New Paper “Climatic Trends In Major U.S. Urban Areas, 1950–2009″ By Mishra and Lettenmaier
By Dr. Roger Pielke Senior
There is a new paper
Mishra, V., and D. P. Lettenmaier (2011), Climatic trends in major U.S. urban areas, 1950–2009, Geophys. Res. Lett., 38, L16401, doi:10.1029/ 2011GL048255
which reports on the effect of urban areas on multi-decadal surface temperature trends.
The abstract reads [highlight added]
We evaluate changes in climatic indices for the 100 largest U.S. urban areas and paired surrounding non‐urban areas. During the period 1950–2009, we find that there were statistically significant changes in as many as half of the urban areas in temperature‐related indices, such as heating and cooling degree‐days and number of warm and cool nights, almost all of which are reflective of a general warming.
Similarly, statistically significant changes (mostly increases) in indices related to extreme precipitation, such as daily maximum intensities and number of days with heavy precipitation, were detected in as many of 30% of the urban areas. A paired analysis of urban and surrounding non‐urban areas suggests that most temperature‐related trends are attributable to regional climate change, rather than to local effects of urbanization, although the picture is more mixed for precipitation.
Among the conclusions
Consistent with previous studies [Easterling et al., 2000; Kalnay and Cai, 2003], trends related to temperature minima in the urban areas are generally stronger than those related to temperature maxima.
For both minimum daily temperature based climate indices and precipitation‐related trends, changes in urban and non‐urban areas are generally consistent; suggesting that the trends are dominantly a response to climate [Parker, 2004; Peterson, 2003], rather than local land cover changes during the period of analysis. However, there is somewhat less consistency in urban vs. non‐urban trends in climate indices related to daily maximum temperature, which suggests that land cover change may be at least partially responsible for those trends.
An important caveat to their study is that they have not factored in the role of microclimate changes at the observing sites which we have started to explore, as reported on in our paper
Fall, S., A. Watts, J. Nielsen-Gammon, E. Jones, D. Niyogi, J. Christy, and R.A. Pielke Sr., 2011: Analysis of the impacts of station exposure on the U.S. Historical Climatology Network temperatures and temperature trends. J. Geophys. Res., 116, D14120, doi:10.1029/2010JD015146.Copyright (2011) American Geophysical Union.
Their finding of less of an effect on minimum temperature trends on whether they are located in urban or rural areas is, however, puzzling, as the urban heat island effect on minimum temperatures is very well know (e.g. see EPA heat island effect). Since the spatial scale, density of build-up and type of constructions on urban areas continues to change over the time, the failure to find a difference between rural and urban areas needs more investigation as to why this was found in the Mishra and Lettenmaier analysis.
I recall someone leaving this youtube video at this site. The video is about 3 years-old. It is worth viewing again
http://www.youtube.com/watch?v=LcsvaCPYgcI&feature=player_embedded
Congrats Anthony, Roger, John’s and others!
Is this paper complimentary to Menne et al 09?
constructoins
typo
Edit notes:
“is very well
know” known“build-up and type of
constructoins” construction“as to
whay” whyIt will be interesting to compare these results to those of the Berkeley Earth Surface Temperature project. They’re also trying to determine the effects of the UHI. There was mention at Climate Etc. that papers were close to being submitted, and the work will also be presented at the AGU. The abstracts don’t suggest large differences between their estimates and those of other groups. Some snippets:
ABSTRACT FINAL ID: GC43B-0908
”Applying the Berkeley Earth techniques, we broadly confirm the temperature histories presented by prior groups. However, the improved methodology allows the uncertainties to be reduced (often by 50%) and also has allowed the instrumental temperature record to be extended back to 1800.”
ABSTRACT FINAL ID: GC44B-01
”The results we obtain are compared to those published by the groups at NOAA, NASA-GISS, and Hadley-CRU in the UK.”
These sorts of studies seem to all be using already available data, simply massaging or modeling such data in different ways.
I really am only interested in doing actual experiments, you know, actual science.
The method would be simple, you need, say a camper, and some portable climate measuring equipment (or even just a calibrated temperature measurement device of the same type used in official stations). Then, drive out from the city you are comparing urban versus rural temperature at, find a suitable site in a rural area as close to said urban city as possible, take measurements for say 24 hours (hence why you need the camper), then compare those to the official measurements in the urban environment.
