I’ve been given a link in email today to a public forecast page for July by weather prognosticator Piers Corbyn, which you can investigate in full yourself here. I find his web pages and forecasts hard to read, and even harder to accept any more, because in my opinion, he presents them like a carnival barker with overuse of exclamation points, bright colors, over bolded texts, random font changes, and fantastic claims. It tends to set off my BS meter like some tabloid newspapers do. Here’s his USA forecast for July:
[UPDATE: 7/8/12 – The full USA forecast has been made available by Mr. Corbyn and is available here for your inspection: http://wattsupwiththat.files.wordpress.com/2012/07/usa-1207-july-inc-public-summary-news-page-full-fc-key-usa-maps-and-extremes-slat8a-prod-29jun.pdf ]
Some people say however, that despite all that unnecessary gaudiness, he makes accurate predictions. Because he’s made a public forecast and advertised its availability, urging “people to pass the links on”, here’s a chance to find out if he demonstrates the skill that is claimed.
He made this bold claim yesterday:
“Terrible weather is coming the world over this July so WeatherAction has issued free summary long range forecasts for USA and for Europe…”
He sounds like Joe Romm or Bill McKibben talking about “climate disruption”. Of course, it could just be another July in the northern hemisphere. Here’s the rest:
The USA pdf link is issued today on July 4th to go with the Europe link issued the day before. We urge people to pass the links on.
“We also expect very serious near simultaneous solar-activity driven deluges and stormy conditions around the world during our top Red Warning R5 and R4 periods. Any communication of the forecasts must acknowledge WeatherAction”
– Piers Corbyn, astrophysicist WeatherAction long range weather and climate forecasters
WeatherAction Free Summary Forecast for July USA:-
“Could it get worse? Yes!” – Extreme thunderstorms, giant hail and ‘out-of control’ forest fires’
pdf link = http://www.weatheraction.com/docs/WANews12No32.pdf
(or no links twitpic = http://twitpic.com/a3y28b/full )
WeatherAction PUBLIC warning Europe July 2012 “Off-the-scale” Flood & Fire extremes likely (WA12No31)
pdf link = http://www.weatheraction.com/docs/WANews12No31.pdf
(or no links twitpic = http://twitpic.com/a3p7pm/full )
The USA forecast map he provides is a bit hard to read, since it seems he scanned it in from print…note the dot patterns in the graphics. I present it here from his PDF page.
Here’s his forecast page for Europe:
He lists “off scale” weather in NW Europe is one of the claims. I wonder how one should define “off scale” weather.
As Carl Sagan once said:
Extraordinary claims require extraordinary evidence
So now that Mr. Corbyn has put forth some extraordinary claims, we can catalog here the evidence to support those claims, and revisit the results at the end of the month. I urge readers to continue to post both pro and con evidence here as the month progresses. I’ll put a link to this thread in the WUWT sidebar so readers can add information that might be relevant.
Since Corbyn is a fellow climate skeptic, let’s give him a fair but factual evaluation to find out if these claims hold up, of if he’s simply following the path of some prognosticators of the past, such as Jeane Dixon, who made claims so broad that even a small kernel of happenstance occurrences after the fact were used to justify confirmation of the prediction. According to the Wikipedia page on Dixon:
John Allen Paulos, a mathematician at Temple University, coined the term “the Jeane Dixon effect,” which refers to a tendency to promote a few correct predictions while ignoring a larger number of incorrect predictions.
I don’t know that is what is going on here with Corbyn or not, but since he’s put out an open 
forecast, let’s find out. Inquiring minds want to know.
UPDATE: here’s a video of Corbyn explaining his methods:
Piers is at his best with the long range big picture of general trends for which in the period May to July he has a record of outstanding success.
Traditional meteorology is quite successful with current 5 day forecasts.
Apparently in Britain for longer term forecasts the Met Office are compelled to use the East Anglia CRU models based on IPCC science
Their attempts at seasonal predictions are a standing joke.
In fact it often seems (like present May to July) that they not just inaccurate but seem to be exactly wrong.
Bryan whilst discussing rainfall you refer to the ‘long term UK average’ then give figures for England showing the average number of days with rainfall of less than 1mm.
Farmer Charlie. said @ur momisugly July 5, 2012 at 8:34 am
Then there’s the David Walsh effect. Every time he walks onto the racecourse, the bookies’ odds shorten dramatically, so he (actually his syndicate) employs runners to place the bets. His winnings finance the largest privately owned art gallery in Australia.
