Northolt, Heathrow, Temperature Records and Weather Station Siting

[Mapantz]

9 minutes ago, Aleman said:

But what defines a frost hollow, though? That usually means a dip in the topography or something sheltered otherwise from wind. That seems to be just what Benson is not. It looks well sited in an airport.

https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1256/wea.119.05

Quote

If you mention Benson to a British weather observer, forecaster or climatologist, there is an immediate recognition of a noteworthy site that experiences low radiation-night temperatures. The locality has one of the largest maximum diurnal temperature ranges at an English lowland site not in an incised valley.

 

Edited July 8, 2019 by Mapantz

[Frost HoIIow]

3 hours ago, cheeky_monkey said:

be interesting to see the difference in temps between Heathrow and Northolt?..i would say if anything Heathrow would be effected more by urban environment due to its size, air traffic and esp being right next to the M25

True. There's an awful lot of concrete on the site of Heathrow (and nearby) plus you have heat generated by cars visiting & leaving the airport 24/7 not to mention aircraft. There's a program on ITV that has the goings-on with the airport & it says it's almost like a small city in itself.

Edited July 8, 2019 by Frost HoIIow

[Scorcher]

Anyone playing down the temps recorded at Heathrow and Northolt- why don't they hold the all time records then?

Often in the hottest spells Heathrow is matched or bettered by stations in Surrey- Wisley springs to mind. Maxima at Kew are often on a par with Heathrow as well.

A lot of coldies seem to bang on about urbanisation causing readings to be too high- even if this is the case, what's the problem with it?

The UK is highly urbanised anyway so if the readings are affected by the urban area slightly I see no problem with this- it reflects the true experience of 90 odd percent of the population who live and work in towns and cities.

I see the OP mentioned the Manchester Airport site too- I actually think having Manchester Airport and Rostherne as the 2 stations for Manchester is not a reflection on the city centre for example- it's almost invariably cooler at the airport which is semi-rural. Rostherne is even cooler, especially at night, and is a bit of a frost hollow.

If you put a station a couple of miles further north in Manchester, further into the urban area you would see higher readings which for me are more reflective of the experience of most people in Greater Manchester.

[mb018538]

42 minutes ago, Scorcher said:

Anyone playing down the temps recorded at Heathrow and Northolt- why don't they hold the all time records then?

Often in the hottest spells Heathrow is matched or bettered by stations in Surrey- Wisley springs to mind. Maxima at Kew are often on a par with Heathrow as well.

A lot of coldies seem to bang on about urbanisation causing readings to be too high- even if this is the case, what's the problem with it?

The UK is highly urbanised anyway so if the readings are affected by the urban area slightly I see no problem with this- it reflects the true experience of 90 odd percent of the population who live and work in towns and cities.

I see the OP mentioned the Manchester Airport site too- I actually think having Manchester Airport and Rostherne as the 2 stations for Manchester is not a reflection on the city centre for example- it's almost invariably cooler at the airport which is semi-rural. Rostherne is even cooler, especially at night, and is a bit of a frost hollow.

If you put a station a couple of miles further north in Manchester, further into the urban area you would see higher readings which for me are more reflective of the experience of most people in Greater Manchester.

I agree with this - if we just binned all the stations now because they were ‘too urbanised’ and relocated them all out into meadows and fields where most of the population don’t live, what would that really achieve? Most people live in cities, so the stations in more built up areas are a good reflection of life.

As already mentioned, if Heathrow was dodgy, why doesn’t it hold the all time record? Surely on that day in 2003 if anywhere was going to be hottest it’d be an airport, right? 

Edited July 8, 2019 by mb018538

[Aleman]

Yet again, I did not start this discussion because of urban heat island effects. I presume that has been done to death.

In researching how to site my own new and modest station (so not being an expert), I was sidelined onto a write-up of siting USCRU's new network of clean Class 1 stations. When I had a look at Northholt just after that, I was shocked. Others said but it ties in with Heathrow but I looked at that and was shocked at that, too. Too much stuff within 30m at both! (Nothing to do with runways or planes for either.)

I'm asking why we don;t restrict stations to at least Class 2 and stop using the others? I'm quite happy for there to be stations at Heathrow and Northolt Airports that try to measure temperatures that carry urban influences that make them warmer than rural ones. Cities and airports should be part of the averages as much as any odd cold spot and station there advise locals of conditions. I'm not happy that these two stations might be measuring something other than what people think due to heat influences within the normal exclusion zones. (Slight sarcasm to following:) Benson is clear and looks like it is measuring a roughly typical temperature for that area. Heathrow looks to be measuring the temperature of the perimeter road and Carillion building (and vents) while elimating any breeze that makes it the same as surrounding area. Northolt looks like it measures the temperature of A40 traffic jams and reflections off the car park. The latter two don't look representative of the general temperature at their airports due to encroachment of stuff that should not be there.

