Urban Biases on Surface Temperature Records

I apologize to readers who visit both of my sites for the repetition between them of late, but there is a lot of demand in the community of folks who usually don’t come to climate sites for climate analysis, so I am repeating stuff from here at my other blog.

A kid and his dad manage to do the analysis that NASA, the EPA, the CRU, and the IPCC can’t be convinced to perform. Awesome.

  • Debbie

    Parker, D.E., Large-Scale Warming is not Urban, Nature 432, 290, doi:10.1038/432290a, 2004.

    Peterson, T.C., Assessment of urban versus rural in situ surface temperatures in the contiguous United States: No difference found, Journal of Climate, 16, 2941-2959, 2003.

  • Mark

    @ Debbie

    “Large-Scale Warming is not Urban”

    After adjustments were applied to the data!

  • Excellent video, simply excellent.

    You know, I went and looked at the temperature data for Raleigh, NC, my home town, for the entire year through today earlier, and Surprise! The majority of the time the highs were at or below average. One interesting point was that the lows were at or above average. Could there be less radiational cooling?

  • ADiff

    Debbie, look up Matthew (ch. XV, v. 14)…

    The impact of urbanization on temperature measurement is, ahem…”settled science” (at least as much as anything).

  • Roberto

    Superb choice of music, too. Aptly, it’s “Have You Heard?” by Pat Metheny. NASA, CRU, etc., can’t because their fingers are in their ears and they are humming.

  • ecology student

    Wow, this blog is oozing with sanity! I’ve half-way convinced my ecology teacher that “a skeptics view of anthropogenic global warming” would be a good topic for my 15 minute end of term presentation (next term).

    He’s naturally reluctant, being a low-level warming priest, and wants me to come up with a preliminary bibliography of recent (2007 to 2009) peer-reviewed articles by scientists that support a skeptical view. I’m not finding too many that recent. Can anyone help?

    Also, the focus has not been developed yet, but what skeptic subtopics can you guys recommend that have the most solid (and recent) backing by scientists, that would punch the biggest holes in the AGW belief system?


  • John Anderson

    I was sure this was you and your son. It’s encouraging to see it wasn’t; more people are learning for themselves.

  • Fred

    This doesn’t change anything. Based on these figures the average temperature in Manhattan will hit 212 F by the year 2509. We are doomed. Doomed I say!

  • Keith Hogan

    ecology student, try this for a list of peer-reviewed papers:

    My suggestion would be to focus on two main points:

    1. Near term historical temperature increases may be partially driven anthropogenically, but they are also not outside the realm of naturally occuring climate fluctuations. The MWP and LIA were real events. Loehle published a nice paper showing a climate proxy reconstruction WITHOUT using tree ring data which clearly shows this. My personal favorite way to show at least the LIA is this chart from the US Geological Survey which shows how the Glacier Bay glacial terminus points have been moving since the end of the Little Ice Age in the late 1700s — long before any widespread use of fossil fuels:

    2. There is really very little scientific evidence that current CO2 levels, and any associated warming, will be in any way catastrophic. Temperatures for the entire 30 year satellite record have been increasing at roughly 0.13 – 0.16 deg C per decade. Even if this continued for the next century, this would mean less than 2 deg C of warming. Study things like Beer’s Law, and you will see that, since the absorption of infrared radiation by CO2 increases more and more slowly with increasing concentration, the less than 2 deg C of warming is likely a top end of the realistic temperature increase due to CO2 increases.

    Good luck!

  • brick

    @ ecology student:

    try this one, 450 reports

  • Stewart

    @ ecology student:

    here’s a few more recent ones:
    Anastasios, et. al., A new dynamical mechanism for major climate shifts Geophysical Research Letters(2007)
    Christy, et. al., Surface temperature variations in East Africa and possible causes, Journal of Climate (2009)
    Lindzen and Choi, On the determination of climate feedbacks from ERBE data, Geophysical Research Letters(2009)
    Christy, et. al., A comparison of tropical temperature trends with model predictions International Journal of Climatology (2007)
    Loehle, A 2000-year global temperature reconstruction based on non-treering proxies Energy & Environment(2007)

  • kdk33

    Which data set did these guys download. GISS offers 3: raw, combined, and homogenized – each with varying degress of correction.

    If they video uses raw data, then the counter would be that GISS ‘corrects’ the data – in the homogenized version – so the UHI effect is eventually subtracted.

  • Stonyground

    If temperature data for the last century is available from both rural and urban weather stations, why would anyone attempting to track changes in average temperature want to include the urban ones at all? The urban heat island effect is difficult to quantify and correct for, so surely the obvious route would be to rely mainly on the data from the rural stations where these unwanted effects will be minimal or non-existent.