I am willing to bet that there will be a noticeable and significantly lower temperature overall in the portable rural site compared to the nearby urban site. Any other method other than the above is just shuffling data around and is not an actual scientific experiment or using the actual scientific method which could be said to be “you say you believe X, SHOW ME”.
They just talk about the increases in the urban setting and are silent on changes (if any) at the rural pairings – at least that is what I’m reading in this summary. Am I wrong?
last paragraph, first sentence, fourth word. “less” rather than “leass”?
[Fixed. Thank you, Robt.]
Ray Spencer is vindicated again.
Wear your scars with pride Dr Spencer.
Note also in AGU abstract GC44B-01 that Anthony Watts is acknowledged specifically!
“…We calculate the effect of poor station quality, as documented in the US by the team led by Anthony Watts by estimating the temperature trends based solely on the stations ranked good (1,2 or 1,2,3 in the NOAA ranking scheme). “
I find the wording somewhat confusing frankly.
1. “…there were statistically significant changes in as many as half of the urban areas in temperature‐related indices…”
OK. So what about the other half? And what does “as many as” mean? 40% I understand, 1/2 I understand, but how many does “as many as half” actually mean?
2. “… trends related to temperature minima in the urban areas are generally stronger than those related to temperature maxima…”
OK…was that true for non urban areas as well?
3. “…For both minimum daily temperature based climate indices and precipitation‐related trends, changes in urban and non‐urban areas are generally consistent…”
OK…so it sounds like urban and non urban were about the same then?
4. “However, there is somewhat less consistency in urban vs. non‐urban trends in climate indices related to daily maximum temperature, which suggests that land cover change may be at least partially responsible for those trends.”
Well that one at least makes sense. Maximum temps occur in day time when things like tar and pavement and on and on are absorbing heat from the sun that would otherwise be reflected. But at night, cooling prevails as the dominant process which is going to be predicated on how much heat built up during the day, and how easy it is for that heat to escape. So even with NO effect from GHG’s, I would expect daily minimums to increase more in urban areas than in non urban.
But without knowing exactly what was meant by “as many as half” and what the results were for the OTHER half, not to mention strength of prevailing winds which would move the warm air from the urban to the downwind non urban… which would futz up oh….given the variability of prevailing winds and the likelihood that the rural pairs were downwind…about 50%? As many as half I mean?
Nothing too surprising here. if you select 100 major urban areas in the US and compare them to their rural partners you will find differences. Heck a random blogger did this over a year ago using all of GHCN not just the US
http://residualanalysis.blogspot.com/2010/04/urban-heat-island-effect-model.html
the cutoff he found was a population of 1 million. making a cutoff on raw population size is probably not the best way to do things as it doesnt take area into account, so its probably better to look at population density. density drives the things that drive UHI. It will be interesting to see which 100 cities they used.
Its also import to note ( after Imhoff and Oke ) that the characteristics of the rural environment are just as important as the characteristics of the urban environment.
the hundreds of thousands of tons of asphalt, concrete, sidewalks, and building surface area per square mile of a city absorbs vast amounts of heat that simply must skew the urban temperature record. as air moves over and beyond urban area, it carries that heat elsewhere. i’d be willing to bet that rural areas in the prevailing downwind path of a large urban area also show a similar warming trend. the null hypothesis, i think, would be to compare the temperature trends when the wind is blowing the opposite direction.
it’s fascinating to see a weather front move into the dc area on the local weather radar and watch the storm immediately and severely intensify as soon as it passes into the much warmer urban areas.
thumbnail calculations of the btu storage capacity of a dense, urban fabric indicate staggering quantities of heat storage capacity. rural areas don’t have this component, and it simply has to make a measurable difference.
thanks for all your hard work, anthony – love coming here for a good read.
Their finding of leass of an effect on minimum temperature trends on whether they are located in urban or rural areas is, however, puzzling,
I think that needs to be reworded as its meaning isn’t clear.
