A new paper showing how UHI and energy consumption are linked to long-term temperature change in China

From the paper – map of energy consumption in China

Dr. Roger Pielke Sr. highlighted this major new paper in JGR yesterday, I consider it important as it relates to the works I’m doing on station siting. The key points of the paper are supportive of what I’ve  been saying for years, and show a significant effect on temperature trends, just as we’ve noted in the discussion paper Watts et al 2012.

Key Points

  • Temperature trends are correlated to the intensity of human activity
  • Local warming is highly dependent on energy consumption pattern in urban areas
  • The local effects on temperature trends is 0.14 to 0.25°C/decade

An empirical study of the impact of human activity on long-term temperature change in China: A perspective from energy consumption

Yan Li, Key Laboratory for Earth Surface Processes, Peking University, Ministry of Education, Beijing, China, College of Urban and Environmental Sciences, Peking University, Beijing, China

Xinyi Zhao, Key Laboratory for Earth Surface Processes, Peking University, Ministry of Education, Beijing, China, College of Urban and Environmental Sciences, Peking University, Beijing, China

Abstract: (bold and paragraph formatting mine)

Human activity is an important contributor to local temperature change, especially in urban areas. Energy consumption is treated here as an index of the intensity of human induced local thermal forcing. The relationship between energy consumption and temperature change is analyzed in China by Observation Minus Reanalysis (OMR) method. Temperature trends for observation, reanalysis and OMR are estimated from meteorological records and 2 m-temperature from NCEP/NCAR Reanalysis 1 for the period 1979–2007.

A spatial mapping scheme based on the spatial and temporal relationship between energy consumption and Gross Domestic Production (GDP) is developed to derive the spatial distribution of energy consumption of China in 2003. A positive relationship between energy consumption and OMR trends is found in high and mid energy consumption region. OMR trends decline with the decreasing intensity of human activity from 0.20°C/decade in high energy consumption region to 0.13°C/decade in mid energy consumption region.

Forty-four stations in high energy consumption region that are exposed to the largest human impact are selected to investigate the impact of energy consumption spatial pattern on temperature change. Results show human impact on temperature trends is highly dependent on spatial pattern of energy consumption. OMR trends decline from energy consumption center to surrounding areas (0.26 to 0.04°C/decade) and get strengthened as the spatial extent of high energy consumption area expands (0.14 to 0.25°C/decade).

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 117, D17117, 12 PP., 2012
doi:10.1029/2012JD018132

One of the key passages of the paper Dr. Pielke highlighted was this:

Besides the impact of land use change on climate, the thermal impact induced by human activity within city plays significant role and should not be ignored. One of them is the anthropogenic heat released from energy consumption. Several studies have shown that anthropogenic heat is important to the development of UHI. Simulation results from a case study in Philadelphia suggested that anthropogenic heat contributes about 2~3C to the nighttime heat island in winter [Fan and Sailor, 2005].

Here’s one of the maps from the paper, showing the areas of highest energy consumption with meteorological station locations plotted on the map:

Here’s a plot of Nanjing’s (in the red zone on the coast) temperature from NASA GISS, note how the trend is basically flat until 1990. Obviously the station data didn’t make it to GISS for awhile after 1990, but when it resumed, what a difference there was:

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=205582380000&data_set=14&num_neighbors=1

Now look at Fuzhou, also in the coast, but not in the red zone of energy consumption:

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=205588470000&data_set=14&num_neighbors=1

In the conclusion:

Our results show significant warming has occurred for most stations in China and the magnitude of warming is closely related to energy consumption, which represents the intensity of human activity. For high and mid energy consumption group, OMR trends decline with the decrease of energy consumption. OMR trends for high and mid energy consumption group is 0.20 and 0.13C/decade respectively. Stronger warming is observed for station with high energy consumption, which usually locates in or near cities. Therefore, the strong warming is more likely a consequence of the local thermal forcing induced by human activity.

