XC Weather Disclaimer

[franksingleton]

I use windfinder - but as has been said above, the basic data is the same.
However, you should always look at a grib file as well - as this will give a good impression as to what is going on (particularly if you can animated it over a period of 48h or so).

................................

Windfinder is based on the GFS. If the GFS gets it wrong, so will Windfinder. XCWeather is the GFS.

You really cannot look at specific grid points. You have always to look at the overall pattern. I say again, any model can only define weather on a scale of about 4 or 5 grid lengths. For the GFS and the UK Unified model, that is about 100 to 15o km. For Windfinder, it is about 50 to 60 km. For ECMWF, it is a little less than Windfinder.

However, like all models that are not run by national weather services, Windfinder starts with no detailed data. In the Med, http://212.175.180.126/DENIZ/DTS/sea.php should be the best. It is the ECMWF model that starts with a detailed analysis and uses what is probably the most advanced of all.

Outside the Med, I use the GFS. For small detail, if I really needed it, I would use one of the national weather service sites. In practice, I look closely at GMDSS forecasts. I monitor those continually together with the GFS to 8 days ahead.

 

[Tim Good]

XC can be very good but it is a land based forecast. If you based it on a city then you should always add 20% for being at sea. Another one which is very good is passage weather which i have found to be accurate.

RASP can be very useful an picks up on local conditions. You'll need to select Surface Wind (10m) from the parameters first:

http://rasp.inn.leedsmet.ac.uk/RASPtableGM/RASPtableGM.html

 

[franksingleton]

XC can be very good but it is a land based forecast. If you based it on a city then you should always add 20% for being at sea. Another one which is very good is passage weather which i have found to be accurate.

RASP can be very useful an picks up on local conditions. You'll need to select Surface Wind (10m) from the parameters first:

http://rasp.inn.leedsmet.ac.uk/RASPtableGM/RASPtableGM.html

First, can I say once again, that XCWeather is the US GFS. Passageweather show a number of models including the GFS. There is no difference between the two GFS data. The same information is available on Magic Seaweed, zyGrib, Ugrib and on several tablet apps. Use whichever you like the look of best or can get in the easiest way but do not think that any is “more accurate” than any other. They are all the same data.

I had not seen http://rasp.inn.leedsmet.ac.uk/RASPtableGM/RASPtableGM.html before, thank you. Before I take any forecast seriously, I like to know its provenance. It says that this is the WRF model – as used by Windfinder, Windguru, theyr.com (I think) and some others.

It is not clear who is producing this. Most likely, they are starting from the GFS 0.5 degree grid output. The model will only be run over a restricted area, although larger than that shown. The area boundaries will be updated using the GFS forecasts. In other words, like Windfinder et al it will be highly dependent on the GFS They are unlikely to analyse any detail,

I note that it claims to give forecasts to 5 days ahead. However, any meso-scale model is unlikely to add anything to global eg GFS output beyond 48 hours and maybe less than that.

Note that the UK, high in the modelling league, only runs its NAE model to 48 hours ahead. Its UK area forecast as seen on its app and its website is only run for 36 hours and is not very good after about 12 to 15 hours, sometimes less.

http://212.175.180.126/DENIZ/DTS/sea.php can be used sensibly to 5 days ahead because this is the ECMWF model. It is run on a grid of 0.125 degrees. It starts with an analysis on that scale and, being a global model, does not have boundary effects. If we could get this for Western Europe, it would give the best indication of local variation predictable on a time-scale useful to sailors.

The stream lines shown are fascinating. However, remember that the WRF use a grid of around 10 km. It can only define detail on a 40 or 50 km scale. Use of topography to produce this level of detail must be highly speculative.

 

[Tim Good]

Frank
http://weather.mailasail.com/Franks-Weather/Home

Nice page you got there. Bookmarked for bedtime reading

 

[franksingleton]

Nice page you got there. Bookmarked for bedtime reading

Thank you. Weather is not simple. On my site I hope that I do not talk down to sailors. As Huckleberry Finn said of Pilgrim's Progress, "Some of the statements are interesting, but tough." Hopefully, not too tough.