If it means they found less of an effect (ie increase) in minimum temperatures in rural areas then that doesn’t surprise me as I believe much of the change in minimum temperatures is due to reductions in near horizon particulate pollution (smoke and haze) causing earlier (in the day) and hence higher minimums.
There is a clear seasonality difference between the sources of particulate pollution in urban and rural areas. In the 1950s and 60s smoke haze from fires would be predominantly in the winter in urban areas, but predominantly in the late summer in rural areas from crop residue burning.
There is also a regional difference as domestic coal fires were replaced earlier in the USA than in Europe. And I would expect a larger increase in min temps in Europe compared to the USA since 1950.
Changes in ocean surface temperatures are the principal cause of climate change. In many regions of the world, there is a distinct ‘ocean signature’ in the weather station record. In California it is the Pacific Decadal Oscillation (PDO). In the UK/W. Europe it is the Atlantic Multidecadal Oscillation (AMO). The minimum daily air temperature is basically a measure of the bulk air temperature of the local weather system as it is passing through. The daily maximum temperature is a measure of the convective coupling of the solar surface heating to the air temperature. By comparing the long term (5 year) average of the ocean temperature (PDO, AMO etc.) to the average of the weather station minimum temperature over the same period of record, an estimate of the urban heat island effect can be obtained. This can be done just using a linear fit to the data, although care is needed and there can be other sources of bias in the data. The difference in slope between the ocean data and the weather station min data provides an approximate measure of the UHI effect.
The technique has provided useful UHI data for California and UK weather stations.
Further details can be found at:
http://scienceandpublicpolicy.org/originals/pacific_decadal.html
http://hidethedecline.eu/pages/posts/what-surface-temperature-is-your-model-really-predicting-190.php
Steven Mosher,
I’m thinking of looking in greater detail at all of the sites I sent you from OZ that I had arbitrarily characterised as ‘pristine’. Then it occurred to me that you might have done this in the wider BEST calculations?
Have you, or will I?
It’s handy to have a set of ‘pristine’ stations covering a large land area, where the chances of effects from the hand of man are credibly small. Sets a baseline of sorts. Lets one look at relative trends of Tmax and Tmin. Would suggest this because a preliminary look was fruitful, but personal computing power is small.
This seems to part of a pattern to downgrade the importance of urban size and morphology in urban temperature measurement, and example of this is this attempt to quantify the “weekend effect” by our old friends here. “Trend” is mentioned 6 times in the above abstract and in the title.
I can’t see what the big deal is here, temperature ‘trends’ in rural and non-rural area should follow a similar pattern, be it up or down, we are all part of land based micro-climates.
Trend is a bit of a broad brush, I take it to mean the continuation of a pattern, are they really talking about correlation?
Seems like a justification to continue urban temperature measurement and ignore the critics,
.
At some point there will be a collation of all these different differences between the IPCC CAGW story and the Real Climate story, I expect …
Clouds. Cosmic rays. UHIE. Satellite/sea-level. Deep-sea temperatures/ocean heat. Storm frequency.
To name just a few.
Paul Irwin says:
the hundreds of thousands of tons of asphalt, concrete, sidewalks, and building surface area per square mile of a city absorbs vast amounts of heat
That’s before you even consider that cities tend to also be full of artificial heat sources
The woman in the picture is a climate expert – she is outstanding in her field.
Seems this woman has many sheep to herd 😉 – yawn ZZZzz…
A bit of common sense is called for here. If I was to carry out temperature readings to prove or disprove global warming, the last place I would take measurements is in a city. The buildings, asphalt, cars etc will absorb more radiative solar heat during sunny days and then radiate it at night. This will not happen in rural areas, simply because grass, trees and shrubs could not heat up to the same degree because their biochemical processes keeping them alive would cease. In addition, in summer, heat from air conditioning units would raise the ambient temperature, likewise with central heating in winter. Cars, buses, lorries and to a lesser extent humans will also raise the ambient temperature.
If this pitiful excuse for a study demonstrates “scientific” reasoning, God help us all.
The comparison should not be rural vs urban 1950 to now. It should be *rural in 1950, urban now* vs. *rural 1950, still rural now*. I would expect the trends in places whose land use hadn’t changed over the last 60 years to be the same, whatever they were over the period.