It seems that stations belong to high and mid energy consumption group in this study are affected
by human impact to a discernible extent. Just as De Laat[2008] demonstrated, anthropogenic heat released from energy consumption may very well have contributed to the observed temperature change patterns.Thus, it may raise more attention to consider the influence of human activity on surface temperature records in the past and next decades.

The bottom line – energy consumption matters, UHI matters, siting matters. Stations that are free of those issues should be the ones we base our land surface temperature record on. NCDC has already proven this point by commissioning the Climate Reference Network, guaranteed from its inception to be free of these issues.

About these ads
This entry was posted in GHCN, Temperature, UHI, Watts et al 2012, Weather_stations and tagged , , , , . Bookmark the permalink.

33 Responses to A new paper showing how UHI and energy consumption are linked to long-term temperature change in China

  1. Fan and Sailor is good work. Interesting that simulations are being cited.

    You really should read Sailors work… all of it.

  2. Jeff D says:

    Still anxiously awaiting Anthony’s paper but seeing another paper confirming this effect in a different approach really is heartening. No pressure Anthony but where is your paper!! Really though take all the time you need. We all know that every team member have their sites leveled right at you. The quality of your work must be impeccable because it will not receive the ” pal review ” many of the other papers we have seen.

  3. Anthony Watts says:

    Another crypto comment from Mosher with no citation or URL / DOI

    Sigh…

  4. Kev-in-Uk says:

    with respect to Anthony and others in this field – isn’t this all really about trying to ‘quantify’ an effect that we know for sure is happening? But using simulations and guesswork, reanalysis, etc, etc – is all a waste of time IMHO. What I mean is, why bother?, even warmists know that UHI is real (sure they try to hide it or make it look small) and in the end we should only rely on the good quality, well sited rural stations for any trend analysis.
    FFS, even UHI is variable in itself – a new motorway/underpass, towerblock, traffic lights, etc, etc will affect the ‘local’ temperatures to that ‘event’ – and I’m sure no-one (with half a brain) is going to go down that kind of analysis to look for possible effects of temp changes on already poorly sited stations!
    I’m just a little amazed at the ‘need’ for this kind of work? It harks back to trying to make the land temp record look ‘valid’ – which in terms of UHI, it clearly cannot ever be – putting lipstick on a pig comes to mind! Am I the only one who thinks this?

  5. Jolly farmer says:

    Yes, Anthony is both a target of the team’s sites, and in their sights.

    No worries, though. They can’t shoot straight.

  6. RHS says:

    Since most of our energy is created through release/transfer of heat (burning coal/natural gas), it makes sense energy consumption and heat island effect can be measured hand in hand.
    Looks like the fact we’re releasing plant food into the air, we’re raising temps rather than the plant food itself which we release is raising temps.
    Looks like less than 500 ppm doesn’t affect the remaining 999,500 ppm.

  7. wayne says:

    Anthony, a few comments if these key points are your own paper’s or addressing this paper if not:

    Key Points

    ▪ Temperature trends are correlated to the intensity of human activity

    Agree.

    ▪ Local warming is highly dependent on energy consumption pattern in urban areas

    ???. I really doubt it is ‘human consumption’. It is more like covering/removing vegetation to pave house slabs, driveways, streets, parking lots, that is removing evaporation sources area. Second, it is the darkening with asphalt, dark roofs, vertical walls of rock, brick, concrete, all dry, but ‘human consumption of energy’? I think that would be minimal unless we are speaking of the very frigid arctic regions where even a tiny bit of energy can make sizable changes in temperatures or maybe nighttime.

    ▪ The local effects on temperature trends is 0.14 to 0.25°C/decade

    Seems reasonable, but this would also hinge on the amount of expansion of urban area over a certain time period. For instance, London has been mostly developed for centuries so it shows limited relative change over time. The mid-America cities and many other places on this globe have all developed from almost nothing to huge cities all within just the last century – a huge UHI change.

    Increased energy consumption highly correlates with the amount of development of structures and pavement so it would be so easy to misapply the correct cause. Logically it is some combination of both but I have always felt the absorbed solar radiation trumps any human energy consumption.

