There are often comments here about UHI being -0.10 ± 0.24°C/100yr, that is not measurable. Whilst on the other hand during anticyclonic conditions the Met Office via the BBC says that the difference between rural and urban areas can be several degrees, to me several equates to more than two and less than eight. So for a significant number of days there is a difference of several between rural and urban temperatures with urban being more positive. Does this significant number of days really have no effect on UK average temperatures at least?

I accept than on days with a fair amount of wind there will be no difference between the two I am still left wondering why there seems to be no measurable difference. Is Europe different from the rest of the world in having low wind high pressure periods. Is the rest of the world like Wellington, Chicago, Punta Arenas and Troy which are known for being windy? Are there other factors which mean urban heat is rapidly transferred to rural areas, or does it go directly into space?

I really would like an explanation of this apparent contradiction, a myriad of links won't really help as internet time is strictly limited for the next couple of months.

To summarise, it is about the trend

Urban areas are indeed warmer than their surrounding rural areas, sometimes by several degrees C. Urban areas are over-represented in the surface station record and an argument has been presented that some of the measured planetary warming may be in fact an artifact due to the UHI increasing. (A thought experiment: consider a thermometer in say, rural Wiltshire and another in central London. The London instrument always reads 4C warmer than the countryside. In the presence of a gradual warming trend both would have the same trend and there would be no contamination. If the urban area became more urban over time and hence warmed faster than the actual global trend then it would contribute a non-climatic signal to the measured trend.)

For this to be the case, two conditions must be satisfied, firstly the trend in urban measurements must be elevated due to increased urbanisation over time and secondly the data quality adjustments designed to identify and remove artificial trends must be failing.

Numerous studies have been done, usually comparing urban to rural trends etc, all have concluded that UHI contamination is between zero and negligible and the homogenisation process does a good job of removing it.

From first principles, this is unsurprising, a tiny fraction of the land surface is urban and the land surface is about 30% of the total surface area. No UHI in the oceans.

There are two issues with UHI which tend to be mixed up.

1) UHI is a form of AGW. As the surface area of our cities increases, the surface albedo decreases and more heat is absorbed. Up goes the temperature.

Heathrow is an extreme example. I am amused by golf charlie's suggestion that the station be moved away from the airport. Firstly it allows her to deny that London has a warming effect. Secondly, the temperature data is in continuous use by flight planners calculating takeoff performance. Higher temperatures mean longer takeoff runs. On a hot day an aircraft may have to carry less fuel or payload to achieve a safe departure.

2) Long term temperature trends are best studied using data from stations in which the effect of the surroundings on the measurements has remained stable over time. Since the longest running stations are where people live and need the measurements, a proportion of the longest running stations will inevitably be in urban areas and airports.

If you want to detect long term trends other than UHI, then you need to allow for increasing UHI and apply a cooling correction to those stations.

SandyS, what Phil Clarke and Entropic Man are doing, is acknowledging UHI exists, but then denying it has any distorting consequences for records in global warming.

Phil Clarke then talks about adjustments and homogenisation, which is a bit like Banks making errors that are NOT in their favour.

Entropic Man then waffles about Heathrow with further added lies. When have I stated that the weather station should be moved from Heathrow Airport? I have stated that reporting records from Heathrow, the world record example of distortion of global warming by exaggerated UHI is a farce. This is the sort of blatant lie and distortion that I thought Phil Clarke used but EM didn't.

Obviously UHI is a far more sensitive subject for global warming panic merchants, than was previously thought credible, and warrants more attention to detail.

The basic points Phil C and EM make are - self-evidently - correct. It's the conclusions drawn therefrom that are problematic.

E.g. "Numerous studies have been done, usually comparing urban to rural trends etc, all have concluded that UHI contamination is between zero and negligible and the homogenisation process does a good job of removing it.

From first principles, this is unsurprising, a tiny fraction of the land surface is urban and the land surface is about 30% of the total surface area. No UHI in the oceans."

