Has somebody broken the Met Office weather model?

chriss999

Well-known member
Joined
9 Oct 2012
Messages
5,745
Location
Devon, England
sailing-chinee.com
… I would not regard 15 with gusts to 26 as unlikely. I have seen actual reports of gusts being twice the average wind. Given the quality of Met Office scientists, I suspect that their gust forecasts should no be disregarded.

A good source of latest actual data is pened on Sligo Airport. See the image attached. Gusts nearly double the mean wind!
Looking on the met office app just now at my location for tomorrow, the gust forecast averages three times the wind forecast, and this persists for the whole day. At a nearby location at times the gust prediction is 4x. Surely implausible. This sort of prediction tests my faith in the Met.
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
Looking on the met office app just now at my location for tomorrow, the gust forecast averages three times the wind forecast, and this persists for the whole day. At a nearby location at times the gust prediction is 4x. Surely implausible. This sort of prediction tests my faith in the Met.
Oops! As ever, my proof reading let me down. My rather cryptic post was intended to refer to Meteociel - Observations des rafales de vent en Royaume-Uni en temps réel. Use this to look at Sligo a/p. The midnight observation is probably still there.
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
No, not considered unusual (a) in the stormy weather system passing today (b) in Tiree, one of the most exposed / low lying and hence windiest locations in the UK, (c) in October.
OK but I got the impression that some posters do not understand how gusts are formed.
But still doesn’t explain the sudden change in the UKHO gust forecasts in July / August in more sheltered locations such as the Clyde.
I do not know whether or not the Met Office has changed its gust algorithm. I suspect that they have not and that what we have been seeing is a consequence of the very unstable weather conditions with deep convection. Strong gusts in vigorous convection occur in two ways. First, hail or heavy rain, literally, drags air down. That air spreads out as it hits the ground. The heavier the hail or rain, the stronger the gusts. A second mechanism is the overturning of air in vigorous convection. Air that was at 3 or 4 km above the earth is brought down with those speeds. In both cases, what happens at the surface is not greatly related to the pressure gradient, ie the wind strength at the surface between gusts. This is what happens in so-called “super cells” of “downbursts.”

However, I encourage more posters to write to the Met Office. The message might through that some serious minded and observant people have been surprised by recent gust predictions.
Yes there are alsoways some risks of gusts. But it is a waste of expensive super computers to have a next day forecast for weeks on end with wind forecast such as 6-8 mph gusting 20-26 mph.
As noted, I could print that on a laminated sheet of paper and use that as a daily forecast and be accurate 80% of the time. But equally no use to man or beast - certainly not a sailor considering whether to venture out. :)
I think that there is some hyperbole here. Even if the gust algorithms are wrong it is hardly a reason to junk the whole system.
Incidentally, what is the definition of a gust forecast - the strongest instantaneous gust, even if just for a second or so (which may be captured as Wind Max on my boat instruments), or some sort of percentile to avoid over distortion?
My (imperfect) memory tells me that, for aviation and marine forecast purposes, gusts are defined as sudden increases in wind lasting 3 seconds.

PS. See What is a Microburst?
 
Last edited:

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
There have been suggestions that gusts on the Met Office app have been at variance with gusts on UKV output. Predicteind has been mentioned more than once. I have no idea what PW does. I can only say that, over the years, I have never been greatly impressed by their understanding of NWP. However, I have looked at the Met O app around the UK and used Ventusky to look at UKV output. The gusts in the two were similar.
These are for Dartmouth.
 

Attachments

  • IMG_5740.png
    IMG_5740.png
    264.8 KB · Views: 4
  • IMG_5739.png
    IMG_5739.png
    272.3 KB · Views: 6

FairweatherDave

Well-known member
Joined
28 Sep 2009
Messages
2,020
Location
Solent
Visit site
I've used my same rituals checking the same forecasts for sailing and windsurfing for several years. I don’t go to technical or too far ahead. But the Met office is one of my regular checks and my gut feeling is the same as the OP. Something did change and gusts got much bigger relatively to the wind strength and more often. So I appreciate the efforts examining this- I'd like to keep faith in the Met office...don't want to change my rituals .
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
I've used my same rituals checking the same forecasts for sailing and windsurfing for several years. I don’t go to technical or too far ahead. But the Met office is one of my regular checks and my gut feeling is the same as the OP. Something did change and gusts got much bigger relatively to the wind strength and more often. So I appreciate the efforts examining this- I'd like to keep faith in the Met office...don't want to change my rituals .
Computer weather output, whether in GRIB format or provided in some form of app, is the basis for all marine, aviation and other forecasts issued for both our safety and to help us organise our lives. In the case of aviation, airlines worldwide use the same forecast information from either NOAA or the Met Office depending on the flight area. In the marine world use of forecasts is less prescriptive. However, wherever you are, there are forecasts issued under the GMDSS. The name says it all.