 

[Bru]

I've lately taken to relying quite heavily on the Meteo Consult Marine android app

It breaks down the UK coastal forecast into much smaller and therefore more relevant areas than the Met Office (5 areas to cover the coast from North Foreland to the Wash as compared to just one) and it seems to be generally more accurate at forecasting what we're likely to get rather than the worst case scenario (which is, I'm convinced, what the Met Office delivers)

I do consult the UK forecast of course, mainly on the basis of reckoning it as a measure of how bad it might get at worst! It has to be said that the joke of sitting in a flat calm whilst listening to a strong wind warning for right now is wearing a bit thin though

 

[pagoda]

I've lately taken to relying quite heavily on the Meteo Consult Marine android app

It breaks down the UK coastal forecast into much smaller and therefore more relevant areas than the Met Office (5 areas to cover the coast from North Foreland to the Wash as compared to just one) and it seems to be generally more accurate at forecasting what we're likely to get rather than the worst case scenario (which is, I'm convinced, what the Met Office delivers)

I do consult the UK forecast of course, mainly on the basis of reckoning it as a measure of how bad it might get at worst! It has to be said that the joke of sitting in a flat calm whilst listening to a strong wind warning for right now is wearing a bit thin though

I find XCweather quite useful, though take it with a small pinch of salt. The "gustiness" Red arrows are a handy measure of how nasty in may be and looking upwind of where you may be (later, tomorrow) is helpful. The Met Office plays VERY safe most of the time. If they mention the number 3 and/or Variable I generally expect plate glass and not a useful puff of anything. A good 15-20Kt sailing wind is usually served up as SWW with dire overtones. I use the US Navy WXMAP data to get a feel for 3-5 days ahead, and that's reasonable as well. I don't think you can usefully speculate on where a low sitting off Halifax NS is going to be in a week's time. 3 to 5 days away maybe, but not really more.

Graeme

 

[Daedelus]

I still have a copy of an XC forecast where wind forecast for 10am was 13 mph gusting 16 mph whilst by 1300 it had increased to 244 mph gusting 292mph and remaining above 200mph for therest of the day. On balance I didn't believe them and I understand it remained a lovely day throughout. I wonder if they have a simple program which uses the GFS data and applies a simple algorithm to it which is then unsullied by any sense checking.

 

[franksingleton]

I've lately taken to relying quite heavily on the Meteo Consult Marine android app...............)

As far as I know, Météo Consult is one of those that starts with the GFS. I am not sure whether it runs a meso-scale model or just interpolates from the GFS. If it is a meso-scale model then they will not start with a detailed analysis.

It is entirely automated so treat it with some caution. Last summer, in Audierne, we had heard the Météo France forecast of thick fog. A Brit yacht left intending to anchor off Ile de Sein. When I mentioned fog, they dismissed the idea saying that the forecast did not mention fog. They had read the Météo Consult forecast in the marina office, needless to say, there was fog with Ile de Sein reporting 50 metres.

The wind data are probably OK. I prefer to use the GFS if only because I know what I am getting.

Any GNDSS forecast has necessarily to be brief. It is and only can be a general description. The areas are defined by the user through the UK Safety of Navigation Group. Listened to carefully, particularly hearing each forecast and looking for changes in emphasis. I use it in conjunction with the GFS. I would not go to sea without hearing or reading the relevant inshore and sea area forecast in whatever country I am sailing. For a cross Channel, I would look at Météo France as well as UK shipping forecasts.

 

[franksingleton]

I still have a copy of an XC forecast where wind forecast for 10am was 13 mph gusting 16 mph whilst by 1300 it had increased to 244 mph gusting 292mph and remaining above 200mph for therest of the day. On balance I didn't believe them and I understand it remained a lovely day throughout. I wonder if they have a simple program which uses the GFS data and applies a simple algorithm to it which is then unsullied by any sense checking.