    It is difficult to avoid the impression that someone is trying to pull the wool over our eyes.

  • joshv

    @Stonyground – exactly, I’ve never been able to get a satisfactory explanation as to why we should include data that we know has a warming bias.

  • Ike


    But you HAVE to include the data from the cities, because more people are living there than before!

    In fact, we ought to give the data from cities greater weight in the calculations, to reflect more people are living in them!

  • Kent

    Even Jim Hansen recognized this problem from the beginning. The question has always been to what “degree” is the data skewed by the urban heat island effect? Hansen says the answer is “very little.” Others argue for much greater warming from the urban bias. It is disturbing how little we have progressed on this issue in the past 20+ years.

    From Hansen & Lebedeff (1987):
    “Removal of the city data reduced the magnitude of the global and hemispheric warmings, as illustrated in Figure 13. For example, the global temperature change in the past century was reduced from 0.7 to 0.6 C, where these numbers represent the difference between the mean 1980-1985 temperature and the mean1880-1885 temperature. We subjectively estimate that complete correction for urban heat island effects should not reduce the global warming in the past century, defined as the temperature difference between 1980-1985 and 1880-1885, to less than about 0.5 C.

    As mentioned already, the nature of the observed temperature trends, especially the geographical distribution of the warming, also provides strong evidence that the global temperature change is not a figment due to urban heat island effects. This evidence and the quantitative test just described lead us to conclude that the global warming of the past century is a real climate trend, even though it does contain a significant contribution due to urban heat island effects. More detailed and comprehensive studies of urban influence are warranted; perhaps the dataset we have developed, including the seasonal variation of trends, can contribute to such studies.”

  • Pete


    I think you are confusing our discussion on climate with your discussion on urbanization, we were talking about the surface temp NOT population temp.

  • Chris Hull

    @Ike said: ‘But you HAVE to include the data from the cities, because more people are living there than before!’

    That is the best satire I have seen in 20 years or you have to get back on your meds.

  • W Frawley

    Analyzing the difference of the data pairs (urban-rural) would serve to illustrate the contrast even more.

  • ADiff

    Ike, climate change really doesn’t make any sense at all on a per capita basis. Your argument’s senseless in this context. If the issue where ‘perceived’ change or something along those lines, perhaps…just possibly. But as long as the issue is over actual physical change and accurate measurement of it, then population locale is completely irrelevant, in itself.

    Here’s what one commentator had to say recently about Copenhagen, that I found interesting:

    “The representatives of the current ‘established elites’ just want to try to make change stop, to try to secure their position, especially from all those unmanaged innovative would-be competitors who they otherwise would either have to successfully compete with, or simply lose that current status. The only change or innovation they like is ‘managed’ change and ‘planned’ innovation (how’s that for oxymoron?) that doesn’t threaten their status quo. If they can manage to steal from those same unruly, threatening would-be competitors, more the better, even if they have to share the loot with the unwashed masses of the world, in order to secure their support for these efforts.

    The representatives of the ‘have-nots’ just want any excuse to loot the rewards of accomplishment to try to equalize their outcomes irrespective competitive or productive performance. They’re perfectly happy to ‘cut a deal’ with the previously mentioned crowd, if they think they can expropriate the wealth created by the current and future generation of wealth creators, suiting both their demands and those of the established elites at the same time.

    These elites don’t seem to realize Palace Economies always end in burnt rubble, and the ‘have-nots’ (or more accurately the thugs that purport to ‘lead’ them) don’t care, as loot’s all they know, and it doesn’t matter much to them whether it’s a pay-off or out-and-out rapine. Or perhaps, as suggested by the forwarded WUWT article [referencing the post topic of this blog entry], their temporal threshold is too myopic for that to matter. Those who can so blithely ignore any geological record outside their own life-times might very well be suspected of a similar (and similarly convenient) ignorance of History.

    I’ve always been keenly aware of the limitations of the depiction of market reality in Ayn Rand’s fiction, but Copenhagen makes me almost think all those involved have taken her characterizations to heart, Ellsworth Toohey for the ‘Green’ crowd and Kip Chalmers for the politicians…. Their thinking, actions and attitudes are so patent, transparent and stereotypical it’s hard not to see them as caricature.”

    That’s very harsh, but it certainly seems to ring true!