A class 1 station should have nothing within 100m and Class 2 within 30m yet Heathrow Airport has panelled blast fences within 10m and a road, drive, some kind of boxed and vented equipment,  and the Carillion building within 30m that has an air conditioning outlet, and which recently had it roof repainted from white to pitch black. Northolt has a only a path inside 10m but a car park, a blast fence, an (electrical) outbuilding, a 6 lane highway and a line of 12-15m tall trees blocking wind and summer morning sun within 30m. These do not seem to comply with Met Office guidelines. I think the Met Office used to aim for 100 feet clearance, which is now WMO Class 2, but they seem to have dropped that and are now (oddly) vague on distances. The things I've described put Heathrow and Northolt into WMO Class 3 or more likely Class 4 territory, which I think makes them probably unfit for what they are being used for - but I'm just a layman/novice. It would be bad enough that these are so cluttered now with things that could affect readings but the clutter has increased over time which generates inconsistencies in historical comparison - and not for London growing or airports getting busier or any other growing urban heat island effect - but just for so much undesirable stuff being accumulated within the 30m Class 2 that it could be contaminating the readings more now than before. As a layman, I was shocked. Should I have been or not?

Now, I was not trying to start a riot or damn the Met Office. I was not trying to start a debate on heat islands effects/urbanisation which is a separate topic. I'm just wondering whether station sites that are so contaminated within 30m are actually representating their wider surroundings, and wondering what degree of contamination there is amongst other stations. I accept it's not easy. I've not decided where to site mine yet. I don't have 30m clear in my garden!

 

Edited July 8, 2019 by Aleman

[DaveL]

@AlemanIt's not easy to do meaningful comparisons without having all the data from Met Office climatalogical stations in the area but my thoughts are first, that the effects of urban heat island are well known and can be adjusted for. Site issues do pose more of a problem, I agree, and I think this is more likely to be seen on extremes rather than averages - that is, in the kind of unusual conditions that tend to produce extremes. Anomalous averages from individual stations may be masked by the bulk of non-unusual days e.g. cloudy, rainy and breezy or windy days.

With this in mind, I had a look at the Torro list of daily extremes for maximum temperatures here: http://www.torro.org.uk/hightempsyear.php  and counted the number of appearances since (finger in the air) 1960,   for several stations in the west and SW of London. That's roughly when Heathrow airport development literally took off, so 60 years of comparisons. Here are the results:

Heathrow - 13 days
Wisley - 8 days
Kew Gardens - 5 days
St James's Park - 3 days
Northolt - 2 days

Those are the total number of days for each station when they recorded the sole or joint highest UK daily maximum temperature over the last 60 years. I would expect to see a more even spread than that, given the same general area they are all located in.

Heathrow appears too many times on the list for my liking,  possibly Wisley too (enclosure is next to an orchard) but Northolt, surprisingly, looks OK. It may, however, simply be that nearby Heathrow is usually just a bit higher so ideally, we need more than just the highest max to look at. But to my eyes, site differences are likely to have influenced the list more than urban heating. And *possibly* -  just possibly, the sensitivity and reaction time of thermometer/screen combinations.

Edited July 9, 2019 by DaveL

[Methuselah]

Without any actual data, talking about what 'might be the case', seems like a red herring. First, find an anomaly and then work to explain it? 

And, anywho...somewhere has to be the warmest...?

[mb018538]

https://www.carbonbrief.org/met-office-wind-data-dispels-doubt-about-cause-of-heathrow-high-temperatures

This article is quite good - focusing on the 1st July 2015, when Heathrow recorded 36.7c (hottest July day on record) and about how people called foul play. A little bit of research into the data showed a real spike in solar radiation that wasn't felt at Kew due to cloud cover a few miles away. It can be as simple as that.....hence why without all of the data you can make erroneous judgements.

[Methuselah]

36 minutes ago, mb018538 said:

https://www.carbonbrief.org/met-office-wind-data-dispels-doubt-about-cause-of-heathrow-high-temperatures

This article is quite good - focusing on the 1st July 2015, when Heathrow recorded 36.7c (hottest July day on record) and about how people called foul play. A little bit of research into the data showed a real spike in solar radiation that wasn't felt at Kew due to cloud cover a few miles away. It can be as simple as that.....hence why without all of the data you can make erroneous judgements.

Many thanks for finding that, mb!

But, as the latest on Climategate implies, it's all part of a concerted misinformation-scheme by so-called climate 'sceptics'..it's been ongoing for a long while, and I doubt it'll cease any time soon; the prospect of receiving humongous wads of Big Oil money is simply too tempting for some?

