Guest Post by Willis Eschenbach
[OK, had to start over, had bad numbers for the areas. Graphics have been replaced. I was using “core area” but I should have been using “greater metropolitan area”. All conclusions are unchanged.]
I got to thinking about the phenomenon known as the “Urban Heat Island” effect, or UHI. Cities tend to trap heat due to the amount of black pavement and concrete sidewalks, the narrow canyons between buildings that slow down the wind, and the sides of the buildings reflecting sunlight downwards.
As a result, cities are often warmer than the surrounding countryside. In some cities, it’s hot enough that it affects the local weather. Here’s a simplified diagram:
What I was curious about, however, was another kind of urban heat effect. This is the heat from all of the energy used within the city—electricity, fuel for transport, fuel for heating buildings, all of it. Eventually, almost all energy ends up as heat. So I went and got the energy usage for 27 huge “megacities”, along with the area of the city itself. I then combined the two to give me a measure of citywide energy usage in watts per square metre (W/m2). As a measure for comparison with Figure 2 below, a doubling of CO2 is said to increase the “forcing”, the total radiant energy impinging on the surface, by 3.7 W/m2.
This was surprising to me. I hadn’t expected the effect to be so large.
Finally, I converted the forcing to an equivalent warming. The Intergovernmental Panel on Climate Change (IPCC) fifth assessment report, completed in 2014, gave a likely “Transient Climate Response” of about 1°C to 2.5°C for each additional 3.7 W/m2 of forcing. I’ve used 1.5°C because the heat generation persists over time. As a result, the cities have had time to equilibrate to the additional heating. Figure 3 shows that result.
Not much more to say about all of that. Note that this is a “first cut” analysis, I make no overarching claims about the accuracy of the results. I’ve used conservative assumptions and the best data I could find. It looks to be a significant additional heating source due to direct energy usage in the densest of the largest cities, one that is not generally included in the calculations of the Urban Heat Island.
Here, I’m in the Forest Cool Island, life is good. I spent my morning crawling around under my house successfully putting a new “generator” into my floor furnace so we now have a warm house again, a less than pleasant job that came complete with a veritable plethora of bad words. Then I got out the pumice stone and scrubbed a toilet bowl until it sparkled, and this has been my afternoon project … do I know how to have fun, or what?
My very best wishes to each of you,
w.
[UPDATE] In the comments below, someone asked about human body heat and how that affects the forcing in megacities. Humans on average put out on the order of 120 watts continuously. Here’s how that plays out.
You can see how densely populated Seoul, South Korea is …
DATA: I’ve appended the data below, in comma-delimited form. PJ is petajoules, 10^15 joules.
City, PJ, km2,Population Beijing, 952, 6562, 21516000 Buenos Aires, 702, 10888, 3054000 Cairo, 282, 1600, 10230000 Delhi, 316, 3182, 11034000 Dhaka, 350, 1353, 8906000 Guangzhou, 1474, 7711, 14043000 Istanbul, 464, 5500, 14025000 Jakarta, 589, 5100, 10075000 Karachi, 339, 1100, 14910000 Kolkata, 78, 1785, 4486000 Lagos, 350, 1535, 861000 London, 1065, 11391, 8825000 Los Angeles, 1848, 10780, 3884000 Manila, 918, 2521, 1780000 Mexico City, 1099, 7346, 9041000 Moscow, 1984, 14925, 12197000 Mumbai, 191, 2350, 12478000 New York, 2824, 17884, 8622000 Osaka, 1258, 6930, 2691000 Paris, 657, 17174, 2229000 Rio De Janeiro, 384, 4540, 6718000 Sao Paulo, 589, 8479, 12252000 Seoul, 1848, 5076, 10197000 Shanghai, 1644, 5177, 24256000 Shenzhen, 350, 3051, 8378000 Tehran, 1145, 9500, 8154000 Tokyo, 2438, 8014, 13839000
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Nice work, Willis,
This fits comfortably with a boring, lengthy review of UHI that I posted on WUWT a year or so ago. My strong conclusion was that UHI is an effect that probably distorts land surface records unacceptably, but those in paid research that should research and correct it at trying to avoid it. Mostly, Though I did reference a few papers worth reading. Geoff S
Auckland city is the largest in NZ. Assuming that it is powered by NZ’s largest hydro power plant, I have always assumed that the city is warmed by the amount of power used. I have asked a number of people what would happen if the power to Auckland was cut and the water was allowed to run down the spillway. Would the river heat up? I reckon it would, and electrical engineers agree with me. Everybody else disagreed. Everybody else is wrong!