Yes, but that tiny fraction of the land surface represents a huge - and therefore potentially contaminating - part of the datasets. There is previous little of any use in terms of long-term, reliable and accurate temperature datasets over the almost 70% of the surface of the planet comprised by the oceans. Of the land datasets, there is a huge (and obvious and understandable) bias towards the contiguous states of the USA and the UK, with much of Africa and Asia and South America being a big blank where the records rely on infilling.

I have just got in not long since, and haven't had time to peruse Phil's links in detail, but need to do so to see how robust is the logic and analysis contained therein. They had better be an improvement on Petersen's work (which was generally excellent, until it fell over at the conclusions stage) if they're to convince me.

Mark Hodgson, what Phil Clarke is actually saying is that loads of studies have been carried out to try and prove that UHI can be discounted as a cause for Global Warming being recorded. Along with deleting the Medieval Warm Period, this was necessary to make the Hockey Stick work.

Even before Climategate revelations, Steve McIntyre got stuck into the mysterious goings-on that did and did not happen, before and since Jones 1990.

If the Hockey Stick "settled the science", why so many failed attempts to repeat it?

If Jones 1990 "settled UHI", why so many failed attempts to repeat it?

If you have 10 minutes start here:

https://climateaudit.org/2010/11/03/phil-jones-and-the-china-network-part-1/

you may decide to spend longer ....... I did! Temperatures are not getting hotter, but they may not be quite so cold, and if this is what global warming is, who cares?

golf charlie,
That is the problem, just how much is the data contaminated?

Entropic man says

the surface albedo decreases and more heat is absorbed. Up goes the temperature.

The Met Office/BBC says the urban/rural difference can be as much as 8'C (tonight East Midlands News) if EM is correct and I think that he is then none or very little of this can be attributed directly to CO2. As urbanisation increases reducing CO2 emissions will not have any impact on UHI. Going back to the contamination problem the alleged correlation between CO2 and global temperature will get more tenuous than it already is.

I've just been looking at Phil C's link to the Berkeley Earth paper which supposedly debunks the UHI effect in global temperature datasets. I'm far from convinced. For instance it includes this:

"In Table 1 we see evidence of “global warming.” Using all the records there is a median warming trend of 0.96 ± .04°C/100yr (2σ error). The estimated warming trend for the very-rural group is larger than that based on all records, in the opposite direction expected from urban heating. The difference observed in this simple analysis reflects that there are many sources of variation in individual station trends that contribute larger effects than any effect due to urban heating. To extract the urban heating contribution from the individual trends a careful analysis would involve modeling known sources of variation, such as geographic location and measurement differences as well as accounting for spatial distribution and correlation. Since our primary interest is to evaluate the effect in the global average we do not pursue
this approach further."

Just stop and think about that for a moment. They are actually saying that rural sites show more heating than urban ones. Does that not tell them that there's something wrong with their data?

Look also at the last 2 sentences. They actually say that to extract the urban heating contribution they need to make a careful analysis modelling known sources of variation, but they "do not pursue this approach further". It doesn't stop them concluding:

" Time series of the Earth’s average land temperature are estimated using the Berkeley Earth methodology applied to the full dataset and the rural subset; the difference of these is consistent with no urban heating effect over the period 1950 to 2010, with a slope of -0.10 ± 0.24/100yr (95% confidence)."

And this stuff passes peer review. Bizarre!

SandyS & Mark Hodgson

RAF Benson (inbetween Reading and Oxford) is in a notorious frost pocket, and "often" records the coldest English temperature. It is not in a built up area. It does not have a high number of aircraft movements.

I have no idea where its weather station is, but I would be intrigued by their temperature records going back to the 1930s, but they may have been "adjusted", to reflect operational funding.

https://wattsupwiththat.com/2016/10/18/define-irony-photovoltaic-heat-island-effect/

Apparently farms of solar panels cause UHIs. DOH!