But, hopefully, this particular round of nonsense has been nipped-in-the-bud?

https://www.bbc.co.uk/news/av/science-environment-48925015/climategate-10-years-on-what-s-changed

[Aleman]

On 09/07/2019 at 01:06, DaveL said:

@AlemanIt's not easy to do meaningful comparisons without having all the data from Met Office climatalogical stations in the area but my thoughts are first, that the effects of urban heat island are well known and can be adjusted for. Site issues do pose more of a problem, I agree, and I think this is more likely to be seen on extremes rather than averages - that is, in the kind of unusual conditions that tend to produce extremes. Anomalous averages from individual stations may be masked by the bulk of non-unusual days e.g. cloudy, rainy and breezy or windy days.

With this in mind, I had a look at the Torro list of daily extremes for maximum temperatures here: http://www.torro.org.uk/hightempsyear.php  and counted the number of appearances since (finger in the air) 1960,   for several stations in the west and SW of London. That's roughly when Heathrow airport development literally took off, so 60 years of comparisons. Here are the results:

Heathrow - 13 days
Wisley - 8 days
Kew Gardens - 5 days
St James's Park - 3 days
Northolt - 2 days

Those are the total number of days for each station when they recorded the sole or joint highest UK daily maximum temperature over the last 60 years. I would expect to see a more even spread than that, given the same general area they are all located in.

Heathrow appears too many times on the list for my liking,  possibly Wisley too (enclosure is next to an orchard) but Northolt, surprisingly, looks OK. It may, however, simply be that nearby Heathrow is usually just a bit higher so ideally, we need more than just the highest max to look at. But to my eyes, site differences are likely to have influenced the list more than urban heating. And *possibly* -  just possibly, the sensitivity and reaction time of thermometer/screen combinations.

I appreciate the effort put in there but I don't see how it helps. If Heathrow is a hot spot on average, it will be in the hottest place list frequently. Similarly, a cold spot would appear in the daily coldest list frequently. Although more hot stations might have significant encroachment and more cold ones might have none, the number of appearances in the list of hottest and coldest doesn't tell us anything. However, if a nearby station never appeared - you might then investigate urban heat effects and encroachent for any unusual differences between them. I think you could study historic data to split out a warming trend. That happens now but how due you split out the heating from encroachment and the heating from urban warming?

[DaveL]

8 hours ago, Aleman said:

I appreciate the effort put in there but I don't see how it helps. If Heathrow is a hot spot on average, it will be in the hottest place list frequently. Similarly, a cold spot would appear in the daily coldest list frequently. Although more hot stations might have significant encroachment and more cold ones might have none, the number of appearances in the list of hottest and coldest doesn't tell us anything. However, if a nearby station never appeared - you might then investigate urban heat effects and encroachent for any unusual differences between them. I think you could study historic data to split out a warming trend. That happens now but how due you split out the heating from encroachment and the heating from urban warming?

I don't think I explained my thinking very well.  The thing is, Heathrow, Northolt and St. James's Park are all well established stations within 8 miles of Kew Observatory, and all 4 are in the central or western London urban area, so you'd expect to see a similar level of UHI effect in all 4.  They are also in the warmest part of the UK (except IIRC in late autumn and winter). Wisley is probably too far out to make a good comparison, I accept that.

That leads me to expect each of these 4 stations to have the highest maximum in the UK for similar numbers of days, over a long time scale

I know it doesn't support the Northolt theory very well but that may simply be because when Northolt would be hottest, nearby Heathrow (5 miles) is 'worse' and usually pips it at the post.  Admittedly it's based on a very limited dataset but it seems to me that Heathrow is the outlier here, and by some margin. I suspect that differences in site quality - with Heathrow being the worst - most likely explains this.

More data is needed, certainly, but at least this is a start, and I know from personal experience that when the Met Office check for anomalous station readings and averages, they make comparisons with the nearest surrounding stations. I think this issue that you first brought up definitely needs looking at in more detail.

Edited July 11, 2019 by DaveL

[Aleman]

That makes a certain sense. Thanks. I probably need to check the stations in Birmingham, maybe. Probably plenty of UHI and encroaching clutter there, too.

[Midlands Ice Age]

10 hours ago, Aleman said:

That makes a certain sense. Thanks. I probably need to check the stations in Birmingham, maybe. Probably plenty of UHI and encroaching clutter there, too.

Regarding B'ham...

I live on the outskirts of Bham, near a piece of 'Green' land, on the South East side of the conurbia. Bham Airport and Coleshill are my  nearest official stations.

There are several 'unofficial' temp recordings in this  area, 

In summer my daytime maxes are usually about right. (comparing with the more central areas); night time I do fall much lower.

In winter, I record much colder both night and day  (unless the wind is in the North or North west - when I get shielded from the colder weather by the 20 miles of suburbia to the West and North).  I find I am substantially colder when the wind is in the Easterly quadrant. I have also observed these affects many times with car temperature watching.

My impression is that wind direction can have a big affect on the UHI of a particular site.

