Inshore weather forecast somewhat deceptive.

[franksingleton]

1. It is a topographically complex area. No forecast written in a few words can ever describe what you will experience over the whole area.
2. MeteoFrance attaches a caveat to all its marine forecasts that says that winds might be 40% (from memory) stronger in gusts than the average predicted. That is so obvious from experience that the UK sees no reason to add such a caveat.
3. In weather when there is vigorous convection, I have seen gusts 100% stronger than the mean wind.
4. Normally, when really strong gusts are expected, the forecaster should say so. On the other hand, they assume that mariners are well aware of that effects of both showers and coastal topography and may just say something like “heavy showers with thunder.” Everyone will know that there will be strong gusts.

 

[savageseadog]

I've come across gale force katabatic winds (not gusts) around the Isle of Man when mean gradient wind was 12Kn. This could potentially happen in any mountainous area.

 

[SimonFa]

I am a Clyde man too, and the thing that I find odd is that the Inshore weather forecast, does not agree with the hour to hour forecast from the Met Office web site for the same area, nor the BBC forecast that's provided to them by the Met Office. Why? They are all the same times covering the same place!

You would think in this day and age, with their massive budget and state of art computers, they could provide much better hour to hour forecasts.

Out of interest I Just looked at the Inshore Forecast for the Clyde area. South or southwest "or" being that they are not sure. 4 or 5 - another "or" - seems they are still not sure again, occasionally 6, then becoming 3. So take yer pick.

Following 24 hour outlook - Variable 2 or 3 (or again) then north or northeast, 4 or 5 occasionally 6. So they reckon it's going to be a 2 to a 6, Variable and the north or perhaps northeast.

What use is that to anyone? They would as well saying "the forecast today will be somewhere between nice and calm and blowing old boots and it will be coming from somewhere in the sky"

Perhaps this is what we need?

http://youtu.be/4zlMO4z2T74

 

[franksingleton]

I am a Clyde man too, and the thing that I find odd is that the Inshore weather forecast, does not agree with the hour to hour forecast from the Met Office web site for the same area, nor the BBC forecast that's provided to them by the Met Office. Why? They are all the same times covering the same place!

You would think in this day and age, with their massive budget and state of art computers, they could provide much better hour to hour forecasts.

You are comparing chalk and cheese or, perhaps better, cheddar with gorgonzola. The hourly forecasts are directly output from the computer and are interpolations between grid points of their Numerical Weather Prediction model – the UK version of the US GFS. There is no human input. The Inshore Waters forecast is a best endeavours worded summary suitable for an understandable broadcast verbally or in text.

An exercise that I have often asked sailors to do is to sail for about 12 hours along the coast, maybe 6 in one direction and 6 back, or in an area of complex topography. Then try to describe what you experienced in a form suitable for broadcast by HMCG as a forecast. Then remember that the forecast is for a larger area for 24 hours and a 24 hour outlook.

Out of interest I Just looked at the Inshore Forecast for the Clyde area. South or southwest "or" being that they are not sure. 4 or 5 - another "or" - seems they are still not sure again, occasionally 6, then becoming 3. So take yer pick.

Following 24 hour outlook - Variable 2 or 3 (or again) then north or northeast, 4 or 5 occasionally 6. So they reckon it's going to be a 2 to a 6, Variable and the north or perhaps northeast.

What use is that to anyone? They would as well saying "the forecast today will be somewhere between nice and calm and blowing old boots and it will be coming from somewhere in the sky"

Sailors know all too well how the weather can and does vary in space and time. I always find comments such as those in this and other threads rather strange. Weather can be like that. The forecaster cannot make the weather les complex than it really is. He has to try to describe what he and his computer model expects is as few words as possible and try to cover all the variations. If he is too broad brush as he often has to be then inevitably such criticisms as yours arise. You have the choice of using the computer generated hour by hour computer forecasts or the brief summary. It is your choice. The bottom line is that any serious users of forecasts, such as we sailors, have to put in intelligence and experience. It will rarely be the case that a forecast will describe accurately what you will experience despite the great improvements since I was a senior forecaster at Bracknell.

There have been many times when it has been suggested that forecasters deliberately over forecast wind strength. In this thread there are some complaints of under forecasting. Just an observation.

 

[Moonbeam]

The bottom line is that any serious users of forecasts, such as we sailors, have to put in intelligence and experience. It will rarely be the case that a forecast will describe accurately what you will experience

Well said Frank

 

[Sgeir]

There have been many times when it has been suggested that forecasters deliberately over forecast wind strength. In this thread there are some complaints of under forecasting. Just an observation.