Most UHI will take place as rural sites *begin* to develop. Stable urban will not show much additional warming.
Also, “rural” – what does that mean? Primeval forest is rural, so is open farmland. But the latter has higher daily mean temps and lower night time temps.
Did they archive their sites?
davidmhoffer, I agree with you, the language is not clear and leaves room for interpretation.
I have noticed more and more of this sort of language appearing in papers as the evidence for catastrophic human induced warming fails to make an appearance. We see papers which present evidence in such a way that the language used to describe it can mean “we looked for evidence of increasing man made warming but found diddly squat” but also allows the researcher to dispute that and claim, “that is not what we meant and we support the consensus” when they need more funding.
From Paul Irwin on September 21, 2011 at 9:13 pm:
Heck, just a mere hundred pounds or so of concrete or asphalt can affect the temperature record, when located right underneath the temperature measuring device. 😉
Australia has temperature data for many stations taken at fixed times.
Jonathan Lowe has done a detailed analysis of this time based data going back 60 years.
In summary, there is no warming trend overnight (midnight, 3 am, 6am) but then a sudden warming trend at 9am, which is especially large in winter.
The supposed warmer nights due to AGW simply don’t exist. At least in the Australian data.
The increasing Tmin is due to something that occurs in the early morning. And that something is likely a combination of decreased particulates and decreased clouds causing increased solar insolation resulting in earlier and higher Tmins, and higher temperatures at 9am.
This data also eliminates UHI as a cause of increasing Tmin, because UHI doesn’t suddenly start in the early morning.
http://gustofhotair.blogspot.com/
I am sorry to have to say this but:
“During the period 1950–2009, we find that there were statistically significant changes in as many as half of the urban areas in temperature‐related indices, such as heating and cooling degree‐days and number of warm and cool nights, almost all of which are reflective of a general warming.” is complete bo***cks.
What is wrong with you “scientists?” Statistically significant in as many as half also means no statistical significance in as many as half. It does not take a McIntyre to work out from this mushed statement that they found nothing but are blaming the nothing on regional climate change. Then they have the gall to start on 30% for statistical significance for precipitation. i.e. 70% discarded. The whole bl**dy lot of you should be sent back to primary school for the starter course on statistics. Mann, Dessler, et Al, now this lot. When will you learn that if there is no statistical result you pack it in and say “Oh well never mind, lets try something else.” Not publish mush and draw conclusions which are just not there. You are wasting my valuable time that could be spent chopping wood or de-rusting the gate and oiling the mailbox hinges. Rant over.
PS Thankyou, Anthony, for drawing all this to our attention, but really, somewhere out there there has to be someone doing some useful science.
“Their finding of less of an effect on minimum temperature trends on whether they are located in urban or rural areas is,” indeed, “puzzling.” If I understand CAGW theory, CO2 being gratuitously pumped into the atmosphere by urbans ought to cause urban temperature minima to be markedly higher than rural areas where one would think there would be less CO2. Unless, of course, CO2 instantly disperses. But even then, I would think that if urban areas warmed significantly more than rural areas, the reflective properties of CO2 still would keep urban minimum temperatures relatively higher.
The precipitation findings, however, are unsurprising. A column of hot air rising over any given urban area is bound to produce precipitation. Just watch Broward County, Florida, on any Summer’s day.
there comes a point of temperature increase termination, though the assumption is that during periods of cooling, the urban heat effect decreases, and during warming periods, it increases in step with rural areas
Much ado about nothing really
Sadly Phil Jones 1990 Nature paper declaring that UHI is not a factor in global warming still stands and is quoted routinely.
In the absence of any data, it is not possible to prove that he was not comparing the price lists of urban and rural chinese take aways
That’s AGW science for you
I wonder if they took into account things like where the prevailing wind was from.
We see a picture of a wheat field next to a city, where were their study fields?
Were they close to large urban areas, or even near areas of newly cleared land.
If the prevailing wind is from the direction of a city, there could be significant UHI drag across the study field, whereas if the prevailing wind is towards the city, there would be minimal UHI effect.
And of course, the prevailing winds in many places is often dependant on time of day.
How many other things have they ignored?
Is this really just another superficial study?