To this point a paper was published about 6 months ago  that stated  that the UHI affect downwind had been detected up to 800 yards, and within 100 yards there was still a detectable amount of heating. Within 50 yards up to 1degree higher recordings were discovered at fairly regular interludes by the researchers.

My message is   - look at wind directions associated with any 'errant' temperatures. 

Edited July 12, 2019 by Midlands Ice Age

[Aleman]

I noticed St James has been in the top 3 a few times of late so I looked at that compared to siting recommendations.

Wind: at least 10 times the height of obstruction. FAIL

Temperature: at least 4 times obstruction's height, at least 30m from paved areas. FAIL, FAIL

Solar: objects to be below 10 degrees from horizon. FAIL

Precipitation: at least 4 times obstruction's height. FAIL

Soil Temp: 1m x 1m site typical of area and flat for 10m radius. FAIL, PASS

Also, not sure 1.5m from the corner of moderately thick black railing helps - but some have mesh,  blast fence or white pickets. Can't they even get consistency in the fencing when it might be only a meter or two away?

https://www.meteopool.org/en/weather-station-london-st-james-park-met-office-wmo-03770-id105

 

I used these guidelines since the Met Office seemed to have gone vague on their own.

https://www.campbellsci.co.uk/weather-station-siting

 

 

 

Edited July 12, 2019 by Aleman

[DaveL]

@Aleman Yes,  and that's one of the reasons why I expected St. James's Park to have the record UK temperature more often, compared to Heathrow. There is, however, a lot of greenery around the immediate location (apart from the pathway shown and a few buildings behind the trees), it's over 100 metres to most of the urbanization and the whole of the park lies to its south west and south. Everything is closer on the north side at Heathrow and although there's a grassy gap on  the south side there, it's much less than the south-SW side of the St James's enclosure. Kew Gardens though is likely the best site you could get in this part of London.

Regarding more data, does anyone know of an archive of the UK or SE regional  daily maximums, as reported every day by the Met Office website?

Edited July 12, 2019 by DaveL

[hillbilly]

It may sound crazy but is it possible that other factors could have any impact on temperatures in very hot weather.If you look an a map Heathrow is clearly half surrounded by resevoirs to the south and west.Would that be possible in a southerly breeze to cause differing winds of different temperatures with hot evaporated water that could cause a type of Fohn effect?

[Aleman]

On 12/07/2019 at 19:20, hillbilly said:

It may sound crazy but is it possible that other factors could have any impact on temperatures in very hot weather.If you look an a map Heathrow is clearly half surrounded by resevoirs to the south and west.Would that be possible in a southerly breeze to cause differing winds of different temperatures with hot evaporated water that could cause a type of Fohn effect?

(Since nobody else has offered:) Not as I understand it.  The Foehn effect is moisture condensing out as it rises up a hill releasing heat which reduces the cooling due to adiabatic (pressure) change that one might see on drier air. The air at the top of the hill is then less cooled than dry air would have been rising up the same hill. When it goes down the other side, it warms through the normal adiabatic process again. So less cooling going up and same warming coming down equals a warmer temperature for moist air than dry on the leeward side.

Without a hill there is no Foehn effect.  Lakes around the airport would moderate the extremes by making air more moist but airports remain dry places due to lack of transpiring vegetation with deep roots. Grass does release moisure but in limited quantities so soon dries out. So, I think lakes around heathrow would moderate temperatures marginally through higher humidity but are mostly likely overcome by other factors.

I'm not an expert so anybody that thinks otherwise, please say so.

 

Edited July 16, 2019 by Aleman

[cheeky_monkey]

1 hour ago, Aleman said:

(Since nobody else has offered:) Not as I understand it.  The Foehn effect is moisture condensing out as it rises up a hill releasing heat which reduces the cooling due to adiabatic (pressure) change that one might see on drier air. The air at the top of the hill is then less cooled than dry air would have been rising up the same hill. When it goes down the other side, it warms through the normal adiabatic process again. So less cooling going up and same warming coming down equals a warmer temperature for moist air than dry on the leeward side.

Without a hill there is no Foehn effect.  Lakes around the airport would moderate the extremes by making air more moist but airports remain dry places due to lack of transpiring vegetation with deep roots. Grass does release moisure but in limited quantities so soon dries out. So, I think lakes around heathrow would moderate temperatures marginally through higher humidity but are mostly likely overcome by other factors.

I'm not an expert so anybody that thinks otherwise, please say so.

 

Jet engines emit quite a lot of moisture?

[Derecho]

On 12/07/2019 at 14:34, DaveL said:

@Aleman Yes,  and that's one of the reasons why I expected St. James's Park to have the record UK temperature more often, compared to Heathrow. There is, however, a lot of greenery around the immediate location (apart from the pathway shown and a few buildings behind the trees), it's over 100 metres to most of the urbanization and the whole of the park lies to its south west and south. Everything is closer on the north side at Heathrow and although there's a grassy gap on  the south side there, it's much less than the south-SW side of the St James's enclosure. Kew Gardens though is likely the best site you could get in this part of London.

