is the input from global scale models. A small insignificant error in the large scale can result in a larger and very obvious error in the small scale.

The second issue that I was trying to draw attention to in my first post was the difficult task of trying to express a forecast in words to be heard or read. Whether or not there is a word limit that can be a most difficult job to make it understandable to a user. The imposition of a word limit can, curiously, make the job a little easier in that it compels the forecaster to concentrate on the salient features. Before anyone criticises forecast texts they should try my suggestion to write a script – see my first post yet again.

As Robin says, there is room for improvement. That has to be tackled on two fronts. First, and organisationally, the easier, is further improvement in forecasts. That will occur for two reasons. More powerful computers will allow better models and more use of ensembles. Secondly, data will improve as satellite sensors become more sophisticated. The costs of both activities will be great.

I suspect that the more difficult part will be in communication of the potentially vast amount of information to users. How will we get the information produced by an ensemble of model forecasts on a 1 km or smaller grid? Let alone how we sailors might use such a volume of data. I can see little improvement while we are dependent upon the written or spoken word. It will require the use of computers and high speed, broad band communications with sophisticated software.

When will we be ready to work like that when sailing?
 
Last edited:
You're better illustrating the point I was clumsily trying to make, really. My 5-7's this year have been 99% F4-5, for instance on one 6-hour trip to windward with 5-7 forecast (which spooked our friends on a neighbouring boat) it was mostly 4, with F7 for 3 minutes under one cloud, easily coped with. So if the forecast had said e.g. "F5 with a 10% chance of F7 beneath shower clouds" that gives a much better picture of what to expect. The forecasters knew the 7's would be under shower clouds, but if they don't say then you have to guess that fact and also how often it might occur.
What I'm asking for is just some more detail, aimed at leisure sailors. Surely not much to ask.
 
Cantata said:
You're better illustrating the point I was clumsily trying to make, really. My 5-7's this year have been 99% F4-5, for instance on one 6-hour trip to windward with 5-7 forecast (which spooked our friends on a neighbouring boat) it was mostly 4, with F7 for 3 minutes under one cloud, easily coped with. So if the forecast had said e.g. "F5 with a 10% chance of F7 beneath shower clouds" that gives a much better picture of what to expect. The forecasters knew the 7's would be under shower clouds, but if they don't say then you have to guess that fact and also how often it might occur.
What I'm asking for is just some more detail, aimed at leisure sailors. Surely not much to ask.
Click to expand...


It comes down to money, having dealt with the met office (and I am a supporter, ((ducks for cover)) they will give as much detail as you like providing somebody pays for it, regarding the F7 for a few minutes under shower clouds, I think this is an area that should be an obvious possibility with a little bit of understanding of weather, not being personally critical just would like to press the point that for the minute detail a lot of sailors require/expect the inshore forecast would be an essay.
 
Reply to Cantata

One of my favourite sayings is that weather does not know itself to within one Beaufort force.

I am in no position to comment on specific cases without a great deal of research – getting hold of the forecasts as issued and enough actuals and charts to make a judgement. . I can say that, in my experience, if a forecaster can say something like F4 or 5 increasing 6 or 7 then he will do so. Sometimes, the best that he can do is to say F 5 to 7. Maybe not helpful in one sense. Helpful if you take it as an indication of the uncertainty in the weather over a large area over a 24 hour period.

Whilst not doubting what you say in this instance, I have heard enough cases when the facts have belied the account. One such was a yachtsman who left the Channel Isles on a two day passage to Cork. 12 hours after leaving the Channel Isles he was rescued by the RNLII off the Lizard. He was quoted as saying that he was surprised by the strength of the wind. At the time he left, the shipping forecast said E F 6 to gale 8 perhaps severe gale 9 later. OK, he was caught in a F 10, maybe he was right to be surprised. I should add that 24 hours earlier, it had been 7 perhaps 8 later. Would you have gone? I would not.

In a sense, that gets at what I see as the most important aspects of forecasts. Safety. If forecasts keep me out of trouble, I am happy. Another of my sayings is that it is better to be in port wishing you were at sea than being at sea wishing you were in port. Forecasters do not deliberately over forecast. They try to interpret the computer guidance to the best of their ability and add value where they can. Meteorology is a science but the limitations of the application of the science has to be understood
 
Cantata said:
You're better illustrating the point I was clumsily trying to make, really. My 5-7's this year have been 99% F4-5, for instance on one 6-hour trip to windward with 5-7 forecast (which spooked our friends on a neighbouring boat) it was mostly 4, with F7 for 3 minutes under one cloud, easily coped with. So if the forecast had said e.g. "F5 with a 10% chance of F7 beneath shower clouds" that gives a much better picture of what to expect. The forecasters knew the 7's would be under shower clouds, but if they don't say then you have to guess that fact and also how often it might occur.
What I'm asking for is just some more detail, aimed at leisure sailors. Surely not much to ask.
Click to expand...