“As a measure for comparison with Figure 2 below, a doubling of CO2 is said to increase the “forcing”, the total radiant energy impinging on the surface, by 3.7 W/m2.”
err no. That’s TOA
As for anthropogenic heating?
why guess when there is science?
data I have discussed before
http://www.cgd.ucar.edu/tss/ahf/
other stuff
https://www.nature.com/articles/sdata2017116
one of my favorite authors
https://www.sciencedirect.com/science/article/pii/S1352231015302156
The great thing is the vast vast major of GHCNV4 stations are not in urban areas.
There is large literature on this issue. No envelope backs required.
In fact if you read the primary literature on UHI, as any good researcher should do, you will find references in things called footnotes!
The temperature meters don’t have to be *in* the urban areas, just down wind of them, in order to suffer from UHI.
“It looks to be a significant additional heating source due to direct energy usage in the densest of the largest cities, one that is not generally included in the calculations of the Urban Heat Island.”
huh
The urban heat island is typically measured in one of two ways.
1. A comparison of an urban station with a nearby rural station.
2. A transect run across the city.
The Difficult job is ACCOUNTING FOR THE SOURCES OF THE EXCESS HEAT.
http://meteora.ucsd.edu/cap/docs/sailorandlu.pdf
One of the sources is from Anthropogenic heating.
So you want to know about Seoul? Well go study the science!!
https://link.springer.com/article/10.1007/s13143-015-0065-6
Willis: “It looks to be a significant additional heating source due to direct energy usage in the densest of the largest cities, one that is not generally included in the calculations of the Urban Heat Island.”
Science: “Anthropogenic heating by human activity is one of the key contributing factors in forming urban heat islands, thus inclusion of the heat source plays an important role in urban meteorological and environmental modeling. In this study, gridded anthropogenic heat flux (AHF) with high spatial (1-km) and temporal (1-hr) resolution is estimated for the whole South Korea region in year 2010 using a statistical regression method which derives based on similarity of anthropogenic air pollutant emissions and AHF in their emission inventories. The bottom-up anthropogenic pollutant emissions required for the regression method are produced using the intensive Korean air pollutants emission inventories. The calculated regression-based AHF compares well with the inventory-based AHF estimation for the Gyeong-In region, demonstrating that the statistical regression method can reasonably represent spatio-temporal variation of the AHF within the region. The estimated AHF shows that for major Korean cities (Seoul, Busan, Daegu, Gwangju, Daejeon, and Ulsan) the annual mean AHF range 10–50 Wm−2 on a grid scale and 20–30W m−2 on a city-scale. The winter AHF are larger by about 22% than that in summer, while the weekday AHF are larger by 4–5% than the weekend AHF in the major Korean cities. The gridded AHF data estimated in this study can be used in mesoscale meteorological and environmental modeling for the South Korea region.”
Willis: “It looks to be a significant additional heating source due to direct energy usage in the densest of the largest cities, one that is not generally included in the calculations of the Urban Heat Island.”
the science
http://www.meteo.fr/icuc9/LongAbstracts/gd5-1-2961311_a.pdf
“All cities emit heat into the planetary boundary layer. A small but important source of this heat is human
activity and its associated burning of fossil fuels for urban transport, industrial processing, and domestic
heating and cooling (Oke, 1982). Combustion processes in cities set the anthropogenic heat flux, which is
an important forcing term in models of the urban heat island effect and global climate change (Allen et al.,
2011; Flanner, 2009). In perhaps the earliest study to quantify the thermal effects of human activity on
urban climate, Eaton (1877) calculated the heat released from coal combustion in the Metropolitan District
of London in the late nineteenth century. Accounting also for the ‘vital heat’ of the city’s 3.5 million
inhabitants, Eaton predicted these sources to raise the air temperature in London by 1.4 K (cited in Garnett
and Bach, 1965). In a more detailed study, Ichinose et al. (1999) used numerical models to simulate air
temperature changes of > 2 K by these same effects during winter nights in central Tokyo.
Published estimates of anthropogenic heat release originate mostly from wealthy, mid-latitude cities such
as Tokyo and London, with few comparable estimates from cities of tropical or low-income regions. The
literature could therefore profit from a universal set of anthropogenic heat flux densities representing cities
of diverse geography, and having consistent methods of derivation. Herewith, we initiate the development
of such a dataset. Using an inventory approach, we calculate heat flux densities at metropolitan scale for
the world’s 27 megacities, i.e., urban agglomerations of more than 10 million people. The global distribution
of these cities—and our use of a common methodology—allows for regional insight into the economic,
climatic, and demographic influences on anthropogenic heat release in urban environments.”