I second Disko Troop.
Someone who can’t write an abstract in understandable English probably can’t do anything else in science.
My two cents worth. They were trying to determine whether there is an actual general warming trend or the percieved warming trend is due solely to urbanization.
“…there were statistically significant changes in as many as half of the urban areas in temperature‐related indices, such as heating and cooling degree‐days and number of warm and cool nights, almost all of which are reflective of a general warming. Similarly, statistically significant changes (mostly increases) in indices related to extreme precipitation, such as daily maximum intensities and number of days with heavy precipitation, were detected in as many of 30% of the urban areas…”
That ” as many as 50%” and ” as many as 30%” were red flags for me. It implies that the data
as a whole may have implied that both rural and urban areas were warming ( though maybe at different rates), but the results were not statistically significant. Rather than throw their results away, they went data snooping, and found subsets of just about 50% and just about 30% that WOULD have been statisitically significant if they had been based on independent studies rather than on data snooping. The moral of this is, there MAY be a warming trend, but more data is needed before we can come to any reasonable conclusions.
See this referrence to “Bonferroni correction” and do a little google searching on the subject.
http://generallythinking.com/what-the-hell-is-bonferroni-correction/
“Imagine that we did 20 studies, and in each one we got a p value of exactly .05. A 5% chance of a fluke result over 20 studies means it’s odds on that one of these results really was a fluke. Now think about how many thousands of studies have been done over the years! This demonstrates the importance of replicating studies – fluke findings have definitely happened and will continue to happen.
However this situation isn’t limited to findings spread over multiple papers. Sometimes in larger papers with several studies and/or analyses rolled into one, you might get a similar predicament. Simply, the more tests you do in a paper, the more chance there is that one of them will have come about through pure chance.”
Disko Troop says:
September 22, 2011 at 1:41 am
“When will you learn that if there is no statistical result you pack it in and say “Oh well never mind, lets try something else.” Not publish mush and draw conclusions which are just not there. You are wasting my valuable time that could be spent chopping wood or de-rusting the gate and oiling the mailbox hinges. Rant over.”
Apparently, the goal now is to turn science into something like a sitcom father, the butt of all jokes.
Something to consider: ENSO long term oscillations create statistically significant differences on a large regional basis (ex: La Nina dominant pattern – colder in the west, warmer in the southeast). Do studies such as the one examined above take this into consideration? Might they be violating the random sample requirement by accidently picking stations predominantly located within one large regional area whose warming pattern is connected with ENSO parameters?
I sometimes get the feeling that when studies are done, they concentrate on station selections east of the Rockies and then assume the results apply to all of us in the US. However, ENSO influence across the US demonstrates statistically significant, and often opposite, large regional affects.
How many Rural sites were at airports? How many Urban sites?
Were they all GHCN sites?
Where is the data and methods?
@Roy Clark says:
September 21, 2011 at 9:44 pm
I believe you’re right. We should be closely monitoring the oceans, not landmasses. But urban areas figure prominently in the calculation of global warming. I wonder why that is? Human land bias, or because including urban areas gives the most warming?
@Andrew Harding says:
September 21, 2011 at 11:41 pm
Yet we keep including urban sites when measuring global warming and it seems the Urban Heat Island Effect is estimated as too low, giving a warming bias.
@Steven Mosher says:
September 21, 2011 at 9:01 pm
Yes, and we’ll keep going over this until it starts to sink in. Land use changes do affect temperature and if those changes aren’t properly taken into account in an open, consistent and scientifically justifiable we will continue to get invalid results when trying to track climate change.
If Berkley extend the instrument record back to 1800 (with a note of uncertainties covering the entire period 1800 to 2010), it will be interesting to see that. It should provide further insight into the warming out of the LIA.
@Bruce of Newcastle says: September 21, 2011 at 8:18 pm
/////////////////////////////////////////////
Dr Spencer is suggesting that on his analysis, there are arguments supporting a conclusion that there may have been no warming in the US as from 1970. If he is right on that, that would be a very important finding.
There are a number of weather stations not just in the US but also in other countries that suggest no warming has taken place during that or similar period.
