Exceptionally hot day at Gravesend Met Office site

Posted: November 1, 2014 by tchannon in Analysis, Surfacestation, weather

Last day of October 2014 (31/10/2014) was lovely and warm over much of eastern England, particularly the south east. Mostly weak sunshine and a slight southerly wind. The Gravesend site has in the past been the subject of questioning why it is so warm, estuary, not in town.

There was a south wind too in the Cairngorms where the temperature gradually rose all day, with a twist, the wind was rising from 48 mph to 60+ mph, gusted to 84 before last thing the wind shifted and fell. This suggests a general flow.


This confusing plot shows an unusual situation. The highest temperature was recorded at Gravesend, logged by the Met Office as 23.6C, 0.3C above the hour mean. The oddity is this occurred an hour earlier than the other group of stations, or perhaps is an effect of quantisation.

Temperature does run east to west with the sun but not that much.

I think the explanation was high cloud which can be deduced from the full data although Gravesend is not a full station.

In the past the Met Office have claimed the cause is a Föhn over Kent / Sussex. With difficulty (web site is broken) I managed to get the regional extremes.

Highest rainfall Stowe 0.2 mm

Which is a trace, dew probably and miles away to the west, so whatever caused it wasn’t that. Neither could I find particular hot air flowing across the English Channel from France. (included Shoreham and representative)

The second set of warm readings, Kew, which is not a full station and the airport pair Northolt / Heathrow.

A weird thing I’ve spotted is a single gust reading at Bedford which I took to be an aircraft or some other local happening but Church Lawford not far away also has a gust the same hour. This leads me to wonder whether some kind of local pressure centre was at work.


Plot of station locations copied from the PDF of all stations. (1.4MB)
(time is short, might add a lookup as an update, Gravesend is 3784, you can lookup in the PDF or CSV)

Any ideas anyone? Not particular important but I like explanations.
And all the data as CSV inside ZIP (59kB)

[update, forgot this]

UK extremes

Parameter Location Value
Highest maximum temperature Gravesend 23.6 °C
Lowest maximum temperature Lerwick 11.5 °C
Lowest minimum temperature Altnaharra 6.9 °C
Highest rainfall Lusa 20.6 mm
Sunniest Hurn 7.2 hours

Issued at: 2303 on Fri 31 Oct 2014


Post by Tim

  1. Scute says:

    Gravesend, Northolt, Heathrow. All the usual suspects, then.

  2. vukcevic says:

    It’s all that hot air from Brussels blowing in across the Channel.

  3. Stephen Richards says:

    It’s always a puzzle to me why the UK gets such high temps when the air is coming from us in France. Yesterday we had (SW France) 27°C which as the air moved north gradually reduced but on reaching gravesend etc it then suddenly increases. It arrives on the sth coast at about 19°c having left france at a max of 23.9°C. You then get a plume NNE of London, obviously, but no reason and no sign of heat generation at Gravesend. There is a power station there somewhere ?

  4. wayne says:

    “This leads me to wonder whether some kind of local pressure centre was at work.”

    Guaranteed… or a local drop in density… or a combination of the two explaining the higher temperature for T = P/(ρ·Rs) and Rs is a constant. Once I finally learned this relation and how important it is and firmly planted it in my mind so many of the strange and mysterious occurrences in temperatures here in my neck of the woods are so simply explained. Here we often have micro-bursts, medium to high velocity but very large area downdrafts that momentary press on the surface and raises the pressure for a few tens of minutes and by that equation, yes, the temperature does take a sudden peak. Secondary, this downdraft causes movement of wind radial outward, divergence, that also causes the density to drop momentarily, adds to the same effect, temperature rises instantly.

  5. tchannon says:

    oldbrew, thanks for adding a link to older content.

  6. michael hart says:

    Maybe there is a curse of Pocahontas?

  7. tchannon says:

    I think what you are talking about is micrometeorology, which is plainly obvious to many practical people but doesn’t make much running otherwise.

    It might indeed be there are wider area effects, why I mentioned a pressure centre, little different conceptionally from a cloudburst here but not there, branches ripped off here, just a wind there.