Regarding more data, does anyone know of an archive of the UK or SE regional  daily maximums, as reported every day by the Met Office website?

It's view on google maps would confirm it has a lot of Greenery surrounding it. Here is a pic of the station in 2017 and its opposite side:

  

With regards to the siting there are no large (word omitted from Alemans post) paved areas. Judging by the google maps measurements I'd go for the follows as a rough estimate: 

It's on the NE side of the park so away from the roads with a building 30m away with a small car park. The statement from the EPA says.

Temperature and humidity sensors should be located at a distance of at least four times
the height of any nearby obstruction and at least 30 m from large paved areas [2], [15]. Other
situations to avoid include: large industrial heat sources, rooftops, steep slopes, sheltered
hollows, high vegetation, shaded areas, swamps, areas where frequent snow drifts occur, low
places that hold standing water after rains, and the vicinity of air exhausts (e.g., from a tunnel or
subway) [2, 9].

So what is classed as a nearby obstruction? The nearest tree is roughly 25m away so it isn't 'nearby'. It may have implications for wind but that is it.

Wind: at least 10 times the height of obstruction. Perhaps a fail?...

Temperature: at least 4 times nearby obstruction's height, at least 30m from large paved areas. PASS, PASS

Solar: objects to be below 10 degrees from horizon. PASS

Precipitation: at least 4 times nearby obstruction's height. PASS, nothing to get in the way of rainfall?

Soil Temp: 1m x 1m site typical of area and flat for 10m radius. PASS, PASS surely? the fence is well in excess of 1m*1m

 

With regards to the Foehn effect you often get a large amount of moisture (wayyyy larger then that from a plane engine). The moisture laden air heading up the mountain cools at a slower rate because it is saturated. The air dries out at the top and then the temperature increases at the faster dry adiabatic lapse rate as it heads down the mountain, as summed up by this diagram below (I think):

Edited July 16, 2019 by Quicksilver1989

[DaveL]

Looking at the various site descriptions and location links for west and central London sites, I think Kew Gardens is going to be the most 'standard' site. It's only 7 miles due east of Heathrow,   so I have obtained from the CEDA website the daily maximum and minimum temperatures for Kew Gardens and Heathrow, going back many years and have compared them. I cannot obtain monthly averages for  Kew so I'll have to work them all out, therefore my initial comparisons are based on daily maxima, which are downloadable in spreadsheet format.

I  previously suspected that the biggest differences in daily maximum temperatures between the two sites are likely to have been on the warmest days of the year, so I compared the differences in maximum temperature (Heathrow minus Kew) in a table sorted by maximum temperature, highest to lowest. I have two tables, the first based on the Heathrow maxima,  and the second based on the Kew maxima  (there are differences). Here is a short summary of my initial results - based only on 2017, which is the most recent year available. I looked at nothing other than the daily maxima.

Heathrow minus Kew based on the Heathrow highest maxima

For the 10 warmest days of 2017: Average difference +0.49°C; Heathrow warmer on 9 of 10 days (Kew 1).
For the 20 warmest days of 2017: Average difference +0.36°C; Heathrow warmer on 15 of 20 days (Kew 5).
For the 40 warmest days of 2017: Average difference +0.35°C; Heathrow warmer on 29 of 40 days (Kew 10, 1 equal).
For the 60 warmest days of 2017: Average difference +0.31°C; Heathrow warmer on 43 of 60 days (Kew 15, 2 equal).
For the 90 warmest days of 2017: Average difference +0.30°C; Heathrow warmer on 65 of 90 days (Kew 18, 7 equal).
For the 180 warmest days of 2017: Average difference +0.23°C; Heathrow warmer on 115 of 180 days (Kew 44, 21 equal).
For all of 2017, Average difference +0.14°C, Heathrow was warmer.

Heathrow minus Kew based on the Kew highest maxima

As expected,  the differences skew towards Kew except for the entire year - but Heathrow is still well ahead e.g. for the warmest:

10 days +0.42°C; H 8, K 2.
20 days +0.30°C; H 14, K 6.
40 days +0.24°C; H 27, K12.
60 days +0.29°C; H 41, K16.
180 days +0.21°C; H68; K21.
All 2017 +0.14°C, Heathrow warmer.

I hope the abbreviations make sense, my fingers are overworked!

So Heathrow maxima on at least the warmest 180 days of 2017,  were mostly higher than the better sited Kew Gardens. The exception I noticed is on the coldest 30 or so days of the year, when Kew was a bit warmer. There's enough there for me to be interested in digging deeper and for other years, with monthly means as well, but it may take a while.  However, if the above results are supported, there is the thorny question of... why? We have, of course, previously discussed that point.