Due to the wonders of modern technology, we can see both the basic computer output and the derived GMDSS information. Which you use is a matter of personal choice. NWP is under continual review and development. Weather being what it is, there is no such animal as a perfect forecast. That is well known to the professional meteorologists but glossed over by third party providers which are the de facto sources, as far as we are concerned, of much useful data.

We have the choice of what we use and how we use it. No responsible forecaster makes extravagant claims for their products, No national weather service nor ECMWF claims to be the best. None provides incorrect descriptions about other models. None claims that their models use any unique data or information source. Third party providers would do well to follow suit. Caveat emptor!
 

chriss999

Well-known member
Joined
9 Oct 2012
Messages
5,745
Location
Devon, England
sailing-chinee.com
Are these typical examples, for instance 25% to 50% of all locations on a particular day, or are these readings rarer than that, in that you have to look at quite a few records before you find them?

Yesterday I looked at the met office forecasts for the day for a location local to me which had wind speed forecasts of about 4-6 and gusts of about 16-20. These gusts didn’t happen. Gusts maybe up to 8.

I’ll get together a proper data set and follow it up with the met office.
 

lustyd

Well-known member
Joined
27 Jul 2010
Messages
12,239
Visit site
I’ll get together a proper data set and follow it up with the met office.
They won’t be interested. They are the experts and instead of listening to a valid point they will instead spend their time explaining that you don’t understand how weather forecasting works. They’ll explain that your narrow data set for a specific point is invalid (they’d do this regardless how large your dataset is).

Unfortunately experts have spent a great deal of time learning the status quo and have a vested interest in keeping it. Only one industry is an exception to this, aviation forces open minds through strict reporting standards. Unfortunately this is likely only because the pilots are in danger too. There’s a book called black box thinking that explains this phenomenon.
Generally in areas with experts involved progress is only made through new ideas when the old guard retire.
 

Marsali_1

New member
Joined
8 Jan 2021
Messages
22
Visit site
"Unfortunately experts have spent a great deal of time learning the status quo and have a vested interest in keeping it. ... Generally in areas with experts involved progress is only made through new ideas when the old guard retire." I suppose this means that the experts who warned of global climate change 40+ years ago were all wrong? I can't wait for the current crop of experts to retire then because they seem to be faithfully maintaining the status quo and are clearly wrong according to your standards.
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
Are these typical examples, for instance 25% to 50% of all locations on a particular day, or are these readings rarer than that, in that you have to look at quite a few records before you find them?
I simply used Meteociel - Observations des rafales de vent en Royaume-Uni en temps réel and chose a few locations at random. Many gust/average speed ratios were not exceptional enough were large enough to make some assertions in earlier posts invalid. In other words, gust speeds exceeding twice the eversge wind are not rare.
Yesterday I looked at the met office forecasts for the day for a location local to me which had wind speed forecasts of about 4-6 and gusts of about 16-20. These gusts didn’t happen. Gusts maybe up to 8.
How do you know that they did not happen? I can accept that you did not experience them.
I’ll get together a proper data set and follow it up with the met office.
I suggest that you you look at the Meteociel site or the XCWeather list of actuals.
 
Last edited:

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
Not sure where these are from, but today is a bit exceptional - certainly pretty gusty in Tiree today - but exceptional enough to be a named storm Ashley :)


Bit different from the weeks in summer when this was first noticed.
I have given the link before. Meteociel - Observations des rafales de vent en Royaume-Uni en temps réel. You can try XCWeather as another source of actuals. The data are as reported from stations reporting to WMO standards and taken directly fro the WorldWideWesthrr channels.
I am showing them because some posters were querying that gusts could be so much stronger than average winds. That is so generally as I have said on several occasions.
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
They won’t be interested. They are the experts and instead of listening to a valid point they will instead spend their time explaining that you don’t understand how weather forecasting works. They’ll explain that your narrow data set for a specific point is invalid (they’d do this regardless how large your dataset is).

Unfortunately experts have spent a great deal of time learning the status quo and have a vested interest in keeping it. Only one industry is an exception to this, aviation forces open minds through strict reporting standards. Unfortunately this is likely only because the pilots are in danger too. There’s a book called black box thinking that explains this phenomenon.
Generally in areas with experts involved progress is only made through new ideas when the old guard retire.
Well, I do not know your background. Mine includes weather observing networks, instrumentation and observing standards. Instrumentation and definitions (including wind, of course) have evolved to meet user needs, including aviation as a main driver and definer of needs along with shipping in general, and leisure sailing in particular, as other important customers.

The discussion here began with comments about what some posters perceived to be a change in prediction of gust speeds. As I have explained, these are derived from NWP output using algorithms. Models do not predict gusts. In fact, they do not even predict the wind at the nominal 10 metres. The bottom level of all models that I know about is at a height where the pressure is 0.998 od surface pressure. Algorithms (statistical relationships) are used to give surface wind. If you think that this is a bit odd, then just give a thought to modelling topography.
You will probably say, “Roll on AI.” Maybe you are correct but that will be longer to achieve than you think and probably less than I think.