Please note that even though David’s syndicate wins more than it loses, the winnings represent a small percentage of their annual turnover of something north of a billion dollars per year. Similarly, for Piers to win at Ladbroke’s, he doesn’t need to win every single bet. The idea that he is “banned” from betting seems somewhat ludicrous given that Piers could do what David Walsh does: employ runners. I can’t imagine Ladbroke’s banning every single punter (turning away business).
This being such a long thread I haven’t read all comments, but Piers’ forecasts were the subject of analysis in a peer reviewed paper some years ago that concluded his predictions were better than chance.
@Willis Eschenbach says:
July 7, 2012 at 11:17 am
“So … was Piers’ forecast correct? Overall, I’d say no in both cases.”
You need to refer to the “R” periods and not the forecast windows for the rain events. They were June 29-Jul 1, July 3-4 and July 6-7. The heavier UK rain was all within these impact periods. The indicated weather types are clearly what we are getting for the nation as a whole, but the regional disparity is no surprise considering what the NAO has been up to in the last week:
http://policlimate.com/climate/cdas_ao_nao_daily.png
So Ulric, we need to refer to the R periods for rain? If these are wrong we refer to the forecast linked here http://twitpic.com/a4q45r/full and that covers all bases, giving a successful forecast – smart!
Rhys, Bryan, thanks for your useful comments.
I have to say I find it pathetic that some ‘mark’ our long range forecasts – on which it (eg for Brit+Ire) explicitly states “Likely possible weather map scenario…; +/-1 day and gives a confidence (B=75% in case of 1-4th July) – against final outcomes and say therefore we are ‘wrong’.
Its like complaining a punter who got the winning horse right but didn’t get the color of the jockey’s socks right.
By the standards these churls apply all forecasts are ‘wrong’ unless perhaps the UKM0 ‘warnings’ of events as they happen.
The question is:
a) compare us with whatever anyone else said that time ahead [and NO the July forecast was made 15th June and any changes very minor on 28th (issue meaning desk top publishing) re interpretation of maps already defined unless we changed SLAT procedure which we did not in this case. If there are changes we say so].
b) compare with say the last ten years of July periods (eg 1-4th) July.
To set up a competition between M0 one day ahead (or nowcasts) and our long range (which also include much longer ahead at times) is really deceitful.
[And here I jest not; we experienced one smartie, years ago, judging our forecasts from weeks or months ahead against MO one day ahead (that being defined as what happened). He found the M0 one day ahead was closer to itself than we were to it! Wow]
Just be clear WeatherAction forecasts are about getting the best long range picture for useful decision making and they work and win on scientific bets. Farmers use them to see around when is there a chance of dry time for harvest / rain for… etc
In terms of the first week of July Brit+Ire we have had a useful few reports on this on http://www.weatheraction.com/displayarticle.asp?a=471&c=5 – readers comments and note many said 1-4th or 1-5th allowing one day was a very good forecast – see Louise Woods on facebook
Facts:
(i) There was hail
(ii) there were floods – Barnsley (eg)
(iii) there was a tornado on 5th in Yorkshire http://www.youtube.com/watch?v=pcFhC18XK-4
Did MO expect tornadoes?
“Oh”, the standard Met nerds go on “ah youve given yourself one day” – from say 10 days ahead (or it might be months ahead) (and they give themselves more than 2 or 3 hours from one day ahead).
Let’s be clear WE WRITE THE FORECASTS and DEFINE what we mean. We can give as many days uncertainty as we deem appropriate. People re-defining what we say and mean in order to denigrate is totally unacceptable.
Principles
1. Like must be compared with like.
2. Any forecast must be considered as stated with all provisos and compared with all other reasonable possibilities.
3. Betting style measures of skill are the most meaningful.
Thanks PC
@Martin Gordon says:
July 6, 2012 at 1:39 pm
“..the fact that he doesn’t release out of date forecasts for evaluation says everything that you need to know.”
I spent a couple of hours yesterday to see how many long range forecasters have a forecast archive available on their site, I only found one that does: http://www.weatheraction.com/pages/pv.asp?p=wact46
Ulric you might have a peek at my website ^ the maps posted for May, and June, are still there and if you use the way back machine you can find the past four years as well, (due to the upgrade of my site and transfer to commercial server past maps prior to May were no transferred over.)