All comments welcome, this is a first attempt and I am not a statistician (but I can use Excel).

Edited July 17, 2019 by DaveL

[jonboy]

1 hour ago, DaveL said:

Looking at the various site descriptions and location links for west and central London sites, I think Kew Gardens is going to be the most 'standard' site. It's only 7 miles due east of Heathrow,   so I have obtained from the CEDA website the daily maximum and minimum temperatures for Kew Gardens and Heathrow, going back many years and have compared them. I cannot obtain monthly averages for  Kew so I'll have to work them all out, therefore my initial comparisons are based on daily maxima, which are downloadable in spreadsheet format.

I  previously suspected that the biggest differences in daily maximum temperatures between the two sites are likely to have been on the warmest days of the year, so I compared the differences in maximum temperature (Heathrow minus Kew) in a table sorted by maximum temperature, highest to lowest. I have two tables, the first based on the Heathrow maxima,  and the second based on the Kew maxima  (there are differences). Here is a short summary of my initial results - based only on 2017, which is the most recent year available. I looked at nothing other than the daily maxima.

Heathrow minus Kew based on the Heathrow highest maxima

For the 10 warmest days of 2017: Average difference +0.49°C; Heathrow warmer on 9 of 10 days (Kew 1).
For the 20 warmest days of 2017: Average difference +0.36°C; Heathrow warmer on 15 of 20 days (Kew 5).
For the 40 warmest days of 2017: Average difference +0.35°C; Heathrow warmer on 29 of 40 days (Kew 10, 1 equal).
For the 60 warmest days of 2017: Average difference +0.31°C; Heathrow warmer on 43 of 60 days (Kew 15, 2 equal).
For the 90 warmest days of 2017: Average difference +0.30°C; Heathrow warmer on 65 of 90 days (Kew 18, 7 equal).
For the 180 warmest days of 2017: Average difference +0.23°C; Heathrow warmer on 115 of 180 days (Kew 44, 21 equal).
For all of 2017, Average difference +0.14°C, Heathrow was warmer.

Heathrow minus Kew based on the Kew highest maxima

As expected,  the differences skew towards Kew except for the entire year - but Heathrow is still well ahead e.g. for the warmest:

10 days +0.42°C; H 8, K 2.
20 days +0.30°C; H 14, K 6.
40 days +0.24°C; H 27, K12.
60 days +0.29°C; H 41, K16.
180 days +0.21°C; H68; K21.
All 2017 +0.14°C, Heathrow warmer.

I hope the abbreviations make sense, my fingers are overworked!

So Heathrow maxima on at least the warmest 180 days of 2017,  were mostly higher than the better sited Kew Gardens. The exception I noticed is on the coldest 30 or so days of the year, when Kew was a bit warmer. There's enough there for me to be interested in digging deeper and for other years, with monthly means as well, but it may take a while.  However, if the above results are supported, there is the thorny question of... why? We have, of course, previously discussed that point.

All comments welcome, this is a first attempt and I am not a statistician (but I can use Excel).

It would be an interesting exercise to do the same comparison for night time temperatures

[Aleman]

19 hours ago, Quicksilver1989 said:

It's view on google maps would confirm it has a lot of Greenery surrounding it. Here is a pic of the station in 2017 and its opposite side:

  

With regards to the siting there are no large (word omitted from Alemans post) paved areas. Judging by the google maps measurements I'd go for the follows as a rough estimate: 

It's on the NE side of the park so away from the roads with a building 30m away with a small car park. The statement from the EPA says.

Temperature and humidity sensors should be located at a distance of at least four times
the height of any nearby obstruction and at least 30 m from large paved areas [2], [15]. Other
situations to avoid include: large industrial heat sources, rooftops, steep slopes, sheltered
hollows, high vegetation, shaded areas, swamps, areas where frequent snow drifts occur, low
places that hold standing water after rains, and the vicinity of air exhausts (e.g., from a tunnel or
subway) [2, 9].

So what is classed as a nearby obstruction? The nearest tree is roughly 25m away so it isn't 'nearby'. It may have implications for wind but that is it.

Wind: at least 10 times the height of obstruction. Perhaps a fail?...

Temperature: at least 4 times nearby obstruction's height, at least 30m from large paved areas. PASS, PASS

Solar: objects to be below 10 degrees from horizon. PASS

Precipitation: at least 4 times nearby obstruction's height. PASS, nothing to get in the way of rainfall?