The issue raised is if CO2 is a well mixed gas which has properties that lead to an increase in DWLWIR which in turn leads to an increase in temperature, how does the theory explain how a large land mass such as the US is not warming? Ditto, with repect to the other stations.
I know that this is ‘cherry picking’ however, the usual tool used for examining the soundness of a theory is to cherry pick examples and see how the theory copes with the selected scenario. If the theory is sound, it should hold up to the cherry picked case. If it does not hold up then it follows that it is not sound and the theory needs revision (possibly fundamental revision, possibly it cannot be revised to explain the cherry picked example at least not without falling down elsewhere and which case the theory is invalid).
I personally consider that not enough emphasis is placed upon cherry picked scenarios where the theory cannot provide an adequate explanation.
PS; A cherry picked scenario cannot prove a theory correct, however, a cherry picked scenario can prove a flaw in the theory, possible a fatal flaw.j
Evaporation of water, cools. Forests, crops, live by moisture transfer, i.e., evaporation driven. Cities live on pipes {enclosed water}, and wires {electricity}; very little evaporation.
To remove the UHI effect, we need to water our cities. 😉
I fully concur with others who have said that we should concentrate on monitoring and evaluating sea temp data. For a long time I have argued that the land based record has become so bastersized that it is unreliable and since it does not even measure the right metric (joules), and since the oceans store about 99% of the energy in the Earth climate system and land air/atmosphere just about 1%, the land temperature record should be completely ditched, and we should evaluate only ocean temps.
The problem is that we do not have good quality sea temperature data extending very far back in time. Therefore it is very difficult to evaluate what has taken place during the past century, and of course, natural ocean cycles play a role in ocean temperature and you therefore need a long period before and valid extrapolations can be drawn. However, as regards the future to see the heare and now, we should concentrate just on ocean data.
Of course the Team have to much invested in their land based data records and are gate keepers to the adjustments made, so they will never agree no matter how strong the scientif case is for favouring ocean temperature data sets.
The following two posts hit the nail squarely on the head.
Paul Irwin says:
September 21, 2011 at 9:13 pm
the hundreds of thousands of tons of asphalt, concrete, sidewalks, and building surface area per square mile of a city absorbs vast amounts of heat that simply must skew the urban temperature record. as air moves over and beyond urban area, it carries that heat elsewhere. i’d be willing to bet that rural areas in the prevailing downwind path of a large urban area also show a similar warming trend. the null hypothesis, i think, would be to compare the temperature trends when the wind is blowing the opposite direction.
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
/////////////////
I have done something similar by comparing rural locations outside of the St. Louis and Kansas City metro areas using the normal prevailing winds. I found that temps downwind of these metro areas are DEFINITELY higher than their upwind counterparts.
////////////////////
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Jit says:
September 21, 2011 at 11:54 pm
The comparison should not be rural vs urban 1950 to now. It should be *rural in 1950, urban now* vs. *rural 1950, still rural now*. I would expect the trends in places whose land use hadn’t changed over the last 60 years to be the same, whatever they were over the period.
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
///////////////////
Once again I have had a similar look at stations which, BASED ON PERSONAL KNOWLEDGE , of the stations in question, met this standard and found this assumption to be true.
In addition I have made actual temperature measurements in and around the small town I live in (pop 2500) and found the evening temps to be 3 to 5 deg warmer at my residence than just one mile out of town during calm nights and on nights with light breezes from the east ,west,or north east. Winds from the south do not show this difference as I live on the extreme south edge of town and to the south of me there is basically nothing but forest for hundreds of miles.
richard verney says:
September 22, 2011 at 7:19 am
@Bruce of Newcastle says: September 21, 2011 at 8:18 pm
/////////////////////////////////////////////
Dr Spencer is suggesting that on his analysis, there are arguments supporting a conclusion that there may have been no warming in the US as from 1970. If he is right on that, that would be a very important finding.
With all the finagling of the numbers, I have wondered the same thing. The consistent discounting of land use change, UHI, station elimination and moves, and all biases in one direction, all we can come up with is a fraction of a degree in a century. I suspect this doesn’t rise above the noise. So where is the accelerating heating that is a prerequisite for catastrophe?
What dataset are they using? If it is a homoginized dataset then the difference from Urban to Rural is smoothed out.