Yes. It happens. In the past couple of weeks, all the forecasts, from a range of forecasters, have been a bit out, perhaps underestimating wind and gust strengths.

Given the complexity of these islands and often being caught between Atlantic lows and the continental highs, it is not really surprising that there can be variation between forecast and actual weather conditions, IMHO. I think most sailors understand that and take account of possible variation when passage planning.

 

[Stoshak]

Accurate weather forecasts are extremely important to all sailors, and the quality of the service provided to us by HMCG and the Met Off is very poor. I believe the enforced brevity of the forecast is particularly unprofessional when used over such vast sea areas.

This brevity seems to be at the behest of the broadcast service. It is not necessary in a internet context.

In any case, Jersey provides a splendid service. Compare their offering at the end of the Inshore Waters with the grudging scarcity of Bracknell or where ever. They consistently have double the word count of the mainland offering.

I always feel very much better informed weatherise when i visit the CI, and I see no reason why the forecast for other areas should not be as comprehensive or as accurate.

 

[jac]

What concerned me was checking the coastal waters forecast for Solent and Christchurch bay. A relatively small area.
It turned out to be over estimating the wind by about 1 force according to bramblemet and what we observed.
It had the weather wrong. Forecasting heavy rain showers and rain all day.
The same BBC site was forecasting sunny with occasional light cloud for the entire area and the wind was about right.

I can accept that forecasts can be wrong due to local reasons but rain or sunny is hardly driven by local conditions yet if using the same model how can such different outputs occur.

One would almost think that "they" were manipulating the coastal and inshore waters forecasts to discourage wafis from straying too far.

 

[franksingleton]

Accurate weather forecasts are extremely important to all sailors, and the quality of the service provided to us by HMCG and the Met Off is very poor. I believe the enforced brevity of the forecast is particularly unprofessional when used over such vast sea areas.

This brevity seems to be at the behest of the broadcast service. It is not necessary in a internet context.

In any case, Jersey provides a splendid service. Compare their offering at the end of the Inshore Waters with the grudging scarcity of Bracknell or where ever. They consistently have double the word count of the mainland offering.

I always feel very much better informed weatherise when i visit the CI, and I see no reason why the forecast for other areas should not be as comprehensive or as accurate.

Brevity of forecasts is always an issue. First, the BBC will only allow 3 minutes for the shipping forecast. As the BBC has no formal responsibility to broadcast the SF there is no way they can be pressured other than by ministers. NAVTEX has limited time slots. The routine SF is of lower priority than messages A, B, C and D. Om 490 kHz the inshore forecast can take the whole slot. HMCG wants brief forecasts as they have other functions. Also longer forecasts would mean that those at the end of the broadcast cycle would suffer.

In any case, try my suggestion of sailing for 12 hours and then describing what you experience. Imagine that for all the variations in space and time in an area. Then think of attention span and information overload.

The Internet is not a part of the GMDSS. Who is going to pay for a service that would be expensive in human time? If you think that the taxpayer should then take it up with your MP. The UK is not unique in its marine forecast service as anyone will know who has sailed in other European countries.

My advice is to use GRIB data from the US GFS as an aid to understanding and using GMDSS forecasts . I repeat that intelligent users of forecasts, sailors for example, should have an appreciation of weather, its variability and the concept of predictability.

As for any marine forecast that from Jersey can be good or not so good. It is a little more localised but I would never use it blindly to the extent of ignoring the MeteoFrance Bulletins Cotieres or the UK SF.

 

[franksingleton]

What concerned me was checking the coastal waters forecast for Solent and Christchurch bay. A relatively small area.
It turned out to be over estimating the wind by about 1 force according to bramblemet and what we observed.
It had the weather wrong. Forecasting heavy rain showers and rain all day.
The same BBC site was forecasting sunny with occasional light cloud for the entire area and the wind was about right.

I can accept that forecasts can be wrong due to local reasons but rain or sunny is hardly driven by local conditions yet if using the same model how can such different outputs occur.

One would almost think that "they" were manipulating the coastal and inshore waters forecasts to discourage wafis from straying too far.

If you are talking about http://www.bbc.co.uk/weather/coast_and_sea/coastal/11#selected-location, then you should realise that it is computer output for a specific location – a grid point. It is misleading to describe it as for an area. It has no human input or verification/validation. The same is true for non-official forecasts such as https://www.windfinder.com/forecast/solent_mrsc , http://www.windguru.cz/int/index.php?sc=275791 or MeteoConsult forecasts on the Internet.