@ur momisugly Paul Vaughan
Not only does Piers need a CEO, he needs the services of an archivist. That costs both time and money. Difficult when you’re reacting to a constantly evolving situation.
I had the pleasure of meeting Piers and Anthony on the same day in March of 2008. I was impressed by both individuals. Anthony’s presentation of his station site analyses was quite fascinating. Piers was not on the agenda but got about 15 minutes to make a short presentation quite apart from his forecasting skills. He did offer a forecast of a dreadful snow storm that would set all time records in the mid-west around the middle of the month with a +_1 day tolerance. That was in 2008 and he was spot on.
I had received a few forecasts starting in December 2009. He forecasted a snow storm for New England the day after Christmas. He said it would be far worse that standard meteorologists would claim 1 to 2 days in advance. On Chrismas Eve, the forecaster’s predictions went up faster than Hansen’s Hockey Stick. Piers was 100% correct. We had not sen that kind of snow ever! During January, after telling several about the Christmas prediction, people were asking what is Piers saying? He was spot on for the next 3 super storms using the same description and each was on the day he said it would happen. My house in Connecticut was nearly buried by these storms. Very unique in my 65 years on the planet.
I will say that he did forecast a 4th storm and it happened but went just south of us. Not a bad stretch.
Since then I have occasionally seen his forcasts with some mixed results. Sometimes things happen and are not on his maps. I asked Piers about this. He told me that if there was a high uncertainty of an effect, he would just leave it out. The result is some severe weather that he does not predict. The idea of hisextreme forecasts is to identify those events that are most likely to occur. It does not mean that he will get them all. If he does not chose to forecast an event, he does not consider this a miss. He does use these occasions as learning moments.
I have recommended to Piers on several occasions to acknowledge his “mistakes” and perhaps explain what happened, assuming he knows. I have noticed this happens more often these days.
I just checked his forecast for a day in late June comparing the current satellite map to his forecast map. There was an uncanny resemblance of one to the other. The July weather forcast for the planet is pretty wild. It will be interesting to see if it happens. I read that one comment said he was at odds with the skeptics by forecasting extreme storms. He is forecasting events with an assigned cause. Solar activity. There was a piece on the news tonight of significant solar storms. That can’t be good. Warmers are making bold claims with non-specific information with no time frame. I would bet on Piers to get the big ones right. The rest doesn’t really matter.
@Willis Eschenbach says:
July 5, 2012 at 10:16 am
“I don’t understand how I’m supposed to tell if Piers is right or not. He only makes four “forecasts” that are so vague that Nostradamus would be proud of them:”
I would pay particular attention to the R periods. I’m not aware of anyone else connecting solar activities with the propagation of simultaneous weather events globally, and mapping these out months ahead to within a day. After seeing every UK forecast since July 2007, I would conclude that all meteorologists will be employing this method in the future.
If he was correct, I would not care if he wrote them in crayon.
John Wright (July 7, 2012 at 5:03 pm) wrote:
“Not only does Piers need a CEO, he needs the services of an archivist. That costs both time and money. Difficult when you’re reacting to a constantly evolving situation.”
Last sentence: Yes.
First sentence: With an infinite amount of resources: sure.
But with a tight budget & small staff, it might not be wise deflecting resources to cosmetics. (I see your middle sentence.)
Some might argue it would attract resources, but that sounds like a gamble that might clock Piers out, causing him to fail to reach his full learning potential during a finite lifetime.
I’m sure Piers has a better sense than us of what might be best for optimizing the lifetime learning that he passes on. I wish him efficiency.
Bryan says:
July 7, 2012 at 12:16 pm
Thanks, Bryan. Always glad to have someone investigate my work. Here’s my previous graph recalculated as an average rather than a total.
I couldn’t find the rain data at the link you gave, but it is available here. Here’s the data, comma-delimited, to cut and paste into a spreadsheet.
The last column is the average rainfall on rainy days.
You also say:
It doesn’t matter how far in advance he made the prediction. Regardless of when he made the prediction, he gave separate and distinct forecasts for the periods July 1-4 and July 5-7. I am treating them as forecasts for that time and no other. If he is really talking about “July 1-4 plus or minus six days”, he needs to say so. Until then, I will take him at his word that there is one forecast for July 1-4, and a separate forecast for July 5-7.
I am aware that the forecasts are made months in advance. But he is the one specifying the dates of the forecast intervals, not me. If he wants wider intervals to be counted and included, he needs to specify them. You can’t claim them after the fact as you wish.