  • MikeC

    The commenter that pointed outthat there are different sets of data (ie: raw and adjusted) is correct. The father son team here did not use the adjusted data, they used the raw data. The GISS UHI adjustment is better than most but it still lacks the sensativity to do the job by about 3 tenths, or half the alleged AGW signal. When you add in other problems with this very un-studied areas of science then you get the whole picture.

  • kuhnkat

    ecology student,

    drop by http://www.co2science.org and check their current posts and archives. They also have a DB of paleo papers for the MWP.

  • kuhnkat


    You are correct. Use ALL of the data. Unfortunately, if the data you have covers 10% of the area you are sampling and is 75% urban you have a problem.

    Tell me, how would you deal with that issue??

  • kuhnkat


    the GISS UHI adjustment AVERAGES between an alledged rural and a urban station. It might knock the top off of a BIGH UHI, BUT, it warms the rurals station, which probably isn’t really rural anyway.

    Drop by http://chiefio.wordpress.com/ and see what E.M. Smith has been doing in dissecting GISSTemp. The latest problem is lack of stations!!

  • BobbyMcP

    Fantastic! I downloaded to show my grand children. This should be a published paper.

  • Richard

    Very good science for “amateurs”! As they say in physics, “It’s all in the numbers.”

    I did notice, though, what appeared to be a very, very slight rise in the rural average temperature line over 111 years. Not much– hard to eyeball with the screen image– but maybe 0.1 deg or less. Certainly not enough to crank up the sirens, but on a geological scale, this could be significant because the earth’s climate goes through wide and slow-paced fluctuations in average temperature. To paraphrase a political consultant from 1992, “It’s the geology, stupid.” Not the humans.

  • Peter’s dad

    Thank you for all your comments, Peter and I really appreciate it. Just so you know, we did not use the raw data. The data was the “combined data at the same location.” We just learned ,however, that the data we downloaded in 2008 has been changed at GISS and the rural numbers have been recently modified. Curious isn’t it? It seems that the rural sites have suddenly become warmer. It really should not change the overall pattern but we will check. Also, you should not be too concerned about the trendline equations. The r2 values are very low which indicates a high variability in the data. You really can’t make an argument about how much of any increase or decrease there is in the slope. It is not statistically valid. It is similar to arguing over how many angels can dance on the head of a pin. Thank you again.

  • Richard Saumarez

    During the summer, I regularly drove from my home (rural) into medium sized cities. As there is an air thermometer in my car I, out of interest, recorded the temperature on driving in to the city and the temperatures in the city and on returning home. This was to resolve a dispute with my wife, who thought that towns were hotter than the countryside and I didn’t. The mean differences (+-2Sd), despite doing trips at different times of day for 40 observations were:

    Home to Cambridge +2.1 +-0.3 (20 minutes)
    Home to Huntingdon +1.8 +-0.4 (30 minutes)
    Home – city – home +0.3 +-0.25

    I thought that was interesting, but really quite trivial, so I didn’t do anything more complicated as regards stratifying the data into time of day. It seems a very simple experiment to have vehicles with suitable instrumentation to traverse cities at set times of day and nail this down.

  • Sedona Mike

    Where do you start. Raw data is raw data. Massaging it to get a result that you like is hardly scientific. There has to be a little logic here, like use the brain god gave you when looking at all of this. I live in Sedona, AZ, and spend time in the Phoenix area. When it gets to 115 in the day, and does not get below 90 at night, but in the desert, just out of town, where it is 10 degrees or more cooler, you have to wonder why. For the same reason that computer processors have massive heat sinks on them with fans, that is where the heat is stored. Phoenix and all cities to various degrees are heat sinks. The buildings, concrete, asphalt and lack of ways to dissipate the heat of the SUN create temperatures that are higher. Any anyone who thinks that cities are the cause of global warming (the earth has not warmed in 10 years but has gotten cooler) needs to get in an airplane and look down. You can by the way, as I did the time to do the calculation, put the entire planet’s population, assuming a family of 5, in 950 square foot houses, and still have room for shops, roads and schools etc., in the state of TEXAS, and use the rest of the earth to grow food and provide natural resources. For 200 years the Vikings raised crops in Greenland, the glaciers that are receding in Europe are uncovering past civilizations. The ice on the polar caps is not melting, and you to can monitor it! http://nsidc.org/data/seaice_index/daily.html Core drilling in the Amazon, where areas have been untouched for millions of years have shown that the earth goes through 36,000 year cycles, where the oceans rise and fall 360 feet. So 18,000 years, up 180’, 18,000 years, down 1`80 feet. And we think WE can do something about that? We are in the upswing now. So if you have property that is less than 180 above sea level, better put up a dike because 18,000 years from now you will be under water. In all these alarmists predictions, why is the sun ignored? It is the source of ALL our energy. Does anyone find that strange? Are all these people wrong? http://www.petitionproject.org/ . OK, I could go on, and discuss the political and financial interests that are pushing this, but there is not enough space here. You must LOOK at WHO is funding the alarmists. Will they have a job if they have findings AGAINST global warming? Have you looked at scientists who have LOST their funding and are now stating that they “created” findings in order to keep their funding coming in? It is very unfortunate that science can now be bought. More CO2 means more fuels for plants. If the planet gets a little warmer, is that a bad thing? Every time you exhale you create CO2, and it is a poison gas? Without CO2, the planet is dead. And water vapor is a greenhouse gas, so do we ban clouds also? The earth is 4 BILLION years old, and we think we have the ability to affect it? Who do we think we are? With 100 or so years of data? As compared to 4 Billion? Where is THAT logic coming from? One volcano eruption puts as much CO2 in the air that human activity does in a YEAR. We are a part of the planet. DO we want clean air, clean water, and a cleaner environment? Sure, no one argues with that. But attacking a building block of the planet? I’m sorry Al Gore, but hopefully you cannot make another $100 million on our backs by destroying our economy. His latest book cover, 4 hurricanes in parts of the world where they are impossible, Florida all but gone, and Cuba gone (which would take a sea level rise of 6,578 feet). Denver – kiss your butt goodbye. Do research yourself, but LOOK AT THE MONEY TRAIL AND THE POLITICS.