    On the last two points, am I missing something here? I so agree with the basic thoughts, we all know UHI is a huge factor when speaking of linear cherry picked period trend lines, all warmists’ bread and butter defense.

  8. oldseadog says:

    Kev in UK:
    No.

  9. Henry Galt says:

    The Chinese have known this for some time, I hazard to guess. Same as India.

    They are very busy with other problems so have none of the ‘concerns’ weighing down our decadent Western politicians.

    Meting out punishment for our AGW? No thanks, we have other fish to fry. ;)

  10. Dan B says:

    It seems to me that this correlates human activity with localized rise in temps.
    I wonder what the political landscape would look like if instead of campaigns and taxes against evil carbon (dioxide) there was a more honest message of “devolve to save the earth”. Would so many people buy into that?

  11. JA says:

    So, city environments – concrete, asphalt, steel, etc., and little to no vegetation – creates a localized warm spot.
    Jeez !!!!!! Who would have thought that. Another great study to show what everybody already knows.
    Does anyone think that asphalt surfaces – whose surface temperature, esp. on a sunny day – can exceed ambient air temperatures by up to 40F to 50F degrees ( !!! ) – has anything to do with warmer city environments. How about concrete and steel as heat sinks.
    More great science from the world of the climate folks.
    Perhaps funding should be provided to see how using seat belts affects the severity of injuries in automobile accidents. Or maybe a study should be launched on the affect of eating 5 pounds/day of french fries (prepared in hydrogenated oil) on blood serum cholesterol levels. Or maybe a study on the affect on the human body of “base jumping” without a parachute.

  12. Steve says:

    Is this significant? UHI is an accepted phenomenon, and climate stations around the world claim to account for it. Additionally, the global rise in temperatures is confirmed by stations well outside of this effect.

    Excuse my ignorance, it appears the majority of posters here have a better handle of this subject, but does this paper not fall squarely into the neutral category along the ideological spectrum?

    Curious to understand where I may be going wrong here.

  13. Edim says:

    Kev-in-Uk, agree 100%. Absolutely the best stations should be used for the analysis, as far as possible from any human influence.

  14. george e smith says:

    Well urban heat islands should not be a problem in figuring out energy flows in the climate system. They are after all, presumably warmer than their surroundings, and hence should radiate/coduct/convect/whatever , more energy than the surrounding areas. The problem arises, when people believe it is ok to use that heat isand Temperature as a valid representation of some place 1200 km away.
    That’s why the first lecture in Climatism 101 should be the general theory of sampled data systems.

  15. Anymoose says:

    Urban Heat Island effect is a product of the mass of concrete, asphalt, steel and other man-made materials which is gathered into one spot. This is a product of urban sprawl. Urban sprawl is a product of population growth. Population growth means more energy usage.

    So, naturally there is a correlation between UHI and energy usage, but if you are doing the same study in the U.S., you could “prove” the same correlation between UHI and sales of toilet paper.

    This “study” belongs in the same trash bin as the one that proved that elimination of the locomotive and the caboose could drastically reduce train wreck deaths.

  16. Kev-in-Uk says:

    Steve says:
    October 2, 2012 at 10:24 am
    ..Is this significant? UHI is an accepted phenomenon, and climate stations around the world claim to account for it…
    picking up on your word ‘claim’ – that is the nub of the matter! claiming something and then demonstrating through correct science is another thing altogether!
    ..Additionally, the global rise in temperatures is confirmed by stations well outside of this effect…
    Hmm, undoubtedly some stations are indeed showing a temp rise; but then you have to check the reasons behind that and compare with other equally good stations and see if there are some local effects (especially sites near the sea for example), etc – and overall whether these are simply showing natural variation (rather than actual AGW/CC/CAGW or whatever) Again, the nub of the matter is that the rural trend (where seen) is much much less than the UHI affected trends – but which do you suppose are stated in the ‘headlines’ for AGW?