Soil Temp: 1m x 1m site typical of area and flat for 10m radius. PASS, PASS surely? the fence is well in excess of 1m*1m

 

With regards to the Foehn effect you often get a large amount of moisture (wayyyy larger then that from a plane engine). The moisture laden air heading up the mountain cools at a slower rate because it is saturated. The air dries out at the top and then the temperature increases at the faster dry adiabatic lapse rate as it heads down the mountain, as summed up by this diagram below (I think):

Good contribution. Thanks.  I like the pythagoras trick.  But I don't agree with your conclusions at all. I didn't think I needed to elaborate but here goes:

Wind: at least 10 times the height of obstruction. FAIL. Trees are both 15m+ tall and the trunks are no more than 25m away - and the canopy brings them even closer.

Temperature: at least 4 times obstruction's height, at least 30m from large paved areas. Trees are 15m+ tall and no more than 25m away. FAIL. If a large paved area encroached to 25m , it would be a fail. A 4m path plus added 1m bordering the station no more than 2m from the temperature sensor is clearly a FAIL. It's a large paved area - not a stepping stone! Other organisations' station parameters indicate NO heating elements inside 30m (as per old Met Office guidelines) to get a Class 2 and inside 10m to get a Class 3. This site would be a class 4 for paving, railings and lamp post. That's not good enough for what it is being used for.

Solar: objects to be below 10 degrees from horizon. FAIL Tree to east about 40 degrees plus lampost to south is over 10. FAIL

Precipitation: at least 4 times obstruction's height. Still the trees. FAIL.

Soil Temp: 1m x 1m site typical of area and flat for 10m radius. FAIL, PASS. It can' be typical of the area if the area if there is so much paving near. For instance, rain run off could mean extra water added to soil on that side of the site.

 

As an added bonus, there are three rectangular manholes about 20m or so to the southwest. So what is going on underground around the station?

Edited July 17, 2019 by Aleman

[Aleman]

4 hours ago, DaveL said:

Looking at the various site descriptions and location links for west and central London sites, I think Kew Gardens is going to be the most 'standard' site. It's only 7 miles due east of Heathrow,   so I have obtained from the CEDA website the daily maximum and minimum temperatures for Kew Gardens and Heathrow, going back many years and have compared them. I cannot obtain monthly averages for  Kew so I'll have to work them all out, therefore my initial comparisons are based on daily maxima, which are downloadable in spreadsheet format.

I  previously suspected that the biggest differences in daily maximum temperatures between the two sites are likely to have been on the warmest days of the year, so I compared the differences in maximum temperature (Heathrow minus Kew) in a table sorted by maximum temperature, highest to lowest. I have two tables, the first based on the Heathrow maxima,  and the second based on the Kew maxima  (there are differences). Here is a short summary of my initial results - based only on 2017, which is the most recent year available. I looked at nothing other than the daily maxima.

Heathrow minus Kew based on the Heathrow highest maxima

For the 10 warmest days of 2017: Average difference +0.49°C; Heathrow warmer on 9 of 10 days (Kew 1).
For the 20 warmest days of 2017: Average difference +0.36°C; Heathrow warmer on 15 of 20 days (Kew 5).
For the 40 warmest days of 2017: Average difference +0.35°C; Heathrow warmer on 29 of 40 days (Kew 10, 1 equal).
For the 60 warmest days of 2017: Average difference +0.31°C; Heathrow warmer on 43 of 60 days (Kew 15, 2 equal).
For the 90 warmest days of 2017: Average difference +0.30°C; Heathrow warmer on 65 of 90 days (Kew 18, 7 equal).
For the 180 warmest days of 2017: Average difference +0.23°C; Heathrow warmer on 115 of 180 days (Kew 44, 21 equal).
For all of 2017, Average difference +0.14°C, Heathrow was warmer.

Heathrow minus Kew based on the Kew highest maxima

As expected,  the differences skew towards Kew except for the entire year - but Heathrow is still well ahead e.g. for the warmest:

10 days +0.42°C; H 8, K 2.
20 days +0.30°C; H 14, K 6.
40 days +0.24°C; H 27, K12.
60 days +0.29°C; H 41, K16.
180 days +0.21°C; H68; K21.
All 2017 +0.14°C, Heathrow warmer.

I hope the abbreviations make sense, my fingers are overworked!

So Heathrow maxima on at least the warmest 180 days of 2017,  were mostly higher than the better sited Kew Gardens. The exception I noticed is on the coldest 30 or so days of the year, when Kew was a bit warmer. There's enough there for me to be interested in digging deeper and for other years, with monthly means as well, but it may take a while.  However, if the above results are supported, there is the thorny question of... why? We have, of course, previously discussed that point.

All comments welcome, this is a first attempt and I am not a statistician (but I can use Excel).

Great effort! But could that just be down to Kew having significatly more moisture in the air thanks to tree transpiration and shading?  It would not heat as much in summer or cool as much in winter.

[B87]

Kew Gardens has slightly warmer summer highs than Heathrow, on average. It is cooler at night though, the area around Richmond Park having the lowest minima of any part of London.