Without seeming to plug the Meteo France forecasts yet again, their inshore forecasts do mention gust conditions like F5-6 'avec rafales' which I find is a useful warning that I can factor in, especially if contemplating a windy upwind passage. We often forget that windforces are averages and can be a steady F5 or a very gusty one going from F3 to F7. OK so in this case we don't get the usual F7 seastate but we will nevertheless need sail area allowing for periods of F7 unless we want to be periodically overpowered. The difference between a steady wind and a very gusty one can maybe be inferred from the general synopsis information or that thundery showers are possible but it isn't always that obvious.
 
This is a most worthwhile thread and I'm grateful for your time and interest that you're putting into it.
I do know a bit about met in an amateur way but I'm trying to keep my user's hat firmly fixed on here.
So far as I'm concerned, I would like to be able to rely on the CG MSI broadcast inshore and shipping forecasts, I really don't want to mess about with internet info when I'm at sea if I can help it.
I'm just looking for more detailed info than what is offered, maybe it's increasing the number of words (it didn't occur to me that there's a limit) to add detail that the forecaster already has on his desk.
Something I've tried this year is to also catch the regular BBC forecasts as well to try to paint in more detail about the general weather behaviour but again the detail on offer is often not enough.
Say for instance the 24-hour I/Waters describes an area of low pressure to the north, and says 'SW veering NW later'. I know enough about met to deduce this is probably a cold front passing. And I know it will be between 12 and 24 hours in to future. But it would be useful to have an estimate of just when that front will pass, e.g. 'cold front clearing the Norfolk coast by 1500'.
I can't see that I'm asking for a lot here, just to be told some more detail that must exist already, and would enable me to make better judgements about my proposed passages.
 
Gust and other detail in weather forecasts

The French VHF forecast always includes the following text

Attention: en situation normale, les rafales peuvent être supérieures
de 40% au vent moyen et les vagues maximales atteindre 2 fois la hauteur significative.

That is very fair and sensible. I wish that the UK would do sp as well. It might be that HMCG would say that it would add that bit extra onto already fairly long texts that they broadcast 8 time a day. and sailors should know that anyway. I have heard the UK mention gusts from time to time. I am never sure whether F 5 to 7 is meant to include gusts or not. One practice that the UK follows uniquely, I think, is to issue a gale F8 if gusts are expected to F9 strength but the mean wind is F7.

Yes, we would always like more detail in a forecast and there is always the temptation for the forecaster to say more than he really knows – or thinks he knows. Is it helpful for the forecaster to give a false idea of what is achievable? As a senior forecaster I sometimes used to have to restrain forecasters on my watch from going beyond what we really thought. This is one reason why I am personally loath to use meso-scale forecasts from some of the private sector firms. It is my belief that they are offering more than is sensibly possible. The input from Tim Thornton supported that.

The general synopsis should include information about the passage of fronts where that would help in the interpretation of the forecast.. I guess, and I see this in French forecasts as well, that this either gets overlooked or the forecaster up against his word limit has to decide what he must say and what he can omit.

I started this thread partly because of comments that I saw on these forums and partly because of a recent letter included in the PBO November editorial. The input and discussions will be useful to me in talks that I give from time to time and as background for my website. It will also help when I next attend a MCA MSI sub group meeting.

Thank you everyone – even, or especially, those with whom I disagree.
 
#54 Today, 15:35
franksingleton
Registered User Join Date: Oct 2002
Location: UK, Mediterranean
Posts: 42

Gust and other detail in weather forecasts

--------------------------------------------------------------------------------

The French VHF forecast always includes the following text

Attention: en situation normale, les rafales peuvent être supérieures
de 40% au vent moyen et les vagues maximales atteindre 2 fois la hauteur significative.

That is very fair and sensible. I wish that the UK would do sp as well. It might be that HMCG would say that it would add that bit extra onto already fairly long texts that they broadcast 8 time a day. and sailors should know that anyway. I have heard the UK mention gusts from time to time. I am never sure whether F 5 to 7 is meant to include gusts or not. One practice that the UK follows uniquely, I think, is to issue a gale F8 if gusts are expected to F9 strength but the mean wind is F7
Click to expand...

Actually not only does the French forecast give the general health warning you refer to above every time, the reference to 'avec rafales' is also added into the broadcast text whenever they think gusts are especially likely and that extra 2 words shouldn't be too difficult to fit in.

I have always assumed F5-7 probably means very gusty conditions but that might be a wrong assumption on my part and I always assume the worst. On one occasion though I remember we left Cherbourg for Poole (60.5nm entrance to entrance) with a broadreach forecast of SW5-7 and had a yeeeehaaaaa crossing in under 8 hours with the wind steadily between 35 and 40 knots for 7 of those hours. If we were going the other way I might well have been very rude about the forecast, albeit unjustly.:)
 
Saving words.

I find the Inshore Forecast to be generally very helpful. I get it from the CG and/or Navtex.

One way to cut down the number of unnecessary words in the Inshore Forecast, would be to give the areas short names. At present the areas are generally defined as "Point X to Point Y including Z". The Shipping Forecast areas are either one or two words. Surely the Inshore areas could also be given shorter, one or two word names.