  • MikeC

    Hi Peters dad, Its great to see ya getting your kid into something educational. The data you used, “after combining at the same location” is still unadjusted data. All they have done here is combine data from different sources, whether its a station that has moved several miles over the years or using data from different stations in the same immediate area (i believe the rule is 5 miles).
    You want to use the data that is after homoginization. This is the final data after Hansen applies his night lights (urbanization) adjustment. You will notice that the numbers tighten up a bit. If you go further into the science you will notice that the whole area of climate monitoring is poorly researched and that many of the adjustments that need to be made are not. Additionally, the statistical methods they use to adjust temperatures, particularly those based on the data from surrounding stations, is not sensative enough.

  • Bob

    _global_ warming IS real and settled science … but only available in _few_ places 😉

  • orthodoc

    This is terrific work. Peter and his dad should be proud.

    Richard, there is a very small trend upward (check the regression line’s slope on the slides) but this would not be surprising, since the earth did warm slightly over the time frame, and the r-squared values reflect that.

  • orthodoc

    This is terrific work. Peter and his dad should be proud.

    Richard, there is a very small trend upward (check the regression line’s slope on the slides) but this would not be surprising, since the earth did warm slightly over the time frame, and the r-squared values reflect that.

  • Ike

    I apologize for not using the Sarcasm tag in my previous post.

    I was in no way being serious, and am a little disappointed that no AGW supporters picked up the bait.

  • Ike

    @Chris Hull – thanks. I was rather proud of its power and brevity, only matched by its brazen inanity.

  • Dave Dardinger

    One point which might be made is that there’s a major advantage to using raw data and paired urban/rural stations. Much of the problem with the adjustments is that they’re not generally detailed or justified. But if you take raw data and pair it, it’s unlikely that even if there is legitimate correction like TOBS, the difference between the paired stations would not cancel this out. (sure an individual pair might have the urban station change first, but eventually the rural station would change too. So at worst you’d have a concave or convex portion but the overall results wouldn’t be affected. Seeing as large a difference between urban and rural stations as is shown here, even though the rural stations would probably have had some urbanization, means we’re looking at the minimum UHI, not the maximum.

    If I were a warmer, I’d probably try claiming cherry picking of the pairs. But I say fine, find all the pairs you can and lets look at what the raw data says. IMO this is a much more honest way of looking at/for UHI than looking at satellite night photos or low speed winds.

  • @Ecology Student – Good for you. For the past several years, we’ve been doing an AGW debate in my honors section of introductory biology. Sometimes the consensi win; sometimes, the skeptics win. But everyone, myself included, learns a lot. And it reinforces two of the things I try to emphasize.

    1. Science evolves. Every paper that’s ever been published is wrong. The only things in question are how wrong and how long it takes for the errors to be detected.

    2. When ideology gets mixed in, both the errors and the time to detect them are invariably magnified.

    Best wishes.

  • julieg

    Congratulations Peter and Dad.

    I only wish that I had any mathematical ability at all as I would love to try something similar in my area (Sydney, Australia).