SandyS

UHI was quoted above as producing around 0.1C/century warming. The overall warming since 1950 has been 0.8C, equivalent to 1.2C/century. Unless you can produce something else wit comparable warming power it seems premature to reject CO2.

golf Charlie

Doh, indeed. Solar farms are designed for minimum albedo and whatever is not converted to electricity end up as locally released heat. Of course they produce a local heating effect similar to UHI.

From a warmist perspective the key point is that this heating is less than would be produced by the CO2 produced by equivalent fossil fuel generated electricity.

Osseo

See what I mean. Another foolish comment from golf charlie.

Entropic Man, actually another dumb series of remarks from you. You really are descending to Phil Clarke's level of infantile snark.

Thank you for emphasising that climate science ran out of science. Shame that taxpayrs still have to fund it.

Entropic man
My original problem is that for many days (periods of 24 hours) when there is little or no wind there is a tacitly acknowledges difference between rural and urban temperatures of up to 8'C. When calculating monthly temperatures using an average value of 0.1'C - 0.5'C, or what ever very small value is used gives an erroneous value to daily/weekly/monthly/annual temperatures. As you agree the UHI is not a result of CO2.

So unless Europe is unique in having periods of low wind lasting weeks then the contamination of global temperatures through albedo changes is quite significant. So far you haven't given me a reason to change my mind that getting 25% of daily temperatures wrong by a significant amount makes the whole data set worthless. 25% is a value I derived from looking at wind generation data from the National Grid. I'd be interested in your estimate of the annual percentage of low wind days.

EM

The Berkeley Earth report I criticised above acknowledges a 2C UHI/per century effect in Tokyo, yet concludes by dismissing UHI as insignificant. So much for 0.1C per century of UHI warming.

Mark Hodgson, climate science has had to acknowledge UHI as a factor in recorded global warming, and then produce methods of adjustment to dismiss it as a factor, with peer reviewed "climate science".

Unfortunately, "peer review" in climate science has been persistently tarnished by the wholly inadequate standards of professional climate scientists, and the Gold Standard of Peer Review has been devalued to that of pigswill.

Mark Hodgson
2C UHI/per century would seem reasonable to me.

My thought process is as follows

1. Other than in exceptional circumstances urban temperatures will always be greater than rural. Exceptional circumstances might be a chinook wind in an exclusively rural area in northern Canada for example.

2 For 75 days in every year there is not enough wind to disburse air warmed due to decreased urban albedo.

3 During those days the measured increase of urban temperatures depends on strength and direction of wind

4 During those days the measured increase of urban temperatures depends on the strength of sunlight reaching the ground. (mist and fog reducing the solar input to virtually nil potentially)

5.The increase in measured urban temperatures depends on the land area of urbanisation

6 During those days the measured temperature also depends on the amount of artificial heat (planes, buses, cars, air-con) added.

7. For the rest of the year when reasonable amounts of wind move the warmed air into rural areas and mix it with cooler rural air the UHI temperature will still be higher than rural but perhaps not accurately measurable.

In order to calculate the affect of UHI on averages you have to account for these factors on an daily and hourly basis, National Holidays, school holidays, Sunday/Sabbath, in working hours and so on. Taking a number and applying it as a blanket adjustment creates more problems than it removes.

I still can't balance the two issues I initially highlighted as being correctly adjusted for in "global average temperature"

Sandy S

I am inclined to agree with your thought processes.It seems to me that UHI is in many ways the Achilles heel of climate alarmism. That is why, no doubt, so many papers have been written saying either that it doesn't exist or (a mutually contradictory position, but when did that stop "climate science"?) or that to the extent that it exists, it has been compensated for by adjustments which identify and eliminate it.