The other possibility is to stop giving the sea state in a weather forecast. Sea state in coastal waters depends on so many factors, like tide, fetch, and depth, that to give a generalisation based purely on weather, is unhelpful.

If either, or both of these suggestions were used, perhaps there would be more space for more information
 
Inshore waters forecast

The area names were under consideration some while ago. I do not where that one got to. It is a problem that we share with the French. One problem is that you have to have areas that can be easily recognised by those who do not know the area. For example, I know Cap de la Hague and Pointe de Penmarc'h.

Sea state is something that I have argued could be omitted except for times when there is a big swell running, say 2 m or more. However, I think that users wanted sea state. My thinking was that sea state is a complex result of tidal stream, bottom topography, coastal topography and wind plus, of course swell. You are quite correct to my mind and, like you, I think that a sea state forecast is rather meaningless.

Any sailor with some experience will quickly learn what gives a rough sea. Swell is a different matter and can make entry to some harbours difficult or dangerous. We sailors cannot predict swell. I have no experience of this around the UK but plenty around the west of France and Atlantic Iberia. Météo France does give swell but not sea state and I think that is the right thing to do. However, as I said, feeds from users have dictaed otherwise.
 
Last edited:
Weather Forecasts - Météo France

Météo France has been mentioned several times in this thread. I have used them a great deal and am always surprised and worried that so many British sailors seem to think that all Frenchmen speak incredibly quickly.


On the whole, CROSS VHF broadcasts are good, with exceptionally clear diction. They are easy to follow even for a non French speaker. Two hints. First, use the links at http://weather.mailasail.com/Franks-Weather/European-Marine-Weather-Forecast-Texts to see the texts before you go to France. Get a hang of the terminology used and, more importantly, the places that they refer to.


Secondly, use http://weather.mailasail.com/Franks-Weather/French-Marine-Weather-Terms to see the terms that they use. One criticism is that they are not seem not to have a well defined set of words.

I always prefer to use local services and, apart from Italy, have not usually been disappointed. One example is that in shipping forecast terms, French and Spanish NAVTEX divide Biscay into 4 areas compared to the BBC one and Fitzroy into two.


I would like to drop sea area Trafalgar from the midnight shipping forecast becasue there are better services available but user demand seems to prevail.
 
Météo France has been mentioned several times in this thread. I have used them a great deal and am always surprised and worried that so many British sailors seem to think that all Frenchmen speak incredibly quickly.
Click to expand...

We have met quite a few Brits in France that never use the Meteo France forecasts broadcast by CROSS on VHF although will view them in the harbourmasters offices, albeit that some places even translate them into English. We have even met several Brits in Southern Brittany who only ever listen to the British Shipping Forecast for 'Biscay' which rarely bears any resemblance to the weather inshore along the West France coast. Even met one chap who was waiting and waiting in Camaret because of the forecast for gales in Finisterre when the French district known as Finisterrre has nothing to do with sea area Finisterre!

For those not accustomed to Meteo France, it is helpful to remember that CROSS forecasts on VHF are transmitted at set times and then goes out from successive transmitters around the coast at around 30 minutes intervals until all have been done. At any one time it is usual to be in range of 3 transmitters, the one nearest obviously but also the ones either side of your position. So if you turn on to the 1st of the 3 you can take down the forecast and if any bits were missed you still have 2 more chances to fill in the gaps.

The CROSS format is always the same. Firstly any gale/strong wind warnings, then the general synopsis, then the forecast first for the daytime, then overnight and then the outlook. They then give the weather weather (cloudy, sunny or whatever) and the visibility. Then follows actual conditions at the various stations in their forecast area, wind direction and strengths and also barometric pressure.

What can vary according to locality is that some local names are used, for instance NW could be Nord Ouest or in Brittany, 'Noroit' (pronounced norwah), SW could be Sud Ouest or Suroit' (surwah). No problem really as long as you know what might come. There are also several variations of rain, drizzle, thundery rain and thunderstorms.

Instead of 'seastate' which I think is unhelpful as it depends on wind with or against tide, local conditions and so on they give the swell height forecast 'La Houle'. This is useful as some harbours and anchorages can become inaccessible with a big swell, whilst at sea clear of land the swell is rideable and more academic although very capable of rolling the wind out of sails. We once were caught out in a really bad gale between L'Aberwrac'h and Guernsey (missed by both Met Office and Meteo France but not Jersey we discovered later as we got in their VHF range). On the tops of the swells we had 45kts over the deck and that whilst going downwind at 10kts SOG yet in the troughs the wind was less than 20kts. I can't remember the predicted swell height now but the actual with seastate on top was frightenly huge. We had sat out 3 succesive gales in L'Aberwrac'h which had built up the seas. What we had experienced was I believe a secondary front which had formed quickly and unexpectedly and we had W-SW 9 instead of the expected NW going SW 5 dec 4 and the actual W3 we had for the first hour.

Anyway I would encourage Brits in French waters to make use of their VHF forecasts, they are not that difficult to master with the aid of a weather terms cribsheet and a bit of patience.
 
You must log in or register to reply here.

Share this page

Top