“Mark and two Cats says:
September 21, 2011 at 11:17 pm
The woman in the picture is a climate expert – she is outstanding in her field.”
Excellent.
*****
Article says:
the failure to find a difference between rural and urban areas needs more investigation
*****
Does it? Dr Spencer found the lowest populated areas experienced the greatest UHIE increases. IOW, it’s almost impossible to find truly rural, unchanged sites, and the “rural” stations examined in this article are actually significantly affected by UHI.
Dennis Dunton says:
September 22, 2011 at 7:32 am
“I have done something similar by comparing rural locations outside of the St. Louis and Kansas City metro areas using the normal prevailing winds. I found that temps downwind of these metro areas are DEFINITELY higher than their upwind counterparts.”
St. Louis is an excellent place to do UHI studies. There is the old city, the old suburbs, the new suburbs, the newer suburbs, and the newest suburbs. I lived on the border of the old city and the old suburbs for many years. I had many friends in the newer suburbs. In summer, the difference in temperature between old city and newer suburbs ranged from 6 to 12 degrees Fahrenheit. In the winter, the difference ranged from 4 to 8. Of course, the biggest difference was in summer evenings because the old city and the old suburbs cool very slowly. But they have always cooled very slowly because they were designed that way. That area is dominated by row houses.
In St. Louis, the “micro” aspect of temperature measurement becomes very important. Scientists must be sure that there is something like a fair distribution of thermometers among old city and the several suburbs. Doing this requires some scientific investigation of the several thermometer sites. Scientists must know the physical influences on individual thermometers just as Briffa should have known the physical influences on individual trees.
Scientists must report their raw data and intelligent summaries of it. Substituting statistical trend analysis for raw data proves only that the scientists are intent upon hiding the pea. (I know, working with raw data is fatiguing, but this is the age of computers.)
beng says:
September 22, 2011 at 8:39 am
“Does it? Dr Spencer found the lowest populated areas experienced the greatest UHIE increases.”
You need to present his entire system of classification to make your claim meaningful. The lowest populated areas might be one percent of the total. In any case, their specific conditions must be known. If the thermometers were placed in villages of 100 and are now in towns of 1000, that could merit a special category of rapid growth.
“IOW, it’s almost impossible to find truly rural, unchanged sites, and the “rural” stations examined in this article are actually significantly affected by UHI.”
Really? Have you ever driven from St. Louis to Atlanta? There is one city on one hill, Nashville, and one large town in one gully, Chattanooga. Have you ever driven across Kansas? There are signs saying “No gasoline for 80 miles.”
From Paul Irwin on September 21, 2011 at 9:13 pm:
“the hundreds of thousands of tons of asphalt, concrete, sidewalks, and building surface area per square mile of a city absorbs vast amounts of heat that simply must skew the urban temperature record.”
This points needs to be amplified by considering that all these tons are above-ground, with heat-sink style geometry presenting a much greater surface area to ambient air than the mere acreage they sit on. This means that urban heating by dark surfaces is actually ameliorated by the heat-storage capacity of all that stuff, storing some daily heat and releasing it at night.
Somebody try comparing two cities with the same total albedo
(determined from aerial photographs)
but greatly differing in their mass of above-ground build-up
(gauged by average building height).
The two places should have differing daily temp curves.
I think these guys must really live in their own little worlds and never think about the real one outside, in which the rest of us get in our cars on winter evenings and drive out of town, with the temperature dropping half a degree every few miles we go. How long will we have to put up with this nonsense?
And why is it apparently only children (link at top of this thread, and that 11 year old girl who did much the same kind of project) who do the oh so simple studies of rural vs city temperatures, using GISS’ own data, and find, oh wow, cities are getting hotter but the countryside isn’t?
Perhaps someone can answer whether any of the analyses of Anthony Watts’ Surfact Stations data address the quality of rural vs. urban stations. I.e.: what percent of rural stations held CRN rankings of “4” or “5”, versus the percent of similarly low-ranking stations in the urban areas.
The lowest populated areas might be one percent of the total. In any case, their specific conditions must be known.