The problem is:

1. Statements that it doesn't exist are clearly bonkers. Alarm bells should ring long and loud when papers claim that the countryside is warmer than the towns, or is warming faster than the towns. But apparently they don't - they print this cr*p with an apparently straight face;and

2. The alternative way of getting the problem out of the way (adjustments) is fraught with difficulty. Petersen's paper attempted to go down that road a bit (whilst,in direct contradiction, also adopting in part argument #1) but failed, to my mind at least, to be persuasive and convincing regarding the accuracy of the adjustments.

Given the refusal/failure of the "consensus" to deal properly with this issue, the whole pack of cards is in danger of tumbling down.

Mark Hodgson
I'd like Entropic Man, who usually gives a civil answer, to explain without internet links where I'm going wrong.

Sandy S

Unfortunately Phil C and Entropic Man don't seem much interested in UHI threads - because they're on a hiding to nothing? Neither of them made a single comment on the UHI thread I started a few weeks ago.

SandyS, Mark Hodgson

Phil Clarke and I are the only warmists still bothering to engage here. We get spread thin.

The big heat capacity reservoir in the climate system is the ocean. That makes it the place to watch for an overall climate trend.

UHI is one of the minor factors distorting the long term land temperature record, mainly for the historical reason that the most avid consumers of live temperature information are aviators and airports tend to be near cities.

That biases the land record up slightly. In practice, the best indicator of how the climate is changing from the scientific viewpoint are the land-ocean temperature datasets.(which also answers TinyCO2's question)

My advice would be to stop worrying about UHI as a factor in planetary temperature change. If you actually live in a large city, it may be a local problem for you.

SandyS, Mark Hodgson

Phil Clarke and I are the only warmists still bothering to engage here. We get spread thin.

The big heat capacity reservoir in the climate system is the ocean. That makes it the place to watch for an overall climate trend.

UHI is one of the minor factors distorting the long term land temperature record, mainly for the historical reason that the most avid consumers of live temperature information are aviators and airports tend to be near cities.

That biases the land record up slightly. In practice, the best indicator of how the climate is changing from the scientific viewpoint are the land-ocean temperature datasets.(which also answers TinyCO2's question)

My advice would be to stop worrying about UHI as a factor in planetary temperature change. If you actually live in a large city, it may be a local problem for you.

EM

Thank you for your response. I appreciate you taking the time and trouble to respond. However, I cannot stop thinking about UHI (I don't worry about it) whilst those in control of the temperature datasets deal with it so poorly.

I agree that ocean trends are important. I look forward to the day when we have a lengthy and consistent set of ocean temperature measurements over a sufficiently large area as to add meaningfully to the sum of our knowledge.

Sandy S

Currently the Argo network is sampling 3000 locations every ten days, sufficient to give good ocean temperature data.

Statistically 1500 would be enough to reach the natural variability threshold.

We will have 30 years data in the early 2030s.

On the subject of homogenisation, try here.

Note how the difference between raw and adjusted data reduces with time. After 1980 the difference is smaller than the thickness of the lines. The effect of UHI is effectively too small to see.

Mark Hodgson, SandyS

Damn, that should have been addresed to you both.

EM:

No worries, and I appreciate the link to Victor Venema's blog. I accept that he is the guru of homogenisation etc, and I will read the link carefully. However, having got this far, I'm already struggling to take it all too seriously:

"The best known inhomogeneity is the urban heat island effect. The temperature in cities can be warmer than in the surrounding country side, especially at night. Thus as cities grow, one may expect that temperatures measured in cities become higher. On the other hand, with the advent of aviation, many meteorological offices and thus their stations have often been relocated from cities to nearby, typically cooler, airports (Trewin, 2010). It would be worth studying which of these two effects are strongest for urban stations. My European colleagues expect it is the cooling due to the relocation to airports."

"cooling due to relocation to airports" indeed. In the UK we seem to have heat records reported regularly at Heathrow. The idea that airports are cooler (and presumably should see their temperatures adjusted upwards?) isn't one I can take terribly seriously, and I'm surprised Victor can write this apparently with a straight face.