Perhaps he is, and I’d be overjoyed if that could be shown to be true. However as in this case, sometimes when we actually compare his forecast with the observations, they are wrong. In this case the intervals on either side of the forecast “Heavy rain with thunder, hail and floods” for July 1-4 were wetter than the forecast time of his previously predicted pluviation …
And sometimes he claims successes when none are there, as when he has predicted wildfires in New Mexico and Arizona, and then taken credit for wildfires in Colorado.
Look, I wish Piers well and I’d love to believe … but I don’t believe anyone until the results are checked. I checked this one. I’m not impressed with this particular forecast. What can I say? I just follow the observations.
w.
Piers Corbyn says (in reference to the forecast period 1-4 July)
“Facts:
(i) There was hail
(ii) there were floods – Barnsley (eg)
(iii) there was a tornado on 5th in Yorkshire http://www.youtube.com/watch?v=pcFhC18XK-4 ”
I cannot find any flood reports for Barnsley for that period, levels on the River Dearne were below flood level.
Having watched the video there is not enough detail to confirm it was anything more than a scud cloud.
The forecast indicated that *most* of the UK would be suffering from heavy rain and floods.
David.. “I haven’t yet seen him put his hand up and say, ‘I got that one a bit wrong’.”
You’re not looking very hard David; yes he does!
>
>
Pokerguy.. “I really dislike this style of forecasting as it gives the impression of rooting for catastrophe.”
I suggest you borrow a dictionary and look up the word ‘catastrophe’ !
Plenty of catastophes around the world due to flooding because of heavy rains right now. Hundreds of deaths, thousands of homes and businesses flooded out, millions of $$ worth of stock and furnishings destroyed. Millions homeless or without power for weeks. Thousands of livestock and masses of wildlife drowned. Is this not catastrophe?
>
>
Lex.. “His forecast were all extremely incorrect….”
I find that very strange, because I have monitored this past years forecasts and I found that they have been extremely(?) correct. Maybe Piers personal learning curve is shaped like a hockey stick!
>
>
AJB.. “Doesn’t that mean Piers must be able to predict solar flares and coronal holes?”
Yes it does, and he has the scary ability to do just that. I’ve been an active amateur astronomer for over 35 years and I am truly shocked at how Piers can predict periods of solar activity.
>
>
P. Solar.. “False logic. Have the Met Office gone out of business?”
The Met Office are paid £170 million a year by the government in compulsory tax payer donations.
Where’s your logic?
>
>
MikeA.. “I have to admit he fits the Steve Jones description of ‘Pratt.”
Oh lord here we go with the ad’ hominem attacks. I knew the serene atmosphere of this thread couldn’t last forever.
>
>
Pamela Grey.. “Often severe weather occurs at night.”
…and? Are you saying that the sun switches off at night?
>
>
Willis Eschenbach.. “..and I don’t have the $$ to do so..”
What a cop out! For less than 3 gallons of gas/petrol for a whole months forecast? You have got to be kidding me? Buy a whole years forecasts and they work out to around 2 gallons per month.
When we bought the full years 30 day forecasts they cost about the same as just one weeks supermarket food shopping.
Also
“..other than mathematical theorems nothing can ever be proven in science..”
Water is wet and destructive…falsify that! (hehehe)
Also
“he NEVER FORECAST FOREST FIRES IN COLORADO”
As I remember it, Colorado is joined at the hip to New Mexico.
In the meantime:
http://fires.globalincidentmap.com/home.php
Click on Arizona and scroll down the pop-up, then read the date. Case Closed!
>
>
Doug Proctor.. “Piers seems to be out there with crystals.”
Nasty! Of course geologists have a complete understanding of their chosen subject don’t they Doug? Ahem!
Is Richard Holle using crystals too, because using your logic he must be?
>
>
Louis Hooffstetter.. “Anyone who claims that climate influences earthquakes and/or volcanic eruptions is a witch doctor, not a scientist.”
Without having to re-read this whole thread, could you point out the post where this was proposed please Louis?
>
>
Richard Holle..
Well, you are certainly on the right track with your research Richard.
>
>
Carrie.. “I don’t like it, this all feels a bit school playground’ish to me. Sorry I thought WUWT was better than this”
I totally agree.
Eccentric? Can’t build a fashionable website? I believe Einstein was called eccentric and was an awful cook, plasterer, bus driver and nanny!