I will repeat what I have said on many occasions, the forecasters writing shipping or inshore forecasts have no agenda to over or under predict wind or any other element. It is always difficult to discuss any particular forecast in terms of what is experienced by a sailor in a relatively small part of an area for a relatively short time. Forecasts cover large areas or long stretches of coast for 24 hour periods. Because the forecast does not match what you have seen does not mean that it is a poor forecast.

 

[Cantata]

?.............Because the forecast does not match what you have seen does not mean that it is a poor forecast.

Sorry Frank but that is exactly why this subject comes up again and again. It rather suggests that we poor customers of the Met Office should just put up and shut up, the customer is just a nuisance and the forecaster is always right.
There's a lot about Shipping and Inshore Waters forecasts that could be improved if anybody could discover how to get it done - the word count, for a start, far too short to allow likely variations across large sea areas to be described.
We are not all numpties, we do understand a lot about the weather, we just get tired of listening to forecasts where we have to work out the likely detail for ourselves. The service is not user friendly.

 

[bitbaltic]

My advice is to use GRIB data from the US GFS as an aid to understanding and using GMDSS forecasts . I repeat that intelligent users of forecasts, sailors for example, should have an appreciation of weather, its variability and the concept of predictability.

This hits the nail on the head. I have always felt that the best way to get an understanding of the weather expected on a given day is to look at multiple sources, apply local knowledge, chuck in a margin of safety based on experience, and produce what is effectively my own forecast which goes into the passage plan. I have in the past tried using various sources; these days I use PredictWind, because I like the derived models but more importantly because it gives high-quality access to two different models (American GFS and Canadian CMC). The important thing to realise is that what it's presenting is data, not a synthesised forecast- creating that i's my job. The key aspect of then making a decision about the weather (='my own forecast') is the degree of consilience between the two PW models. Less consilience means erring on the side of the worse forecast and applying a greater margin of safety. I am also always interested in the difference between the predicted steady wind and the predicted gusts, because in the inner Bristol Channel experience tells me that the wind speed is often maintained closer to the gust speed.

Once I have reviewed all this data I can write my own forecast, n paper on in my head, and I then read the GMDSS Inshore Waters in the light of this. Again, I'm looking for consilience and any reason to revise my opinion. I might perhaps go on to look at the synoptic charts in I am planning a longish passage or if I still can't formulate a decided opinion on the day ahead. Once all this is done I have the understanding of the weather that I'm going to base my passage on.

In practice and in the sort of settled weather that we all try to sail in, this process does not take any significant time or require any kills beyond those aquired on a DS theory course and frankly I assumed all sailors applied a process like this. The other day I was showing a pal and experienced sailor the Predict Wind models and there was a discrepancy between them. His remark was to enquire how you would then know which one to believe; so perhaps not everybody works this way, which may explain some of the apparent incredulity re. under and over forecasting on this thread.

 

[franksingleton]

I will put this on two threads as both are discussing the emotive but poorly understood problem of predictability.

First, one of my more pithy remarks. “Weather does not know itself to within one Beaufort force.” This is due to natural small scale structures within the system.

Second is an example that I have often given. Assume straight isobars, E/W up the English Channel. A pressure of 1000 at Southampton and 1004 at St Malo will give an average W F4. If pressure at Southampton changes by 1hPa the wind will be change by one force. In other words a change of 0.1% in the total mass of air over Southampton will lead to a one Beaufort force change in the average wind. Just think about the implications on smaller scales.

I am not defending the Met Office. Through my writings I have tried to goad them into issuing forecasts such as those by Denmark http://www.dmi.dk/en/hav/#danmark , Turkey http://212.175.180.126/DTS/sea.php and others. They will not do so unless someone brings pressure to bear. Who will? Not the RYA. Not the Yachting press.

But, even the most detailed weather models eg http://www.meteociel.fr/modeles/wrfnmm.php?ech=3&mode=3&map=330 or http://www.meteociel.fr/modeles/aromezoom.php?mode=3&x=384&y=384&ech=26&zoom=0 show patterns that are highly smoothed compared to what we experience. The scientific problem is that there are small structures that we can barely and may not be able to observe. These have short lifetimes. A Boscastle type storm might be predictable implicitly 12 or 24 hours ahead but not explicitly until a few hours ahead. Met services are trying to predict such events as severe storms, localised fog for airports and motorways. Details on such size scales could not usefully be included in any broadcast forecast.

It is a combination of a scientific and communication problems. What may seem obvious and frustrating to the user is far less obvious to the scientists but equally frustrating.