I cannot explain that one. I have heard of such nonsense from another forecast service. XCWeather should just be the GFS. Their forecasts at specific locations are simply interpolations from the GFS grid points.

Are you sure that it was XCWeather? I recollect some excitement when another firm did something as stupid as you describe.

 

[Bru]

It is entirely automated so treat it with some caution..

I treat all weather forecasts with some caution ... if not with outright scepticism!

It may well be based on GFS (it doesn't say on their website) and to what extent their claims of interpretation by expert forecasters are valid I've no idea. All, to be honest, that I care about is that the info is presented in an easy to read format where I can see what I need to know at a glance and that so far experience has found it to be accurate within the limits of reasonable expectations.

I would never rely totally on one source of info and it's somewhat a case of the Shipping Forecast for the worst we might encounter, Meteo Consult for what might actually happen and look up to see what is really going on! Got half a dozen other weather apps and web links that'll get a scan if there's a major disparity between the major two or three

 

[Cantata]

I have taken notice of XC Weather more than any other forecast this year and have found it generally much more accurate that the UK Met Office for local conditions (usually comparing with the Inshore waters F'cast, usually in the old 'Thames' area). As with all of them I find it worth starting to look several days before, and start to believe in it if the forecast remains much the same as the due date draws nearer. I took a decision to turn back and wait 12 hours once this year based solely on what XCW had been saying for several days about a short-lived improvement in conditions very locally, it wasn't mentioned by the MetO but turned out to be spot on.
Living in East Kent and no doubt influenced by geography, I have long observed that while the MetO seem quite good at getting it near enough right when the weather systems are trundling across UK west to east, if there is any different sort of pattern then whatever methods they use don't seem to work so well.
And I totally agree with earlier poster who observed that MetO forecasts always seem to assume worst case. I was 'sailing' last Weds in a flat calm all day, and the forecasts kept mentioning 3's and 4's. "Look out of the window", we cried.

 

[NormanS]

Where I sail, internet access is an occasional bonus, and generally very slow. Of all the web forecast sites that I have tried, XC downloads far faster than any of the others. It is extremely easy and quick to read, and provided that you remember that it is only a "forecast", I find it very useful.

For the Inshore Forecst, I mostly rely on Navtex. I find that on-line, the IF takes ages to download.

 

[christhefish]

Yesterday I ventured out with a friend and her two boys 12 and 14, neither of whom had been on a yacht before.

Preparing for the trip to sail them back to the island they came over from in the morning, I checked XC weather among others. 15mph gust 28mph reducing to 13 and 25 respectively was the wind forecast. I set off reefed in with the idea of later taking the reefs out.

Instead of decreasing, the wind increased and INCREASED! At the point we had a 47knot gust over steady 35 knots it was time to return. Took the main down and sailed ion a run with a 2/3 reefed jib doing 9 knots!

Now while it was rather exhilarating for me, the youngsters, although saying they enjoyed the day, were rather nervous at the point we turned back.

So I had a look at XC weather on return and now they had a wind indication which more reflected the conditions. So it is not a forecast, but a late response to what actually is happening. Their disclaimer http://www.xcweather.co.uk/Info/Disclaimer which basically says 'We are rubbish! Do not make any plans on what you read here because we are not a weather forecast site."

I shall place less reliance on the site as I have had up until yesterday.

years ago I used to tow a small boat all over the country fishing,one weekend the met office forcast was NE F1 or 2,afer towing 90 miles we were met with SWF9/10
the following day I was sent to do some work in the met office I mentioned the weekend and was told with a serious face the forecast was spot on its just the weather changes!!!!!!!!!!!!!!!!!! check a few sites out and decide for yourself,,,,I have found XC pretty good

 

[franksingleton]

I treat all weather forecasts with some caution ... if not with outright scepticism!
.......................

There is no such animal as an accurate or a precise forecast. Anyone who claims to provide such simply understands neither meteorology nor weather models. I make that clear in http://weather.mailasail.com/w/uploads/Franks-Weather/bookcover.pdf.