>
>
Pertinax.. “Again from my operational experience (for the US at least) Piers’ forecasts are at best worthless.”
I found your post intriguing up to the point you said “Piers forecasts are at best worthless”. At that point I lost all faith in your credibility.
It doesn’t matter how far in advance he made the prediction. Regardless of when he made the prediction, he gave separate and distinct forecasts for the periods July 1-4 and July 5-7. I am treating them as forecasts for that time and no other. If he is really talking about “July 1-4 plus or minus six days”, he needs to say so. Until then, I will take him at his word that there is one forecast for July 1-4, and a separate forecast for July 5-7.
Willis, did you read Piers’ reply?
I think it does matter how far in advance he makes these predictions. If he says a significant weather event will occur, and give a confidence level for when it occurs, I think it only fair to judge him on 1) whether the event occurred within a reasonable period of time and 2) the accuracy of the timing.
If he is able to predict a significant weather event well ahead of time when no one else is, then it somewhat misses the point to use the accuracy of the timing as a way discrediting the forecast. If he says there is an 80% chance that the significant weather event will fall within this time frame, and you only look at one example, then you have a 1 in 5 chance of picking an event he will get wrong in timing.
That’s not entirely fair I would submit.
He is saying well ahead of time, expect shed-loads of rain with the risk of flooding at around about the period 1-4 July, and it turns up 5-7 instead, then you ought to be able to say that he predicted the event accurately but not the timing. From a real-world usefulness perspective, I think it still important to know that you are going to get a deluge and prepare for it accordingly, even if the exact period it occurs is slightly out. That he got the deluge correct in his predictions a month away is pretty significant IMO. He didn’t say ‘a bit of rain’ or ‘the possibility of rain’, he said it would effectively p-ss it down and it most certainly did.
So if it transpires he is predicting these events well ahead of time, and despite other confounding factors and the possibility that timings may be off at times, then it stands to reason there is something to his methods. Like a lot of commentators I have a hunch that he is onto something. Some one like you would be ideal to take a serious look at his accuracy of his predictions, taking into account confounding factors such as the system he uses might not be perfect but still have interesting and useful skill that could be improved. If you were to throw yourself into a skeptical analysis, I’m pretty sure I would not be lone in following it with immense interest.
Mr Watts, you are using cheap shots. Not difficult in this case, I know, but a pity.
Corbyn is a member of the Royal Metereological Society and he called on that body to organise systematic trials of weather forecast skills. That is clearly the way forward. It will involve a clean definition of methodology and statistical analysis. Your once off test has no status. FWIW, I first started taking notice of Corbyn when he correcly predicted the tornado on Helgoland July 12, 2010. What does that prove? Nothing at all, any more than your ignorant and arrogant polemic.
Regarding earthquakes, I have done a little work on this, and believe there may be a correlation beween deep quakes and the fast solar wind, modulated by the moon, as Corbyn predicts.
However, there is a lot to do before that postulate can be confirmed.
If you want to test Corbyn, please do it properly or not al all.
REPLY: My goodness, you sound like one of the people that regularly defend Michael Mann. I don’t care if Piers is a member of the International Society of Dog Catchers, or a Noble Laureate, or Holder of the Sacred Chalice of Rixx. If his work can’t withstand some scrutiny then it isn’t of much use to anyone. I’ve emailed Piers privately on many occasions that his presentation is subpar and tried to help him produce a more professional product. He’s ignored me. Personally, I think it looks like weather astrology the way it is presented. You are welcome to get your own blog and do/say/test whatever you wish. The post stands. – Anthony
Agnostic says:
July 8, 2012 at 6:03 am
Yes, I did read his reply. I also read his prediction, which is why I don’t have a clue what you mean when you say “If he is able to predict a significant weather event well ahead of time” … did you notice that his prediction was made on the 28th of June? That’s a whole three days in advance of the start of the period he is forecasting, which was July first to seventh.
No, in this case he is not saying anything “well ahead of time”. He made his prediction a mere three days before the start of his forecast period. From that distance, if you are out by three days in your forecast, it’s a joke.
A month away? A month? It was three days away, and he still got it wrong. Nor was it a “deluge”, it was one day of rain on the 7th which was about average for a rainy day in July, plus less rain on other days. In addition, the interval predicted for the “deluge” was drier than the period before and the period after. How you count that as a successful prediction is beyond me.