“The urban heat island is typically measured in one of two ways.
1. A comparison of an urban station with a nearby rural station.”
It’s not just sufficient to use a nearby rural station. The rural station has to be located outside the impact zone of the UHI.
Tim
You said, “The rural station has to be located outside the impact zone of the UHI.” That means it has to be upwind, or orthogonal to the wind direction.
Thanks, Mosh. You keep saying that you know all about UHI, and how Berkeley Earth has removed the UHI from their data. Perhaps you’d explain this, then.
The “WORLD AVG” is the global average temperature trend per Berkeley Earth … and of all of the datasets, Berkeley Earth has the highest trend.
Regards,
w.
Sorry the science is way ahead of you.
http://www.meteo.fr/icuc9/LongAbstracts/gd5-1-2961311_a.pdf
I see you’ve provided a clone to repeat your talking point – I’d say the propaganda is way ahead.
To quote the given reference on its methodology:
The bald assumption that ALL energy consumed produces ONLY heat–and no work or light–is symptomatic of Mosher’s grasp of “way ahead” science.
Not sure what you are using for city areas.
city area can be defined two ways.
1. ADMINISTRATIVELY. the land the government defines as belonging to the city
2. BY LAND COVER. the land that actually represents the urban cover and not, for example, the
bordering regions which may or may not be administratively defined as being part of the city.
Mosh, thanks for all of that, always good to hear from you. You say “the science is way ahead” of me and that this is all well known. But the link you provided with the claim that science is way ahead has a total of 9 citations. The UCAR paper is about country-level energy use, not cities. And the “new global anthropogenic heat estimation based on high-resolution …” paper has 10 citations.
The 2015 Elsevier paper is the only one that’s on point and has gotten traction. I was surprised to see that the Elsevier paper gives values much, much higher than the ones that I calculated. For example, they say that anthropogenic heating in NY City is 63 W/m2 … this is about 40% of annual average absorbed sunshine in NYC. And in the winter, anthropogenic heat is given as 97 W/m2, which is larger than absorbed sunshine at 80 W/m2 for DJF. That seems awfully high.
Meanwhile, here’s a paper with some guy called “Mosher” as one of the authors … it says:
Notice what’s missing in your description? The guy busting me didn’t consider anthropogenic heat important enough to even mention. To make sure I wasn’t missing anything I followed the reference [7], which is to the seminal Oke paper that says:
So Oke is not considering anthropogenic heat, and even when he does he only includes direct heating of the air, and then only what escapes through doors, windows, etc.
Finally, you’ve busted me for saying:
“It looks to be a significant additional heating source due to direct energy usage in the densest of the largest cities, one that is not generally included in the calculations of the Urban Heat Island.”
Since neither Oke nor you included anthropogenic heat in your discussion of UHI, I’d say I was not too far wrong … I’d say we’re both half right on that one.
In any case, thanks for the interesting links.
My best to you as always,
w.
I also note that in the paper by the Mosher fellow it says (emphasis mine):
Now Mosh, I assume that this is the real finding from the real data. You’re too good a scientist to do otherwise. But I gotta ask … with the huge anthropogenic heat effect in the papers you cite, plus the effect from asphalt and concrete, and the large number of temperature stations in urban areas (10% or more), what’s your explanation for y’all finding “no urban heating effect”?
One possibility is that it is an artifact of the “scalpel method” used by Berkeley Earth. I’ve asked you and Zeke a couple of times about this, for example in my post “Problems With The Scalpel Method“. The nearest thing to an answer I got was that you’d analyzed this and dismissed it. I asked where I could find that analysis … crickets.
Finally, theres an interesting and informative post about UHI by Mosher here on WUWT, worth reading. No mention of anthropogenic heat in it, but fascinating otherwise.
w.
As an engineer, if the readings before and after up to the next significant digit and go from there. That would truly be all the accuracy you could expect from that measurement device if aging has an impact on the readings.
The problem that causes for the CAGW alarmists is that this would make it impossible to see the heat increases they say we are having on a global basis. It’s why taking temperature measurements only accurate to the nearest degree or the nearest tenth of a degree and averaging them to produce a figure supposedly accurate to the nearest hundredth of a degree is such a joke! You simply cannot increase accuracy by implementing averages. 2+2 is not equal to 4.0!
yes the science is STILL way ahead of you
http://www.meteo.fr/icuc9/LongAbstracts/gd5-1-2961311_a.pdf
You still dont get it.
We measure the UHI as a temperature differential.
say its 5C worst case.