    The infamous **Brogdale** is close to the Gravesend station (which is not in Gravesend where there used to be a station). Brogdale was within a horticulture microclimate and human incompetence climate, 2003 a probable microevent again. (I included Kenley with missing data for that reason, radio links break under ionising events). I have the evidence to pour heavy scorn on Brogdale but not copyright release for site photos. Some of this is new information I unearthed in addition to that.

    Worse, there are site cessations, removing evidence. This included withholding data and then station removal.

    I included Wisley because it is a dubious horticulture site and several others of that nature, with near controls. Nothing much shows.

    Right now by best guess is 1/f at play.

  8. tchannon says:

    New snippet.

    Gravesend is Gravesend-Broadness located on a loop in the Thames estuary, former tidal marshes.

    The tide was turning at about 12:00, low tide. Roughly at neap tides.
    http://www.pla.co.uk/Safety/Tide-Tables (PDF for above and below)

    Lack of hourly is a problem…
    Same here

  9. Phill says:

    Had a fiddle with the data. I took the 17 stations +/- one degree from Gravesend. Gravesend starts with the 3rd warmest minimum, behind Woburn and Shoreham so it is starting off with a head start on most other stations. Shoreham is on the shore facing south so with a southerly wind overnight temperatures are moderated upward by the sea (and downward during the day). Gravesend is very close to the sea and similarly stays warmer at night, but with a Southerly wind acts like an inland site during the day. It has the 7th largest range so even though it was the hottest it is not the place that warmed the most. That prize goes to Northolt that warmed by 12.1 degrees compared to only 9.7 at Gravesend. On this basis, and on this occasion, Gravesend was the warmest because the proximity to the sea kept it warm at night and then the sun warmed it exactly as it does other inland sites.
    IMO the Met are partially correct about a Föhn effect. As the warm air crosses the sea the lower layers cool with contact at the surface but as the airmass moves inland it remixes with the higher and warmer and drier air. During the early afternoon the humidity % in the data is considerably lower around London than it is at the south coast.
    The whole area of climate studies is plagued by issues of micro climates. The data indicates that the wind was very light at Gravesend on this day which allows greater scope for local effects. Just when most other places recorded their hottest temperatures the wind speed increased at Gravesend and the temperature dropped slightly. Apart from the usual issues of buildings, concrete and soil structure which can retain, reradiate and reflect heat, there are also issues of the chemical reactions going on inside the local vegetation. During the photosynthetic process plants literally gobble up a % of the incoming photons. These are simply taken out of the equation and used by plants to grow. A photon captured and stored in vegetation has disappeared, perhaps temporarily, out of the system. It cannot heat anything unless the plant is digested or burnt. The overall efficiency of photosynthesis is small, probably less than 5% but together with the cooling effect of transpiration adds another layer of complication.

  10. tchannon says:

    Updated the main plot… x-axis scaling had been lost during graphic export.

  11. Kelvin Vaughan says:

    Doesn’t this site have a high metal fence all around it?

  12. Scute says:

    Remember that Northolt Littlebrook power station bathes the Gravesend site on three sides with warm water from its outflow into the Thames (1 mile upstream).

    I notice that, as in the 2012 post on Gravesend that oldbrew linked, there is an uptick in the early evening just when the power station kicks in. I commented on that at the time in response to Anthony Watts suggesting a look into it. I basically verified his suspicions by checking on the Northolt power station site that Tim had linked and put a few numbers on it. I seem to remember that there was some discussion in the comments that the 6PM uptick was quite early because the 2012 record was on a Saturday and that it would probably be a little later if it had happened in the week and was related to the power station cranking up its output. Indeed, this time it was a weekday and the uptick in temperature was at 7PM, just when you’d expect it.

    However, I am aware that quite a few records show an uptick in the early evening (Kenley was one I noticed at the time) for some reason so maybe it’s not as clear-cut as it would appear.

  13. Scute says:

    Correction to above: I mean Littlebrook power station not Northolt.

  14. tchannon says:

    A very good comment Phil.

    The matter of the response of vegetation or for that matter soil is important and in part why met sites need so much clear space and contiguous ground cover. Not so at this site.

    A peculiar situation.