I don’t think the Mishra research looked that carefully at their rural siting selections. Their paper makes no references to night lights, and the only criteria I can find is that their “non-urban” sites be 15 or so km outside the city, so that there would be no spillover of UHI effects there.
Mishra’s paper gives equal weight to the trends in both city and country, pairing them, then claiming that a statistically-significant trend in one is a verification of a similar trend in the other. Voila! Warming in both!
As noted by Beng (above), Spencer and others have pointed out that rural stations can be just as bad as, if not worse than, poorly sited stations in the city.
Anthony also addressed this question in the previous posting of this article (a few weeks ago).
Richard verney – the extended record should indeed be interesting, as should their sensitivity to station quality (with important contributions from Anthony Watts, apparently), UHI effects, and their apparently new statistical approach. The impression one gets from the AGU abstract is that, when all is said and done, the overall signal is similar to the other 3 products.
Very much looking forward to the details.
Bill Parsons says:
September 22, 2011 at 10:52 am
“I don’t think the Mishra research looked that carefully at their rural siting selections. Their paper makes no references to night lights, and the only criteria I can find is that their “non-urban” sites be 15 or so km outside the city, so that there would be no spillover of UHI effects there.”
How quaint. Where are these guys from, Edinburgh? Fifteen kilometers outside the city limits of Atlanta, St. Louis, Chicago, and most vibrant American cities is a new downtown more densely built than the old downtown in the city limits.
On the sea temp data history: I believe some at this site have pointed out in the past that sea captains traversing the seas have kept daily logs, including meteorologic data, for centuries.
Though creating a useful data base out of these logs might be hard work, time consuming, and expensive, the consequences of government decisions re CAGW are so great, that funding for such tedious research ought to be readily available.
Sorry.. their “band” of separation between urban and “non-urban” was 25 km, not 15. I was thinking of miles.
I still think they should have selected 4’s and 5’s from the Surface Station “approved” sites. Any studies which deliberately ignore Watts’ research do so at their own peril.
Bill Parsons says:
September 22, 2011 at 12:48 pm
“Sorry.. their “band” of separation between urban and “non-urban” was 25 km, not 15. I was thinking of miles.”
Won’t matter much. Fifteen miles north of the Atlanta city limits is still downtown Atlanta.
What we need are scientists who have instincts for the empirical. Scientists who respect the authority of actual observations of reality. Instead, we have a bunch of self-taught wannabe statisticians who are interested only in trends and never in empirical observation. In fact, they are repulsed by empirical observations. In the last couple of months on this site, Mosher did an informal statistical analysis of a temperature station at an airport near San Francisco. The station was interesting because it was reading 4 degrees higher than nearby stations. Mosher’s analysis made all 4 of those degrees go away. My response to such analysis is “Thank You, Sir, for exhibiting your great skill at hiding the pea.”
I see a trend in “science”, especially “climate science”, especially among the pro AGW crowd, to what can only be called white collar science. Basically, instead of actually doing anything themselves, getting their hands dirty with actual experiments and gathering real data and finding out what the actual local conditions are (so they can factor those in), they instead stay in their nice cozy bureaucratic office environment and play with statistics from other peoples data. This is what they know, from their ivory tower existence inside universities, where they never have to be in contact with reality, but can instead only concern themselves with the theoretical rather than ever getting their hands dirty with, say, raw nature.
If they ever got out of that nice comfey air conditioned office environment, they would know that cities are hot and the country is cool. I doubt that many of them ever even leave the city, some may never have. They are basically like the new ruling liberal elite, the noble lords who live in the castle, while only the peasants live out in the fields and country. Like nobles, they don’t want to get their hands dirty with real work and leave their comfey academic environment (except for an occasional junket to exotic vacation spots called climate conferences). They also hang out only with their own kind, and thus may never even hear that anyone has a different opinion than what is the current peer pressure enforced (and grant money reinforced) local academic one. In short, they live in their own little world.
Thus we have this study, yet another playing with statistics of other peoples data, about which the players actually know little about, such as what the local conditions are or how it was gathered. As such, it is virtually irrelevant to the real world. Not that these people care, they don’t live in that world.
Where was that photo taken?
People will always look for the next new thing. Today it’s content; yesterday, it was social media