There, I say that it is necessary to have an idea about what forecasts can and, equally important, cannot do for you. Forecasts have to be used with care, in a pragmatic way, using commonsense and experience. GRIB files, usually from the GFS, are a useful tool in whatever form you get them. But, they do not provide an answer in isolation any more than the shipping or inshore waters forecast.

There is better information than the GFS or any of the other services such as Météo Consult, Windfinder, Windguru etc. The Turks have led the way for the whole of the Mediterranean. I am hoping that some pressure can be brought on western European Met Services for there to be as good a service west of Gibraltar.

The suggestion that forecaster only give the worst possible or deliberately over-forecast is a hoary old chestnut and simply incorrect. Nevertheless, a statistical study does show marine forecasts over-forecasting by about one force. The reasons are twofold.

First, a gale warning MUST be issued if a gale is POSSIBLE somewhere in a sea area. Similarly for a strong wind warnings in inshore waters. Warnings can only be cancelled or allowed to lapse when the forecaster is sure that there are no gales. Of necessity, that means that warnings are kept in force until after the gale/strong winds have ceased.

Secondly, at the low end of the scale, in a 24 hour forecast over a sea area or coastal area, it is impossible to differentiate between Forces 0 to 3. As a result, words such as “variable 3”, possibly “variable 2 or 3 “will be used. Have you ever heard a forecast of Force 0? Yet we have all been out there in a Force 0.

 

[franksingleton]

....................I find it worth starting to look several days before, ..........

That is good practice. If you keep on top of it you minimize the cahnces of being caught out.

I think that many people like XCWeather for its presentation. It is not my choice but the data are just the same as any of the many GFS services available to us. Always remember that the GFS will often under-estimate the strongest winds. This is because the computations are done on a grid of points about 27 km apart although you only get the output about 50 km apart.

I have commented on the over-forecasting aspect in another post. I will just emphasise that forecasters fo not set out to over-forecast.

 

[Bru]

I have commented on the over-forecasting aspect in another post. I will just emphasise that forecasters fo not set out to over-forecast.

I don't suggest that the Met Office sets out to deliberately OVER forecast. I do suggest that they err heavily on the side of caution

My belief, along with many others from posts hereabouts and conversations elsewhere, is that the UK Met Office shipping and coastal forecast wind speeds are based on the maximum gust speed produced by any of the model runs. Or to put it another way, if any model run predicts gusting to F6 anywhere in a coastal forecast area at any point in the 24 hour period, F6 is what will be forecast and there'll be a strong wind warning for that area for the entire 24 hour period of the forecast

Part of the problem is that the Met Office are restricted, by the very limited format of the shipping and inshore forecast, to a very coarse resolution in both time and space. Additionally, as the national weather service for the UK they are duty bound to warn of the worst case possibility

My admittedly limited and relatively short experience of using the UK Met Office forecast is that approaching the forecast with the above in mind generally makes sense of what otherwise often appears to be arrant nonsense!

From the point of view of planning a days sailing, the Met Office shipping and inshore forecasts are all but useless. Empirical and apocryphal as that statement may arguably be, that's my experience and that of many others I suspect.

The GFS based forecasts in their various forms tend to be a more accurate indication, within the limits one might reasonably expect of an imprecise science such as predicting the weather, of what the sailor might encounter in his local area. We each tend to find our own favourite interpretation and presentation of the model and as I have already mentioned I'm currently enamoured of the Meteo Consult output

Since it doesn't slavishly follow the GFS data (which I do also look at I might add) they are clearly "adding value" whether automatically or by human intervention and whatever they're doing to the data they're doing it pretty well as far as the Thames Estuary is concerned because they've not been far out all summer!

I don't think anybody expects miracles of accuracy from any forecast. If I see SW F3 gusting F4, I might expect anything from F2 to gusts of F5 and I won't be massively surprised if the direction is anywhere between a Westerly and a Southerly! I plan accordingly.

What is utterly useless to me is a Strong Wind Warning, which keeps me in port on a perfect sailing day, issued because just one of dozens of model runs predicted a brief gust of F6 somewhere on the coast between North Foreland and the Wash at some point over a 24 hour period.