If anyone wants to take up the matter of what the Met Office does and does not issue freely to all, then get to your MPs. Make a concerted effort to get more out of them. In the end, it comes down to money and resources. Please all of you that really care keep pushing your MPs. That is probably a request made in vain. Also, even were far more issued we would probably do no better than at present. That is the nature of the problem.

In a post a couple of days ago, I said that we would probably be leaving Dartmouth for Guernsey on Wednesday. As of this morning, that still looks to be the case. On Wednesday morning I doubt whether we will not be really sure e just what wind strengths we will experience.

 

[Cantata]

That's all well said, Frank.
Normally I'd have expected a viable route to change to be through the RYA. Why do you feel that is not the case?

 

[chanelyacht]

the word count, for a start, far too short to allow likely variations across large sea areas to be described.

The word count is there due to the broadcast time slot via Navtex for each service in Europe (due to the distance Navtex can travel). We often used to have to chop off older WZ warnings on the main MSI due to this, especially if the forecast was longer that day (e.g. a lot of gale warnings).

It's not something the UK could change unilaterally - although I would like to see the UK operating a rolling wx service in the way the USA do.

 

[franksingleton]

That's all well said, Frank.
Normally I'd have expected a viable route to change to be through the RYA. Why do you feel that is not the case?

I have tried to raise the issue of the issue of wind forecast charts but had no response from the RYA. The CA showed more interest but, as yet, no intent as far as I know to take any action. I understand that there have been some discussions with the Met Office but have no knowledge of any progress. I have written articles in the yachting press referring to the Danish and Turkish wind charts. I have written a somewhat caustic letter in Weather, a Royal Met Society publication all to no avail. I know the attitude of Treasury to services to the general public and to sectional interests – that includes us sailors. Unless and until someone with a big stick comes along I doubt that there will be any change.

Further, as I have said, even if the Met Office, MeteoFrance, DWD, et etc were to issue more than they currently do I doubt that we would be much the better informed. They would not provide the kind of forecasts that some seem to want. To some extent weather forecasting is the art of the possible.

I would like to see GRIB type output from the Met Office although it may not be significantly better as far as we are concerned than the US GFS. I would like to be able to see their detailed model on the 4 km grid. If we could get continual updated 1.5 km output it would be interesting to know how useful it was. But would I have time to still go sailing?

I am a realist and not an idealist. It is up to us to use what we have as effectively as possible. It has long been my view that many do not do so and some seem to want to be spoonfed. That is one reason why I was pleased when I was asked to write Reeds Weather Handbook. I tried to convey the need for pragmatism when using forecasts to make decisions that affect the safety of us and our boats. My criterion for the value of forecasts is whether or not they keep me out of trouble while sailing where I want to go – as long as I do not have tight time deadlines.

 

[Stoshak]

Well, here is the current forecast for the Channel Islands. You will see that it is produced by someone who is interested in the subject, and keen for us to know what is happening.

Channel Islands - Strong winds are forecast

24 hour forecast: Southeast to east 3 to 5, occasionally 6 with gusts to 35kt for a time late afternoon, backing east to northeast in the evening and decreasing 2 to 3 after midnight, becoming cyclonic variable Tuesday morning. Slight, locally moderate in the north and west of the area at first, becoming smooth or slight after midnight. Scattered thundery showers, locally heavy and prolonged for a time this evening and tomorrow morning becoming misty with sea fog patches overnight. Good, becoming moderate to poor in showers and in mist, and very poor in fog.

Outlook for the following 24 hours: Cyclonic variable 2 to 4 becoming easterly Wednesday morning. Smooth or slight. Thundery showers, locally heavy, becoming isolated by mid-afternoon mist or haze with sea fog patches. Moderate to poor, locally very poor.

On the other hand, here is the forecast for our area. Turned out by the sausage machine.


Selsey Bill to Lyme Regis

24 hour forecast: East 4 or 5, becoming variable 3 or 4. Slight or moderate becoming smooth or slight. Occasional rain, fog patches later. Moderate or good, occasionally very poor later.

Outlook for the following 24 hours: Variable 3 or less, increasing 4 at times. Smooth or slight. Rain or thundery showers at first, fog banks. Moderate or poor, occasionally very poor.


Why do the CIs get a good service, and the rest of get minimalist crumbs?

 

[john_morris_uk]

Well, here is the current forecast for the Channel Islands. You will see that it is produced by someone who is interested in the subject, and keen for us to know what is happening.