I am taking “a serious look at his accuracy of his predictions”, and I’m taking a lot of flak for doing so. Is there “something to his methods”? Quite possibly, as I’ve said from the start. It’s just very difficult to determine, for reason I explained above. One of these is his habit of claiming that close is good enough.
You don’t seem to understand that if the Met Office says it’s going to be a nice weekend and it rains instead, nobody says “well, they predicted nice weather and it was nice on Tuesday, give the Met Office a break, they were close.” Why on earth should we not hold Piers to the same standards to which we hold the Met Office?
Look, I’m willing to give Piers as much slack regarding the time interval as he asks for. But he has to ask for it. If he’s predicting someone for a four-day period, and then something else for the following three days, then that is what I will judge him on. He is the one that is setting up the dividing lines, not me.
Now, as you point out, the further ahead you predict things, the less accurate they are likely to be. But that’s Pier’s lookout, not mine. If I predict right now, this minute, that there will be storms on July 3rd of next year and that the 4th of July will be clear and people can barbecue on Independence Day, then that is my prediction.
And if it turns out that the 3rd of July is clear, and that everyone who set up their barbecues based on my prediction gets rained on, then my prediction would be wrong. Not close. Not “cut me some slack, I predicted it a year ahead and it was clear on the day before the 4th.” I would be wrong, and people who depended on my prediction would justifiably be upset.
Look, if someone is foolish enough to make predictions for a three-day window two months from now, or one week from now, then that’s their business. And if they do so, then that’s what I will judge their results on. Those three days. If Piers want to be judged on getting it right within ± 3 days as you claim in this case, then he should have made it a ten-day window and not a 4-day window.
It’s like Piers forecasting forest fires in Arizona and New Mexico, and then claiming success because there were forest fires in Colorado. Sorry, but if he meant “the American Southwest”, then he should have said so, and he would have been right. He didn’t do that, he said New Mexico and Arizona. If you do that, I don’t care if there are fires in Colorado, or Utah, or California, or Texas, or Nevada, or Oklahoma, all of which adjoin the predicted states. If the fires don’t happen in New Mexico or Arizona, you don’t get to say ‘oh, I really meant fires in those two states plus the six adjoining states, so I was 100% right’.
Because once you start down that road with the nonsense about ‘he predicted a rainstorm in the period from the 1st to the 4th, and there was a windstorm on the 7th, that’s close enough, he was 100% right’, then just about any prediction is correct.
Me, I assume Piers means what he says, and nothing else. If he says tornados, I assume he means tornados, not rain, not snow, but tornados. And if he says the first to the fourth of July, I assume he means July 1-4, not the 7th. If he wants rain on July 7th to be included as a correct forecast, then he needs to include July 7th in his forecast interval.
That’s why, when Piers offered to bet on whether “the Olympic opening ceremony in London on 27th July will suffer disruptive downpours etc.” I said I’d be glad to bet US$100, but as part of the bet he had to specify exactly where and when and how much downpour there would be. I don’t want him claiming success because there were downpours the day before or the day after, or because it rained in Colorado, or because there was a mild rain on the Olympic Stadium. Predictions need to be specific to be falsifiable.
w.
… did you notice that his prediction was made on the 28th of June? That’s a whole three days in advance of the start of the period he is forecasting, which was July first to seventh.
No mate – he made his prediction at the beginning of June and updated it on the 28th. Read his reply again. He made no major changes to the prediction he had issued a full month earlier. You could criticize him for not adjusting the exact time closer to the time, but I don’t think you can call him out for having gotten the prediction wrong within such a close period time so far ahead. And as I said sometimes he does get the timing right. Whether the exact period is correct or not is not as interesting as him making a specific prediction about a significant weather event. We aren’t just talking about a bit of rain on a day you might want to BBQ, we are talking a serious weather event that causes floods.
Now, as you point out, the further ahead you predict things, the less accurate they are likely to be.
What do you mean by accurate? If he had predicted snow, or a heat wave, or strong winds, then you could say he had been inaccurate. But he predicted deluges and they did occur albeit later than he thought. And you are looking at a very isolated period. I submit to you that whether he is accurate to within a week is not as significant as his being able to detect an event so far ahead of time.
You are really good at skeptically looking into claims in scientific papers, and checking peoples sums and assumptions, but the big issue here is whether or not Piers can reliably detect significant weather events well ahead of time, and do it where others cannot. It seems to me anecdotally he might just be able to do that, and if he can then that’s really very interesting. I just getting caught up with how accurately in time he predicts them (within reason obviously) is missing the big point. Accuracy with timing is interesting too – but it should be looked at separately.