  17. As usual the Chinese have gone straight to the scientific source of the data and events and have not been deflected from logic by the ludicrous breast-beating concerns clogging up the thinking of our decadent political leaders. The Indians have also see through the foolish game of CAGW too. They are getting on with business and life.

  18. ferdberple says:

    According to the power company we use 1000 Kwh of electricity a month at our house, and 10 gigajoules of gas. This is a total of about 3778 Kwh of power per month.

    There are 720 hours in the month, the lot the house sits on is about 15M x 38M = 577 M2. So, doing the math, this works out to 3778Kwh/720/577M2 = 9.1 W/m2

    According to wikipedia, the forcing for a doubling of CO2 is ” In 2001 the IPCC adopted the revised value of 3.7 W/m2,”

    So, the heating of the property our house sits on is 9.1/3.7 = 2.5 times greater due to our energy use than that predicted for a doubling of CO2.

    This is worth repeating. The heating of the land on which our house stands is currently 2.5 times greater due to the use of energy at the house that is predicted for a doubling of CO2.

  19. TimTheToolMan says:

    Apparently policy warms more than UHI effects. Policy that min/max readings should be taken in the morning rather than the evening.

  20. ferdberple says:

    Steve says:
    October 2, 2012 at 10:24 am
    ..Is this significant? UHI is an accepted phenomenon, and climate stations around the world claim to account for it…
    ============
    Not correct. Phil “the dog ate my homework” Jones showed in his paper that UHI did not exist in China. BEST showed the same thing. Mainstream cliamte science holds that the effects of UHI are at best minor. Thus only CO2 can explain the increase in temps.

    As I have shown above, the heating at our property due to annual energy use significantly greater than that predicted for a doubling of CO2. That doesn’t include the 2kwh of fuel we burn in each vehicle each month, which heats the roads.

  21. Leo G says:

    Isn’t one of McKitricks papers along similiar lines? Using population density to show temp bias?

    Slowly the worm turns….

  22. I think this is one of those cases where correlation between A and B isn’t proof/evidence A causes B.

    Other studies show waste heat has a small effect on UHI, which is due to reduced evapotranspiration (humidity) and the Urban Canyon Effect, with some contribution from additional boundary layer mixing.

    Without knowing how they calculate energy consumption, I’ll assume it’s primarily electricity consumption. Electricity consumption replaced coal burning in stoves and hearths as China developed and urbanised, especially in the colder north.

    For this reason, electricity consumption in China is a good proxy for aerosol/particulate reductions, and this is the likely (IMO) cause of the warming, ie increased solar insolation from decreased aerosols and aerosol seeded clouds.

    Waste heat will have minimal spatial extent (try detecting the waste heat from your house 500 meters away). Whereas aerosols have well documented effects tens to even hundreds of kilometers from their source.

  23. Simon says:

    Why should all of the UHI be removed from the temperature record? It is a real increase in heat so it should be factored in. There is bias as urban areas are over-represented in the number of stations, but some of the UHI temperature increase should count in the temperature record.

  24. techgm says:

    Noteworthy: the map appears to imply that Tibet is not part of China.

  25. David Cage says:

    They knew for certain this was the case back in 2002 if not before, so I assume they will continue to keep the majority of the public unaware of this in the future just as successfully as they have done for the past decade. Facts count for nothing in the climate studies world.

  26. wayne says:

    Simon says:
    October 2, 2012 at 11:58 pm

    Why should all of the UHI be removed from the temperature record? It is a real increase in heat so it should be factored in.

    This is very picky subject to be sure! Could I take a very idealistic example to show you what I think is correct and how any adjustments should be made?

    Take 10,000 equally spaced points on this globe, 1000 of those coincide with population centers. the thermometers recorded to represent an average global temperature are all at the 1000 cities. If all cities have a population of one a century ago and if all 1000 said 25C the average would be 25C and correct. Now fast forward to today. Each city has one million and even though the 9000 say 25C the cities are now 27C.