If the Met Office were to publish, even if they cannot include it in the actual shipping forecast itself, more detailed information on mean and gust wind speeds over somewhat smaller areas and if they were to publish a confidence assessment in the forecast (as Meteo Consult and other services often do) I'd have more faith in their output.

 

[franksingleton]

I don't suggest that the Met Office sets out to deliberately OVER forecast. I do suggest that they err heavily on the side of caution
...........

Your long and thoughtful post needs an equally long and, I hope, equally thoughtful reply.

As someone who spent a third of his working life at the sharp end and still has contacts in the Met Office. I can state categorically that the forecasters do as honest a job as possible. They definitely do not base the forecast on expected gust speeds. If strong gusts are expected, then some words such as “squally showers,” would be used. That is rather more useful than the blanket statement that accompanies all French marine forecasts – “en situation normale, les rafales peuvent être
supérieures de 40 % au vent moyen….”

Of course, you are correct about the limitations imposed by the brevity of any forecast and the straitjacket of fixed sea areas and coastal zones. The shipping forecast has to be written in 330 words. That is a tall order; try it for one or two sea areas when you know what actually happened..

There is a fine balance between giving a sensible and meaningful description of the generally expected weather and the possibility of strong winds, or fog. I could write at length, and have done in the past in these forums, on using forecasts to best advantage. I can only suggest that you get my book when it comes out in early January.

If I had my way, unfortunately not the case, we would be able to have the UK version of the GFS available to us as well as the ECMWF output. They are at the top of the weather model pecking order. I do wish that the yachting press would take up the cudgels on this issue. There are no technical or political reasons, as far as I know.

Worded forecasts with their human input are necessary despite their shortcomings. Computer output from these weather models would be a great help and far better based than anything else that I have seen.

The GFS may not be the best weather model but it performs pretty well and it is readily available for whatever area I want. Using zyGrib, UGrib or a tablet App, there is no need to have the area limitations of XCWeather or Passageweather; it is easy to see the large scale pattern . Like all forecasts it has to be used with care. Like radar and AIS, using forecasts systematically, using commonsense and experience, will minimise risk. Risk when sailing will never be eliminated.

After 14 seasons in warmer climes we have just returned to the UK. Being 80 with a wife not much younger, we want to avoid being out in gale force winds or thick fog. We also do not want passages much over 12 hours. So, we left Rochefort (Charente Maritime) and coast hopped round to St Peter Port. Our last Channel crossing was in 2000 so we were aiming for a passage to Dartmouth entirely in daylight.

Whilst still in St Helier, on the 29.30 July, it was looking as though August 3 or 4 would be good days to leave St Peter Port. SW winds, not too strong. After that it did not look good. We watched the forecasts closely over the next few days. On 2 August the 4th seemed slightly the better day, winds SSW 4 or 5 would be great. A touch of 6 would be possible. We opted for the 4th.

Late on the 3rd, the UK and French shipping forecasts upped the ante to SW 5 or 6. The UK inshore was slightly stronger and the GFS agreed. There was a mention of F7 but the UK shipping and inshore texts both made it clear that the F7 would be over the west of Plymouth and the Lye Regis to Land’s End areas. In this case, the shipping forecast and the inshore forecast were pretty specific.

We thought about delaying but the outlook was not good. We did have time to spare. We thought about rigging the inner forestay and the storm jib but, on the basis of the forecasts decided not to.

In the event, halfway to the Casquets TSS, we put in the second reef and had a half furled Genoa. It was a good crossing. We were doing 7.5 knots in our HR 34 for long periods and averaged 6.5 door to door. The wind was a good deal of F5 with a spell in mid-Channel of a top end of F6. Gust reached F7 speeds. It was never too strong and we never felt the need for the third reef.

Had I only see the forecast late on 3 August, I would have not known what to do. Having lived with the forecasts over the preceding period there was no real doubt.

I could give many other examples of such decision making.