 

 

 

[Derecho]

44 minutes ago, Aleman said:

Good contribution. Thanks.  I like the pythagoras trick.  But I don't agree with your conclusions at all. I didn't think I needed to elaborate but here goes:

Wind: at least 10 times the height of obstruction. FAIL. Trees are both 15m+ tall and the trunks are no more than 25m away - and the canopy brings them even closer.

Yes I did say for wind it is dubious but do they monitor wind anyway? I can't see an anemometer

Temperature: at least 4 times obstruction's height, at least 30m from large paved areas. Trees are 15m+ tall and no more than 25m away. FAIL. If a large paved area encroached to 25m , it would be a fail. A 4m path plus addition 1m of paving no more than 2m from the temperature sensor is clearly a FAIL. It's a large paved area - not a stepping stone! Other organisations' station parameters indicate NO heating elements inside 30m (as per old Met Office guidelines) to get a Class 2 and inside 10m to get a Class 3. This site would be a class 4 for paving, railings and lamp post. That's not good enough for what it is being used for.

The guidelines on the link you posted are vague, just what exactly is an obstruction. The WMO class an obstruction as follows:

"An obstacle is an object (ie. tree, building, etc.) with a horizontal angle of 10 deg. and wider that can shade the sensor enclosure. ". Given the google maps image clearly shows the site away from the shade, is it an obstruction?

A lot more information can be found from the WMO guidelines. For class 1 these are:
 

 

Solar: objects to be below 10 degrees from horizon. FAIL Tree to east about 40 degrees plus lampost to south is over 10. FAIL

Precipitation: at least 4 times obstruction's height. Still the trees. FAIL.

Soil Temp: 1m x 1m site typical of area and flat for 10m radius. FAIL, PASS. It can' be typical of the area if the area if there is so much paving near. For instance, rain run off could mean extra water added to soil on that side of the site.

 

As an added bonus, there are three rectangular manholes about 20m or so to the southwest. So what is going on underground around the station?

Class 1 - the highest

• T1) Ground within a 10m radius ‘mostly’ covered with lawn or natural and low vegetation (generally less than 10cm) representative of the region. Probably a fail as the path covers a good deal of the 10m radius 
o Open patches of bare ground acceptable if natural and representative of the region.

• T2) Generally flat area within 10m of sensor. This area surrounded by generally open space with a slope of less than 1:3 (19°) that is considered to be representative of the large scale area. PASS, its flat

• T3) No irrigation or routine lawn watering system operational within 30m radius. Pass

• T4) Sensor located at more than 100m from significant heat sources or reflective surfaces (buildings, roads, concrete surfaces, car parks, unnatural exposed ground areas, etc.). This also includes expanses of water even if only seasonal. Fail

o Open water can be excluded if representative of the region such as open coastal or large scale inland water/swamp/muskeg (ie. 25% of 100km radius, approximately equal in most directions).

A ‘significant’ source of heat (or expanse of water) is defined to have an impact if it comprises more than 10% of the horizontal surface within a 100m radius surrounding the sensor, comprises more than 5% of the horizontal surface within 30m radius or comprises 1% of the horizontal surface within a 10m radius.
10% of 100m radius circle = 3142 sq.m (32,212 sf)
5% of 30m radius circle = 141 sq.m (1368 sf)
1% of 10m radius circle = 3 sq.m (33 sf)

• T5) Sensor away from any subjectively determined heat source such as nearby infrastructure (ie. electronics boxes, metal mounting structures, etc.) that will have any affect on sensor. Minimum distance from sensor to nearby other objects at least 2x the height of the object away (2:1). The fence around it probably doesn't fall into this class of objects, so PASS?

• T6) Site not subjected to any jet blast, prop wash or helicopter downwash. No routine significant aircraft ground movement/parking or significant paved public highways within 100m. Fail, it's a class two.

• T7) Sensor in a ventilated solar radiation shield or enclosure. PASS

• T8) Sensor mounted between 1.2m and 2.0m above ground level. PASS

• T9) The maximum anticipated (subjective and non-extreme) snowdepth will remain at least 1m below the bottom surface of sensor shield or enclosure. Maybe a fail if a 1947 style winter came along.

• T10) Sensor radiation shield away from all significant shade when the Sun is higher than 5 deg. above local geographic horizon. Fail due to the shrubbery behind the sensor.

More detailed info on classification systems can be found here...

https://www.wmo.int/pages/prog/www/IMOP/SitingClassif/Canada/Siting Classification System7 - Sep 2012.pdf

I'd say its a class 2 station. If everywhere was a class 1 station how limited would we be with regards to potential site placements? These sites have been there for many decades (St James Park 1903 I think). The main thing is that the temperature data is homogenous for long term climate studies. The readings may not be top notch perfect from an open station point of view but surely they are representative of the overall area they are in.