    Ok, are the readings in the cities correct. Yes. Should they be adjusted? I say no. They are scientifically correct. Do these small warmer areas really affect the global average? YES! The average global temperature now says 27C with a 2C per century slope, all readings are in the cities. Is that the REAL average global temperature. NO! The real unrealized average is nine tenths 25C plus one tenth of 27C or 25.2C with a slope of 0.2C per century.

    That is the real situation. It is the calculation of the mean that is in error, not the individual station readings. Do you ever see this type of adjustments to the calculation being performed? NO! That is the problem.

  27. wayne says:

    Also Simon, I didn’t play in the area weighting for the small area occupied by just the cities and this magnifies that error even more.

  28. Crispin in Singapore's perpetual summer says:

    @Philip Bradley

    >Waste heat will have minimal spatial extent (try detecting the waste heat from your house 500 meters away). Whereas aerosols have well documented effects tens to even hundreds of kilometers from their source.

    Someone with knowledge about aerosols (condensed particles) and other (agglomerated or accumulated) particles is Dr Tami Bond (U of Illinois) who is quite an expert on the subject of black carbon and heating potential downwind. Without offending most people (unlike the majority of bloggers :) she has brought this carefully to the fore including testimony to Congress. Another contributor to the science is Prof Philip Hopke (Clarkson) who develops aethelometers for measuring BC nanoparticles, heretofore missing from the measurements because they do not interact with visible light. It may interest researchers to consider that as the UV radiation level increases during solar max (faculae and all that) these BC nanoparticles, collectively called ultrafine particulate matter (UFP), come into play. As the wave length descreases, the UFP’s play a more important direct atmospheric heating role. There are masses of UFP’s in the atmosphere resulting from burning any and everything. Bursts of EUV light up these particles, literally (if you could see them) and they happily float around the entire atmosphere.

    For those not following the above, the maximum size of particle can will react with visible light is about 0.1 microns and the efficiency of doing so is low (1-5%?). Thus black carbon (BC) is considered an important heating influence on the atmosphere but discussion to date has not recognized what were ‘invisible particles’. Organic Carbon (OC) is whitish and has a cooling effect. Both are created by burning biomass (badly), as in forest fires. Diesel engines, even ‘very clean ones’ produce large numbers of BC UFP’s and are rated ‘clean’ because the compliance equipment used does not detect nanoparticles. It borders on humorous.

  29. Kev-in-Uk says:

    Crispin in Singapore’s perpetual summer says:
    October 3, 2012 at 6:59 pm

    that’s interesting, but I’m not sure of the implications.
    so, in a nutshell, there seems some indication that BC and UFP’s are potentially ‘very warming’? Is it being suggested they are the ‘equivalent’ of a GHG as they are widespread in the atmosphere and are absorbing UV energy? If so, are these particles not doing something that would happen to the UV anyway (i.e. absorbing it) when it reaches the surface? It strikes me this could well account for atmospheric warming to some degree, whilst preventing energy reaching the surface and thereby causing ‘surface’ cooling?
    I’m a bit curious, as this also harks back to the ‘absorbing’ atmosphere issue again – whereby our amazing atmosphere has this ability to trap heat and never release it back to space (/sarc?).

  30. Brian H says:

    Human patterns of energy consumption have a large effect on surface temperature trends — as misrepresented by urbanized stations. The actual amount of energy introduced into the troposphere by said consumption is too trivial to create any real general effect.

  31. Crispin in Singapore's perpetual summer says:

    @Kev-in-Uk
    >that’s interesting, but I’m not sure of the implications.
    >so, in a nutshell, there seems some indication that BC and UFP’s are potentially ‘very warming’?

    It has been known for a long time that BC particles are very warming, but it was always played down (see early IPCC references to them falling to the ground early (within 10 days of source) and only having a ‘local effect’. Per kg (which is not a fair comparison) BC has about 640 times the heating effect of CO2 but recall my contribution above that it only picks up a wavelength that is related to its physical size.