Channel Islands - Strong winds are forecast

24 hour forecast: Southeast to east 3 to 5, occasionally 6 with gusts to 35kt for a time late afternoon, backing east to northeast in the evening and decreasing 2 to 3 after midnight, becoming cyclonic variable Tuesday morning. Slight, locally moderate in the north and west of the area at first, becoming smooth or slight after midnight. Scattered thundery showers, locally heavy and prolonged for a time this evening and tomorrow morning becoming misty with sea fog patches overnight. Good, becoming moderate to poor in showers and in mist, and very poor in fog.

Outlook for the following 24 hours: Cyclonic variable 2 to 4 becoming easterly Wednesday morning. Smooth or slight. Thundery showers, locally heavy, becoming isolated by mid-afternoon mist or haze with sea fog patches. Moderate to poor, locally very poor.

On the other hand, here is the forecast for our area. Turned out by the sausage machine.


Selsey Bill to Lyme Regis

24 hour forecast: East 4 or 5, becoming variable 3 or 4. Slight or moderate becoming smooth or slight. Occasional rain, fog patches later. Moderate or good, occasionally very poor later.

Outlook for the following 24 hours: Variable 3 or less, increasing 4 at times. Smooth or slight. Rain or thundery showers at first, fog banks. Moderate or poor, occasionally very poor.


Why do the CIs get a good service, and the rest of get minimalist crumbs?

See post No29 paragraph 1?

 

[franksingleton]

Well, here is the current forecast for the Channel Islands. You will see that it is produced by someone who is interested in the subject, and keen for us to know what is happening.

Channel Islands - Strong winds are forecast

24 hour forecast: Southeast to east 3 to 5, occasionally 6 with gusts to 35kt for a time late afternoon, backing east to northeast in the evening and decreasing 2 to 3 after midnight, becoming cyclonic variable Tuesday morning. Slight, locally moderate in the north and west of the area at first, becoming smooth or slight after midnight. Scattered thundery showers, locally heavy and prolonged for a time this evening and tomorrow morning becoming misty with sea fog patches overnight. Good, becoming moderate to poor in showers and in mist, and very poor in fog.

Outlook for the following 24 hours: Cyclonic variable 2 to 4 becoming easterly Wednesday morning. Smooth or slight. Thundery showers, locally heavy, becoming isolated by mid-afternoon mist or haze with sea fog patches. Moderate to poor, locally very poor.

On the other hand, here is the forecast for our area. Turned out by the sausage machine.


Selsey Bill to Lyme Regis

24 hour forecast: East 4 or 5, becoming variable 3 or 4. Slight or moderate becoming smooth or slight. Occasional rain, fog patches later. Moderate or good, occasionally very poor later.

Outlook for the following 24 hours: Variable 3 or less, increasing 4 at times. Smooth or slight. Rain or thundery showers at first, fog banks. Moderate or poor, occasionally very poor.


Why do the CIs get a good service, and the rest of get minimalist crumbs?

OK. Let me act devil’s advocate.

First, CI CG only have one forecast to broadcast and are allocated the time. HMCG have to schedule several broadcasts om a 3-hourly schedule all round our coasts without the risk of over-speaking each other. Longer forecasts would take more time and delay forecasts to some areas. I was on the MCA MSI sub-committee when the current schedules were introduced but Chanelyacht may wish to clarify that.

Secondly, yes, they use more words but do they really give more information? Is the information any better? In my experience of the area, the answer to both is NO.

In order to meet time constraints of HMCG, NAVTEX and the BBC, the Met Office marine forecasts have been developed to give the maximum amount of useful information in the minimum number of words. As a Senior Forecaster in the Bracknell era, I used to try to ensure that my forecasters only wrote what they were reasonably sure about and avoided unnecessary padding. For example, is it necessary to say 35 kt gusts? There are showers in the forecast. We all know there will be gusts.

 

[Stoshak]

See post No29 paragraph 1?

Not very helpful. Frank's concerns relate only to Navtex and the BBC broadcast. I do not believe the Coastguard broadcasts are constrained and could give more detail. Similarly on-line resources could be more useful.

The insistence on brevity is no longer necessary. Most of us use the Met Office website which has plenty of room.

The RYA, MCGA and the rest of the Establishment seem to have lost sight of the fact that accurate weather forecasts are one of the strategies which are intended to prevent fatalities. My impression is that the Met Office look upon it as one of the less interesting jobs and would prefer to spend the time on more lucrative contracts.

It is time that the whole issue was re examined in an effort to more closely align the product with what is appropriate. The current product is not fit for purpose (except in the Channel Islands).