@ur momisugly Agnostic
To be accurate, the forecast is an update of the forecast made on the 15th June.
As Willis Eschenbach says, what is the point of including dates in the forecast if those dates can be ignored. Widespread flooding was forecast for 1-4 July from a forecast made 2 weeks before and updated 3 days before. Flooding actually took place on the 7th/8th July during a period forecast to be showery.
Russ says:
July 8, 2012 at 5:51 am
Cite?
Cite?
Cite?
Russ, you need to understand that this is a scientific site. As such, your claims that Piers is right and can predict solar flares and coronal holes are interesting but anecdotal. Do you have, for example, Piers’s last ten predictions of the “solar flares and coronal holes” so that we can compare them to the actual observations?
Best regards,
w.
Agnostic says:
July 8, 2012 at 10:52 am
Thanks, Agnostic. I read his reply correctly the first time. It doesn’t matter if it’s an update of a forecast from last year or last week. The actual forecast that we are looking at, the one that is quoted from and being discussed by a number of people above, was made just three days before the start of the forecast period.
So it is not a long range or months-ahead forecast as you claim. The forecast we are discussing was written on June 28th. It was not written, as you keep claiming, a month in advance. That was a different forecast, one which we (or at least I) have never seen, and which is superseded by his later forecast.
It’s the same with all forecasts. Here, the TV weather guy typically issues a forecast for the coming week. Then he changes the forecast as the time gets shorter and shorter. The most recent forecast should presumably be the most accurate, and it supersedes earlier forecasts.
Now, that most recent forecast is, as you point out for Piers, an “update” of the TV guy’s earlier longer-range forecast … so what? The most recent forecast should be the most accurate because it was made most recently. The fact that an earlier longer-range forecast was made by the TV weatherman means nothing regarding his most recent forecast. Nor can the TV guy go back and say “well, I said yesterday that today would be sunny and it rained, but I’m still right because a week ago I said it might rain today”. The more recent forecast replaces the older one.
Similarly, Pier’s most recent forecast, made three days before the start of the forecast period, should be the most accurate. Why? Because it wasn’t made a month before the forecast period, it was made three days before the forecast period.
And since it was made three days before, a three-day error means that it is way wrong. If on Wednesday the Met Office says the weekend will be sunny, but it rains on the weekend and doesn’t clear up until three days later, you don’t say ‘the Met Office forecast for the weekend was right because it was sunny the following Wednesday and Thursday’ … so why are you saying that for Piers?
If Piers wants to break out his earlier one-month ahead forecast, we can discuss that, but at present, all we have to discuss is the three-day ahead forecast.
Finally, I say again that if Piers wants rain on the 7th to be counted as a successful forecast, he has to include the 7th in the forecast period. You can’t predict rain on the 1st to the 4th and claim success because it rains three days later, that way lies madness.
w.
Agnostic says:
July 8, 2012 at 10:52 am
Look at the data, Agnostic. Here it is again:
He predicted “heavy rain with thunder, hail, and floods over most parts of Britain and Ireland”. But nowhere in the data is there a significant heavy rain event. The largest rains in the entire period are just average rainy days, they are not deluges. They barely exceed the average July rainy day. And the rest of the days are, well, just above or below average for July. So I fail to see the “heavy rains” he predicted. The issue is not just the timing, as you keep saying. It’s that even the wettest day (July 7th) is just an average July rainy day … and even that didn’t occur when he predicted the “heavy rain”.
w.
@Willis Eschenbach says:
July 8, 2012 at 10:35 am
“… did you notice that his prediction was made on the 28th of June? That’s a whole three days in advance of the start of the period he is forecasting, which was July first to seventh.”
Piers says:
“[and NO the July forecast was made 15th June and any changes very minor on 28th (issue meaning desk top publishing) re interpretation of maps already defined unless we changed SLAT procedure which we did not in this case. If there are changes we say so].”
Willis says:
“Nor was it a “deluge”, it was one day of rain on the 7th which was about average for a rainy day in July,”
Maximum was Leeds at 57.2mm, several locations had around 40mm, you know where the data is.
Willis says:
“If Piers wants to break out his earlier one-month ahead forecast, we can discuss that, but at present, all we have to discuss is the three-day ahead forecast.”
No you don’t, he said in his own comments above that he made no changes to the 45 day ahead forecast.