    Consider: if there are no UFP’s there is no heating in the atmosphere at that point. If there are UFP’s and there is a high energy, short-enough-wavelength photon incoming it is going to capture it (because it is black). The interesting thing that happens is that the ozone which would have been created instead, is not.

    There are hints that the altitude at which ozone is created varies with solar activity, right? Could BC/OC be affecting this? It would not be for the first time that an element of misdirection was involved in our analyses. The total solar energy output doesn’t vary much, leading to thoughts that it ‘can’t matter’ yet the upper spectrum of that energy changes a lot. So it might matter, or matter a lot.

    >Is it being suggested they are the ‘equivalent’ of a GHG as they are widespread in the atmosphere and are absorbing UV energy?

    The effect per kg is far greater than a GHG or ice but the total mass is such that there is a reduced overall effect. It is presently assumed (hard to test etc) that is it about 20% of the GHG effect but I am not up to date. I believe that privately it is thought to be much higher but the ears cannot bear to hear the message. Suppose it was 40%. What are the implications? The main one is that CO2’s 0.8 per doubliing is perhaps 0.5 deg. To me, that is significant and worth investigating.

    >If so, are these particles not doing something that would happen to the UV anyway (i.e. absorbing it) when it reaches the surface?

    They are certainly interacting with suitably sized particles (anything above a certain size depending on their wavelength). The only thing I have seen referenced is ozone and UV, but BC and OC in the same ‘sphere of influence’ will be in the mix. OC will reflect it, for sure. Remember when conceptualising this that you cannot see these particles at all. They are transparent as far as our eyes are concerned.

    >It strikes me this could well account for atmospheric warming to some degree, whilst preventing energy reaching the surface and thereby causing ‘surface’ cooling?

    Good thoughts and worth considering as part of the whole. It is probably worth honing in on the UV because it is so variable and it is already known to affect ozone altitude of formation. It is now clear that BC and OC are floating around at nearly all altitudes (you can collect them at 40,000 ft sitting in your ‘steerage’ seat watching bad re-runs). They do not ‘fall to earth’ in the vicinity of where they are created.

    Consider: when they do eventually fall or grow in size and fall, they and maybe end up on ice somewhere enhancing the dirty snow effect. But the measurement of dirty snow is limited to large-ish particles so the effect of present/absent UFP is not perhaps considered. But there is less effect on the ground anyway. It is probably going to be the most interesting to find out what happens in the regions where ‘we want’ ozone to form.

  32. Kev-in-Uk says:

    Crispin in Singapore’s perpetual summer says:
    October 5, 2012 at 3:36 pm

    just seen your post….
    you say ”Consider: if there are no UFP’s there is no heating in the atmosphere at that point. If there are UFP’s and there is a high energy, short-enough-wavelength photon incoming it is going to capture it (because it is black). The interesting thing that happens is that the ozone which would have been created instead, is not.”
    so you are suggesting that UFP’s may prevent ozone formation by ‘capturing’ uv that would have otherwise formed said ozone? as for the UFP’s actually capturing UV, you’ve already hinted it will be wavelength dependent? Is ozone forming UV the same wavelength as that potentially captured by UFP’s?
    As a slight aside, surely most atmospheric molecules can technically capture UV? I’m not sure I buy the ‘it has to be black’ suggestion! My physics is old and rusty, but I would have thought the capturing issue is more due to the particle size and molecular constituents rather than its colour! IIRC correctly, the photon capture (whether UV or whatever) is essentially due to the wavelength and receptiveness or ‘need’ to a photon of the molecule itself – for example, is it +ve or -ve charged, etc, etc…..but I have no intention of rereading my old physics stuff as I’ve no doubt its out of date (like me!).
    I think the dirty snow thing is completely different – being that this is mostly absorbing IR – or have I got that wrong?
    I could also query whether these UFP’s are actually persistent in the atmosphere – meaning that, unless they are essentially nonvalent, inert type molecules, they could presumably react with something else? (even if it was simply being absorbed into/onto water droplets!) That’s enough brain/memory thoughts for a saturday night….ouch……
    regards
    Kev

Comments are closed.