Malvern(CET) DCNN4813 versus Great Malvern DCNN4874 – Means of Deception?

MALVERN (Above) 52.11299 -2.30834 CIMO Assessment N/K Temperature data archived from 1/1/1900 to 1/11/2008

GREAT MALVERN 52.12129 -2.30622 CIMO Assessment N/ Temperature data archived from 1/10/1983 to 1/11/2008

There have been two official weather stations in Malvern with one being known as “Great Malvern” after the central and older urban area within Malvern itself. Whilst the older Malvern site ran from 1890 (or possibly earlier), the Great Malvern one ran in conjunction with the older one from 1983 to 2008. The two sites were in very close proximity at just 908 metres/2,982 feet apart and similar altitiudes as seen from this locater map below.

Both sites were manually observed with the older site being viewed once daily at 9:00. The later Great Malvern site was quite peculiarly claimed to be reporting much more frequently as detailed below. I have not been able to either confirm nor ascertain what the timings were of these observations actually were but, as will become evident later in this report, it is quite possible this is an error in the archives. Readings were recorded twice daily in common with multiple reading sites. It is very significant that both sites closed simultaneously suggesting (but not yet confirming) there was a relationship between the two – possibly the same observer.

1996-05-30

2008-11-01

OBSERVING PRACTICE

8 OBS/DAY

The first most important item to note is that the older Malvern station was one of those reporting to the world’s longest running temperature dataset – the Central England Temperature series (CET). Readings from Malvern were used for the CET from 1959 to December 2005.

So it is reasonable to assume that if Malvern was used for such an important and internationally recognised time series it must have been a good quality and reliable site with good historic provenance………wouldn’t it? However, when I noticed such quite alarming notes in the archives as below (along with the headline very concerning image) I started a detailed investigation uncovering some “debateable” issues…..”Unsafe for the Observer”at a weather station site in a middle England town?!

2006-02-01

2006-09-30

MISSING DATA

OBSERVATIONS WERE SUSPENDED DUE TO THE SITE BEING UNSAFE FOR THE OBSERVER. OBS RESUMED ONCE THE SITE WAS MADE SAFE.

What does the Met Office actually claim about the CET? Again I strongly recommend reading their link below in full before progressing.

https://www.metoffice.gov.uk/research/climate/maps-and-data/cet-series

Why is the Met Office stooping to publish more words regarding their perception of “The climate misinformation you may encounter” than the actual subject matter being the CET? A guilty conscience perhaps or trying to justify themselves against criticism?

Those querying the validity of the CET since its upkeep was fully taken over by the Met Office just happen to include the likes of the late Phillip Eden (formerly Principal of the Royal Meteorological Society) who actually established his own competing version as he had lost confidence in the Met office version……….Ah I see, that kind of “Misinformation” – from a scientist of integrity.

The Malvern site had become bluntly junk over time sitting in an area of woodland surrounded by a building site – this image from 1999. If this were a Private Weather Station the Met Office would dismiss its readings out of hand as unreliable.

Latter day readings from this allegedly very long term site would have had very questionable integrity – but what were earlier ones like? Below are the location coordinates shown in the CEDA archives.

And here are the location coordinates shown in the digital archive record for every single year of the record going back to 1900 which are the same as above so obviously that is where it was i.e 52.113 -2.308

Again I will stress the point that I make many times (originally regarding Southampton) that few, if any, modern students would feel the need to question the location of this site – the archives “conclusively” prove that ……only typically for the Met Office this is completely WRONG a.k.a.Disinformation. The manuscript records show completely different coordinates of 52° 08′ 2° 18′.

Confirmation of the different location is handily supplied by “Objects Used for Estimates of Visibility” which gives distances to fixed points.

In similar vein to that I used in my Bala report, I was eventually able to establish the long term location of Malvern weather station was, in fact, as below. The measurements from Edith Walk, Como Road and Church Street place the original screen somewhere under the foundations of the modern day Waitrose site.

This makes the distance between the original location and the final one just under 1.4km. as below.

I have not been able to verify exactly when the relocation took place but what I positively can verify is that in being so far apart it is highly probable that the two sites would have displayed different climatology and certainly should not be considered suitable to derive a continuous unbroken data set. Unfortunately that is exactly what the Met Office data implies by not openly showing that two distinctly different datasets have been seamlessly bonded together like so very many others e.g. Braemar and Stornoway. Thus the obvious question is can such a difference in location make a noticeable difference in recorded temperatures and is siting as critical a factor as I often suggest? Enter Great Malvern supplying comparative data from just 908 metres separation.

61 Charles Way Malvern WR14 2NB is the exact location of that “8 Obs/Day” official Met Office weather station DCNN4874 and this is exactly what it looked like. This image is taken from clearly visible online archived house sales information site that I stumbled across by fluke but have confirmed was the screen location.

