Craibstone DCNN1272 – A detailed comparison between an excellent site and a poor one just 1.25 miles away.

57.186705, -2.214910 No Met Office CIMO Assesment. Installed 1/1/1925 Closed 24/8/2021

Those who have followed my posts from the start may recall my review of Craibstone number 2 from the early days of the Surface Stations Project. I personally felt the new Craibstone 2 site was a good reliable one, well established and maintained by its owner the SRUC – Scotland’s Rural University. My experience of studying weather stations leads me to expect the highest standards from agricultural and associated research institutions and they rarely disappoint. Whilst aviation sites fundamentally exist to assist short term micro scale aviation requirements, agricultural sites record weather in support of climate analysis over wide area representation and long term time scales. Craibstone offers a remarkably good contrast between aviation and agricultural sites given that one of the worst possible aviation sites was under 2km/1.25 miles away – the risible DYCE.

Craibstone was installed in 1925 when the college acquired the large surrounding estate. Digital temperature records are visible from 1931, unfortuanately I have not been able to identify any manual copies from its first 6 operating years. There are no indications of any site moves or significant changes to the surroundings over the intervening years up to closure. The construction of the A90 Aberdeen bypass encroaching with 100 metres of the screen together with a few developments within the estate itself, led to concerns over potential local compromises to the readings. This resulted in the chosen relocation to the new number 2 site.

Starting as a once daily manually observed site with a very good record, the equipment was automated (in common with a large numbers of other sites at the same time) to a PRT measured unit in 2010.

Following strict adherence to the CIMO guide I would judge this site probably just fails Class 1 but only on the basis of changeable ground cover. This is certainly Class 2 at worst. It is worth noting, however, that the Met Office judges the newer number 2 site as Class 4 which again I feel is unduly harsh as I commented on in my earlier report. Met Office assessments regularly surprise me in that the Dyce Airport site is also (incredibly unlikely) rated as Class 4.

Here is another different dated aerial image of Dyce from Google Earth to highlight again all the significant issues. N.B. No this is not a photoshop!

To put the Dyce screen and the original Craibstone site in context here is a reference location map. The Dyce screen is at 65 metres amsl with Craibstone 37 metres higher at 102 metres. The environmental lapse rate (average rate at which temperature decreases with height in the troposphere) is approximately 6.5°C per 1,000 meters (or 0.0065°C per metre). This is the commonly used value accounting for real-world moisture and other factors. 37 metres elevation difference is minimal, however, it should be expected that Craibstone would be fractionally cooler solely based on this cooling at the higher site: 37 × 0.0065 ≈ 0.24°C cooler.

The principle difference between the two sites is the huge development that has occurred at Dyce since the weather station’s inception with temperature readings in 1957. In 1957 Dyce handled 19,761 aircraft movements with 53.683 passengers. By the time of Craibstone closure in 2021 this had increased 4.6 times in terms of aircraft movements to 91,279 with a 57 fold increment in passengers to 3,057,738. This indicates a huge increase in the size and associated power output of the aircraft using the airport. The number of helicopter movements is not exactly known, however, the advent of the North Sea oil rigs is known to have driven an increase of orders of magnitude of larger models.

This is what Dyce Airport looked like in the 1950s.- not even remotely similar to the intensely developed international airport of the 21st century.

It would be reasonable to assume that at the start of the overlapping temperature readings in 1957 (and prior to any significant development at the airport site) that these two sites readings would naturally be very similar especially after allowing for the elevation difference. Any divergence over subsequent time would indicate changes to immediate surroundings affecting readings. iven that Craibstone did not encounter any notable site alteration until closure whilst Dyce changed out of all recognition, this should indicate the effects of the urbanisation of Dyce and its aviation expansion.

N.B. The Met Office main Scottish address is Aberdeen Forecasting Centre: Located at Building 2, Level 1, Aberdeen International Business Park, Dyce Drive, Aberdeen, AB21 0BR i.e. at the airport . The Met Office is very well aware of the details of this area!

Working on the basis that ultimately it is only the actual raw data that can prove/disprove my assertion that there inevitably will be a significant diversion between the two sites I did a basic comparison of the first month of comparative readings in Jan 1957 as opposed to the same month of the last year of such comparative readings in 2021.

My initial basic precis showed Jan 1957:

Craibstone average minimum was 1.67°C whilst Dyce was 1.75 °C a difference of a barely measurable of 0.08 °C .

Average maximum at Craibstone was 7.22 °C versus Dyce @ 7.45 °C – allowing for the lapse rate this noted difference of 0.23 °C indicates almost exact parity.

How different were the January 2020 readings?

Craibstone min mean was 2.11°C whilst Dyce was 2.52 °C. The 1957 negligible difference had grown from 0.08 °C to 0.41 °C

Craibstone max mean was 8.23°C whilst Dyce was @ 8.86 °C – the 1957 differential equivalent to the lapse rate had now widened to 0.63°C.

Obviously such a short comparison period is not conclusive at all but a strong enough indicator to warrant further investigation. As a test example, I have asked Dave Woolcock to run the full periods figures to plot the differences (if any ultimately exist) and this should soon hopefully be available on Dave’s return from holiday. Whether the resultant data proves my hypothesis or not, I will still publish it when available.

Additionally this data should make a good comparison to the homogenised/modified data produced by the Met Office under its Location Specific Long Term Averages for its “gridded squares” hypotheses. After all they seem to conclude Craibstone and Dyce are the same.

Source: https://tallbloke.wordpress.com/2026/08/01/craibstone-dcnn1272-a-detailed-comparison-between-an-excellent-site-and-a-poor-one-just-1-25-miles-away/