  15. Tonyb says:

    For what it’s worth just an hour earlier I heard the BBC say a new record of something like 21.7 had been set at somewhere in Surrey. Why this gravesend reading should have come so late in the day from nowhere I don’t know.


  16. tchannon says:

    Charlwood, Surrey. BBC didn’t check all likely stations.

    Both Charlwood and Gravesend-Broadness are new stations with short records which set records on the day they opened, of course.

    Therein lies the silly matter of apples and oranges. Nothing must change about a station but it does. We compare different situations.

    There are many obsolete sites holding extreme records. What they would have produced if nothing had changed since is unknowable.

    If I had the hourly AWS from it’s inception I would look and see how far out of exceptional the reading was taking into account inherent cyclic variation, This can only apply to a specific site, compare with others is another worms can of.

    More notable is why there was a general warm flow. I did read this from published GCM output
    https://tallbloke.wordpress.com/suggestions-8/comment-page-1/#comment-91493 but only mentioned the possible storm front. This was preceded by warm bands.
    This actually fizzled to much less dramatic

    2014-11-02 zulu, what is left of it is showing the shear zone over Irish sea.

    Craig posted and also commented on Suggestions alerted me to there being enough interest for a post. http://weatheraction.wordpress.com/2014/10/31/hottest-halloween-on-record/

  17. Graeme No.3 says:

    I rather like the idea of the Föhn effect operating as the winds descend from the Alps of Kent.
    They could be there, after all we people who spent years creating mountains out of molehills, so a whole new range is not beyond them.

  18. tchannon says:

    Just a few sites again and again

    UK extremes

    Parameter Location Value
    Highest maximum temperature Gravesend 16.9 °C
    Lowest maximum temperature Dalwhinnie 8.0 °C
    Lowest minimum temperature Lough Fea 4.8 °C
    Highest rainfall Achnagart 41.6 mm
    Sunniest Edinburgh Gogarbank 5.4 hours

    Issued at: 2303 on Sun 02 Nov 2014

  19. tchannon says:

    I have unearthed significant new information about the Gravesend site. If I can put together the graphics will be a new article.

  20. tchannon says:

    Not that anyone is watching this far off the front page, I’ve written some initial code to extract hourly data as a time series from the daily CSV. Not as simple as it might seem. (about missing data)

    A quick check shows no obvious tidal period in the data, although checking with wind direction would be a tricky problem.

    More data integrity checks to be done, don’t like the looks of some of it, might be a time error. There again using a party trick there are surprises such as an abrupt change I think 4th Oct, sudden change in the weather. A quick web search suggests this was noticed and is correct.

  21. Scute says:


    I’m watching. I think the tidal element would be complex. If there was a link it might nevertheless be smudged by erratic power station cooling water outflow.

    I was wondering about the possibility of getting someone to measure the water temps. I did some research yesterday and Thurrock Yacht Club is opposite on the bend. It has a jetty that seems to reach to the low tide mark as you might expect. I wonder if they or anyone else in the area routinely measure the water temps.

    Otherwise, is there such a thing in weather circles as a wireless thermometer that can phone its temp to you every hour (strapped to the yacht club jetty if they were game)?Sounds far fetched, I know, but it won’t be in a couple of years if it’s not yet here. I know there are wifi thermometers with perhaps a 50yd transmission range.

  22. tchannon says:

    Oh. Usually only new or occasional browsers spot old stuff. It can be a bit of a problem hanging on to tech posts. 🙂

    In my opinion getting meaningful data on an estuary is not feasible unless there is a “stonking” effect which is not credible.

    I just deleted a draft reply, far too long and about why I think it is infeasible.

    I suspect it is as feasible but also not worth the effort to monitor close to the met station and even good remote surface sensing, which is getting into investigating the microclimate. Field work on land where the public have access is very tricky.

    Leaving boxes near safety infrastructure is men with guns stuff. Strap a box on a pier leg with phone attached… yeah right, which prison do you fancy?
    Sad state of the world I am afraid.

    However I am not objecting to anyone trying, you never know what will appear.

    There is some news in relation to the Met Office sites, one related to Gravesend-Broadness. Enough I think for a post but first I have some legwork to do, new stations have appeared and at least one probably vanished.