PARTITIONING THAT into various causes has been worked on for ages
“All cities emit heat into the planetary boundary layer. A small but important source of this heat is human
activity and its associated burning of fossil fuels for urban transport, industrial processing, and domestic
heating and cooling (Oke, 1982). Combustion processes in cities set the anthropogenic heat flux, which is
an important forcing term in models of the urban heat island effect and global climate change (Allen et al.,
2011; Flanner, 2009). In perhaps the earliest study to quantify the thermal effects of human activity on
urban climate, Eaton (1877) calculated the heat released from coal combustion in the Metropolitan District
of London in the late nineteenth century. Accounting also for the ‘vital heat’ of the city’s 3.5 million
inhabitants, Eaton predicted these sources to raise the air temperature in London by 1.4 K (cited in Garnett
and Bach, 1965). In a more detailed study, Ichinose et al. (1999) used numerical models to simulate air
temperature changes of > 2 K by these same effects during winter nights in central Tokyo.
Published estimates of anthropogenic heat release originate mostly from wealthy, mid-latitude cities such
as Tokyo and London, with few comparable estimates from cities of tropical or low-income regions. The
literature could therefore profit from a universal set of anthropogenic heat flux densities representing cities
of diverse geography, and having consistent methods of derivation. Herewith, we initiate the development
of such a dataset. Using an inventory approach, we calculate heat flux densities at metropolitan scale for
the world’s 27 megacities, i.e., urban agglomerations of more than 10 million people. The global distribution
of these cities—and our use of a common methodology—allows for regional insight into the economic,
climatic, and demographic influences on anthropogenic heat release in urban environments.”
lastly read my words.
Do you understand what I MEANT by
“INCLUDING”
and what I MEANT by
“OTHER PHENOMENA.”
Anthropogenic heat flux is typically a small percentage of the cause of UHI.
So you are still behind on the science. do keep up
Still no answer to the scalpel question?
And could you cut down on the snark? It makes you look like a dickhead, and I know you aren’t one.
w.
Willis
If it walks like a duck, and quacks like a duck, …
Ah yes, I see that steve is still trying to pull out the “you just don’t understand how science” works BS. It’s easier than actually trying to come up with an explanation that holds together mathematically.
“and domestic heating and cooling (Oke, 1982)”
Raise your hands those of you who believe we use significantly more energy now than in 1982 for heating and cooling… I assume of course that all those power plants built in the mean time was built for a reason.
Willy, you have been tricked with BA’s. The Federal Capital of Buenos Aires has 3 million people.
But this is just the small central area. But the much larger entire mega city of BA’s has over 13 million people.
And there is a major heat island effect happening.I know because I lived there for 3 months in 2014 to learn & dance Tango,
“We measure the UHI as a temperature differential.”
What temperature differentials? Between urban thermometers and rural thermometers within the UHI impact zone? I.e. “nearby rural staitions”?
Or rural stations that are well within the UHI zones of smaller cities outside the big cities.
Just because a station is listed as rural (or was listed in the data base 10 to 15 years ago as being rural) does not mean that the station is pure (nor does it mean that the station is still rural today).
I grew up in the Austrian countryside. Winters were snowy and cold. The landscape was the same as Vienna which was only 60km away. We went to the capital often to see family. Every time we went there we could see a pretty obvious change. Less snow and not as freezing as it was at home. The urban heat island at work. My father explained it to be in absolutely unscientific terms. It was a thing during his childhood as well and my father is pushing 85. Only now we seem to have forgotten about it and think the globe warms. Get out of the city …
Again, it looks to me as though Mr. Eschenbach didn’t treat “forcing” correctly. I’ll provide an undoubtedly oversimplified, radiation-only analysis to explain why I think he may have erred. Executive summary: I think forcing takes its value from the top of the atmosphere, not from the surface.
So what do I think forcing is? Consider the usual case: forcing from CO2.
If from an equilibrium state the CO2 concentration somehow makes an abrupt quantum jump and the atmosphere’s infrared opacity therefore does, too, the effective radiation altitude will abruptly increase. Because of the lapse rate the new radiation altitude has a lower temperature than the old one, so outgoing radiation falls abruptly, changing an erstwhile radiation balance to an imbalance.
It’s that initial imbalance’s value that I think is the forcing value associated with the CO2-concentration increase. The imbalance would decay, of course, but it’s the initial imbalance value that remains as the value of the concentration change’s forcing; even though a new equilibrium is established for the new concentration, its forcing value remains the value of that initial imbalance.