Stop and reconsider this point – this was an official Met Office weather station and just like Ballywatticock , Carlton in Cleveland, Astwood Bank and probably over one hundred others (no exaggeration) is in a totally inappropriate back garden site with far too many compromises to be taken seriously. Could this screen possibly get any nearer to that hedge, suffer from any more sun shading from such tall trees, be in more of a wind-sheltered spot? But this is what the Met Office falsely claims and expects its tax paying funders to believe. This statement below is demonstrably untrue a.k.a. “Disinformation.”

“Placement and standards

Weather stations are found throughout the UK, typically spaced about 40 km apart, allowing for effective tracking of weather systems such as low pressure and frontal movements. Their locations are carefully chosen to ensure that measurements are representative of the wider area and not unduly influenced by local factors. For instance, stations are often found in open spaces like airports, but equipment is placed at internationally agreed distances from runways to avoid interference.

Consistency is vital for both short-term forecasts and long-term climate records. To maintain this, weather stations must adhere to strict criteria set by international meteorological organisations. These standards cover aspects such as grass cover and the need for clear space, ensuring that readings are not affected by buildings, trees, or other non-meteorological influences.“

To confirm the CET details firstly a map of stations used over time detailed in this report entitled “An Update to the Central England Temperature Series—HadCET v2.1”

https://rmets.onlinelibrary.wiley.com/doi/10.1002/gdj3.284

And here is an extract detailing the time scales used.

Crucial to bear in mind is that, in the event of data unavailability from any site, readings from nearby sites are/were often substituted as noted:

“In addition, when assessing output from Met Office weather stations, data is compared to nearby observations, as well as what’s feasible in the synoptic meteorological set-up, to check for any spurious data which needs to be checked, marked as suspect, or not be used in Met Office products and services. Indeed, this does happen on occasion, where an observation has been found to be unsatisfactory, and therefore wasn’t used as an official observation.“

So now for some numbers to compare just how well correlated these two Malvern/Great Malvern weather stations, so close together for such a long period, actually were to either confirm or refute Met Office practises of data correlation and substitution.

Yet again I am deeply indebted to the ongoing and pains taking work of Dave Woolcock who has put in may unpaid hours to download the entire CEDA MIDAS digital temperature record for every single weather station current and past – a major task. From this data Dave is devising all manner of data comparisons, giving full and detailed methodology which I will also include.

Firstly, Dave offered differences in minimums (t/Min) of Great Malvern over Malvern. Should there really be much difference? I have added all Dave’s comments for clarity. Clicking on graphs opens in new window for better view.

“when matching the two series I originally did a left outer join instead of an inner join … inner join is where it only produces an output if BOTH entries exist)

Anyroad up:  here is tMin“

“looks like an annual pattern, so I picked two random years – 1991 and 2003 – to take a closer look:“

“Some outliers in the winter months but not really in the summer months. Is this the same effect as we noted before – tMins sometimes scoring twice in Winter? {Ed note: see references below}

What are the overall stats?“

On to tMax

hmm – looks different to tmin – less peaky, wider spread

and the two years:

yep looks like similar random noise the same throughout the year

what about the stats?

average (mean) difference is half of bugger all

but the standard deviation is more than double tMin at 2.5 degC

and the range of differences is even huger. 

…. and what do we conclude from all that?

on average the stations produce pretty identical results?

but individual daily values can vary enormously. “

A massive “thank you” to Dave for all this work. Both he and I are now trying to further analyse any potential causes for such large discrepancies from a purely statistical representation point of view. There is the issue of the Great Malvern site reporting more frequently during the day and (as Dave mentioned) this can cause the winter minimums to no longer effectively carry over from one day to the next under certain weather conditions. We have previously demonstrated this effect with the Cambridge Convolution Part One and Part Two. The whole “Time of Obs” bias was raised with the issue of why winter nightimes are getting warmer and the step change effect of changing both reporting times and instrumentation type (from LIGT to PRT) in association. This automation/multiple reading times effectively removes the previous once daily reading problems but effectively generates the appearance of warming where none may exist.

If anyone is able to download the MIDAS stats and offer alternative representations I would certainly welcome their interpretation for further discussion.

For my part, at this stage, I feel that the huge discrepancies between the two sites are largely caused by differences in siting standards with both being subject to distorting influences. This demonstrates both how critically important good site location is to the accuracy of readings and thus whether different locations can genuinely be well correlated and absent readings be derived from elsewhere. Despite huge variations in temperature range, long term derived crude meteorological averaging can actually mask these differentials but clearly data substitution is not possible for missed readings if so hugely variable.

This post will be continued with an assessment of another CET former site (Ringway Airport) and I make no apology for the rather long length and details involved as they are necessary to evaluate the problems in full. However, by way of a preview, I offer this image of latter day Ringway which is actually hugely superior to its previous location (also during its CET days) which was on the roof of a building!

Source: https://tallbloke.wordpress.com/2026/05/26/malverncet-dcnn4813-versus-great-malvern-dcnn4874-means-of-deception/