Be that as it may, I do not know if gust algorithms have changed. My instincts say NO but, I do not know. If they have changed, it will be on the basis of experience and only after a suitably long trial period. From hard won experience, models and model output are only modified after extensive trials and exhaustive testing.

From my experience as a meteorologist and my even longer experience as a sailor, dinghy sailor and cruising yachtsman, I still think that there is a perception problem here. But, yes, provide feedback to the Met Office direct. A long time ago, I was involved in meetings chaired by the MCA between users of marine weather information and the Met Office. I do not know if that group still exists. It would be a slow process but, if enough people really do think that this is a significant problem, then that or a similar group would be a way to proceed. Nothing will happen very quickly. There are far too many users and far too much complexity in both the science and the global organisation required to serve the myriad of users.
 

franksingleton

Well-known member
Joined
27 Oct 2002
Messages
3,585
Location
UK when not sailing
weather.mailasail.com
"Unfortunately experts have spent a great deal of time learning the status quo and have a vested interest in keeping it. ... Generally in areas with experts involved progress is only made through new ideas when the old guard retire." I suppose this means that the experts who warned of global climate change 40+ years ago were all wrong? I can't wait for the current crop of experts to retire then because they seem to be faithfully maintaining the status quo and are clearly wrong according to your standards.
We are seeing clime change happening. Of that there is no doubt. Cleverer people than me were saying that well before I was and even longer before Trump displayed his abysmal lack of knowledge.
 

Marsali_1

New member
Joined
8 Jan 2021
Messages
22
Visit site
We are seeing clime change happening. Of that there is no doubt. Cleverer people than me were saying that well before I was and even longer before Trump displayed his abysmal lack of knowledge.
I fully believe that climate change is real and is being experienced daily by people whether they believe in it or not.

I have spent some time reviewing your website since I posed the question about ensembles and I think I understand your explanations vis a vis gusts and forecasting. I don't think others who have engaged in this discussion have done that and are arguing their point from a place of reduced background knowledge. As you have mentioned in other posts, the science of climatology and meteorology is complex and, certainly, not fully understood by professionals which means that their explanations, even when reduced to day to day language, are not easily understood by lay people or are perceived to be incomplete and, therefore, not to be trusted. Unfortunately the trend seems to be blame the messenger (expert) for their own confusion and/or lack of knowledge.

Thank you for persisting with your posts. There is useful knowledge to be gained by those who choose to accept it.
 

chriss999

Well-known member
Joined
9 Oct 2012
Messages
5,745
Location
Devon, England
sailing-chinee.com
Sure, I accept that a lot of scientific knowledge is needed to understand weather and weather models. But the met office weather app (and other apps) are designed to be used by non-specialists. If they fail to explain likely weather in clear enough terms for the average end user to understand them, then they have failed their main purpose. Am I wrong here?

In my limited sample yesterday I read of wind speeds of 6/8 or so and gusts of 18/20 in that forecast area. The observations I’ve seen show winds in that area of 3/8 and gusts of 8/11. As a weather ignoramus I’d call that gust forecast a fail as it was positively misleading about the likely upper end.

But I’ll take Marsali1’s advice and read Frank’s links to educate myself.
 

lustyd

Well-known member
Joined
27 Jul 2010
Messages
12,239
Visit site
suppose this means that the experts who warned of global climate change 40+ years ago were all wrong?
The majority fought strongly against that actually, climate change took a generation to be accepted by most experts, which is kind of what I said.
 

AntarcticPilot

Well-known member
Joined
4 May 2007
Messages
10,420
Location
Cambridge, UK
www.cooperandyau.co.uk
The majority fought strongly against that actually, climate change took a generation to be accepted by most experts, which is kind of what I said.
Even in Antarctica, where the effects of climate change are strongest, it was not possible to state with statistical certainty that the average annual temperature was increasing and that it was caused by climate change until the late 90s or early 2000s (not sure exactly when!). Pretty much all of us in the Antarctic business (so to speak) were pretty certain of it (I have two papers at least on the topic that indicated long-term change; colleagues published many more that pointed to changing climate - but not at the necessary level of statistical certainty) and we could certainly say that the temperature was rising, but because of the relatively short times series (we only had temperature records back to the 40s) it wasn't possible to say it couldn't happen by chance until the late 90s. As I say, we were all pretty much convinced of it on a variety of grounds, but nothing that we could say was outside the possible natural range. Even convincing anecdotal evidence is just that - anecdotal. Of course, as I worked for a government organization, we had to be conservative in our statements.
 

Marsali_1

New member
Joined
8 Jan 2021
Messages
22
Visit site
The majority fought strongly against that actually, climate change took a generation to be accepted by most experts, which is kind of what I said.
Maybe another way of looking at it is that it took a generation's worth of data to irrefutably show that there was a trend and not just randomness. I do question the implications of this, and your previous post (to which I responded) that all experts are only interested in preserving the status quo until retirement. That being the case, which generation of experts do we believe and which status quo do we choose to accept?
 
Top