How is the imbalance redressed? The imbalance causes the surface temperature to rise and (modulo any lapse-rate change) thereby increase the radiation-altitude temperature by the same amount until the radiation out returns to the radiation-in value. The value of that top-of-the-atmosphere initial imbalance is associated with the surface-temperature change needed to redress it.
However, the surface temperature changes from a value higher than that of the radiation-altitude temperature, so the surface-radiation increase that attends the surface-temperature increase exceeds radiation-altitude’s increase: it exceeds the initial radiation-altitude imbalance, i.e., the forcing. The surface-radiation increase may be, say, half again the forcing value.
It’s not clear to me that Mr. Eschenbach took this into account; I got the impression that he was going by surface, not top-of-the-atmosphere values. But if anyone has a different understanding, I’d love to hear it.
Sounds rubbish to me. How does CO2 “force” anything? What was it forcing at 280ppm/v? What is it forcing at 410ppm/v? Forcing as in the term used to describe something acting on something else it doesn’t normally do, right?
Although I have certain reservations about the concept, you may want to read up on what radiative forcing is conventionally understood to mean: https://www.ipcc.ch/site/assets/uploads/2018/02/WG1AR5_Chapter08_FINAL.pdf
What is this “degrees C” you’re going on about? What’s is it in atom bombs / sec (the correct measure of AGW)?
“Cities tend to trap heat due to the amount of black pavement and concrete sidewalks, the narrow canyons between buildings that slow down the wind, and the sides of the buildings reflecting sunlight downwards.”
As a guy who remembers going barefoot and feeling cool grass versus hot sidewalks. I have to ask. Doesn’t the moisture of grass and soil play a role in making dry things hot and wet things cool? Cities with storm sewers tend to dry themselves quicker than most places.
“They paved paradise and put in a parking lot.”
I do know that technically minded people working on farm scale irrigation systems will tell you that wet soil equals cool and dry soil equals hot.
In the 2000’s I was working in Crawley, Uk, the day the highest uk temperature record was set there. For those that don’t know Crawley is where Gatwick airport is. In the evening set off for my drive home around the m25 motorway, 6-10 lanes of carnage. Thermometer dropped from 32 to 30 to 27 deg C. Off the M25 up the M3 temperature dropped to 25. The m3 motorway in to london is bordered by rivers, lakes (old gravel pits) and reservoirs. Once into the suburbs where i lived temperature went up to 30C. That was 4 miles from heathrow where frequent annual records are set.
Thinking on the comments above about solar irradiance dwarfing human energy release and heat island effects, in an ideal stable climate wouldn’t energy emitted be less than incident energy?
From a thermodynamic perspective negative entropy should be observed by photosynthesis, other life forms and chemical and geological processes building higher order.
That would also suggest our global energy release is more significant than any forcing CO2 may cause.
It looks to me that there is an error regarding the area and population data of Los Angeles. You have an area of 10,780 km2 with a population of 3.884 million. The 10,780 km2 is very close to a Wikipedia value of 10,510 km2 that represents the land area of Los Angeles County. However, the population of Los Angeles County is 9.8 million (est 2018 ). The 3.884 million might be an older City of Los Angeles value as the 2018 amount is 3.99 million with an area of 1,210 km2. The official Greater Los Angles area consists of 5 counties. I wouldn’t use this grouping as it is made up of large relatively sparsely populated areas of mountains and deserts. I think the more appropriate grouping would be the Los Angeles Metropolitan Area as it consists of just 2 counties, Los Angeles and Orange that are mainly urban, adjacent to one another, and consist of 12,362 km2 and 13,310,447 people.
Willis, really enjoyed this article as it helped with a longstanding question I had about how much heat is produced by human civilization and how that compares to what we receive from the sun. I’m not there yet, but I think this helps validate that we are heating up locally, but not globally.
A couple of years ago, I was re-reading a science fiction novel that postulated an earth-like planet inhabited by trillions of human-sized beings. It further postulated that the heat generated by the beings, along with their industrial heat by-products was so great that they had to up move the planet away from the sun to maintain their lovely tropical climate. Since I first read this in maybe the 70’s, I wondered if it was possible for our leaking heat to cause this catastrophe.
With these numbers, it’s obvious that UHI contributes to local temperature reporting, which in turn inflates the models, which in turn inflates the rhetoric. I’m a bit disappointed, and correct me if I’m wrong, to see that the UHI contribution to heat, when spread across the entire planet (i.e. dive by planetary meters squared), is not significant.
On the other hand, if I extrapolate Seoul out to a 100x or 1000x globe-spanning megacity, the waste heat alone will make any GHG contribution meaningless.