Our Urban “Climate Crisis”

By Jim Steele

Published in Pacifica Tribune May 14, 2019

What’s Natural

Our Urban “Climate Crisis”

clip_image002

Based on a globally averaged statistic, some scientists and several politicians claim we are facing a climate crisis. Although it’s wise to think globally, organisms are never affected by global averages. Never! Organisms only respond to local conditions. Always! Given that weather stations around the globe only record local conditions, it is important to understand over one third of the earth’s weather stations report a cooling trend (i.e. Fig 4 below ) Cooling trends have various local and regional causes, but clearly, areas with cooling trends are not facing a “warming climate crisis”. Unfortunately, by averaging cooling and warming trends, the local factors affecting varied trends have been obscured.

It is well known as human populations grow, landscapes lose increasing amounts of natural vegetation, experience a loss of soil moisture and are increasingly covered by heat absorbing pavement and structures. All those factors raise temperatures so that a city’s downtown area can be 10°F higher than nearby rural areas. Despite urban areas representing less than 3% of the USA’s land surface, 82% of our weather stations are located in urbanized areas. This prompts critical thinkers to ask, “have warmer urbanized landscapes biased the globally averaged temperature?” (Arctic warming also biases the global average, but that dynamic must await a future article.)

clip_image004

Satellite data reveal that in forested areas the maximum surface temperatures are 36°F cooler than in grassy areas, and grassy areas’ maximum surface temperatures can be 36°F cooler than the unvegetated surfaces of deserts and cities. To appreciate the warming effects of altered landscapes, walk barefoot across a cool grassy lawn on a warm sunny day and then step onto a burning asphalt roadway.

In natural areas like Yosemite National Park, maximum air temperatures are cooler now than during the 1930s. In less densely populated and more heavily forested California, maximum air temperatures across the northern two thirds of the state have not exceeded temperatures of the 1930s. In contrast, recently urbanized communities in China report rapid warming of 3°F to 9°F in just 10 years, associated with the loss of vegetation.

clip_image006

Although altered urban landscapes undeniably raise local temperatures, some climate researchers suggest warmer urban temperatures do not bias the globally averaged warming trend. They argue warming trends in rural areas are similar to urbanized areas. So, they theorize a warmer global temperature is simply the result of a stronger greenhouse effect. However, such studies failed to analyze how changes in vegetation and wetness can similarly raise temperatures in both rural and urban areas. For example, researchers reported overgrazing had raised grassland temperatures 7°F higher compared to grassland that had not been grazed. Heat from asphalt will increase temperatures at rural weather stations just as readily as urban stations.

To truly determine the effects of climate change on natural habitats requires observing trends from tree ring data obtained from mostly pristine landscapes. Instrumental data are overwhelmingly measured in disturbed urbanized areas. Thus, the difference between instrumental and tree ring temperature trends can illustrate to what degree landscapes changes have biased natural temperature trends. And those trends are strikingly different!

The latest reconstructions of summer temperature trends from the best tree ring data suggest the warmest 30-year period happened between 1927 and 1956. After 1956, tree rings recorded a period of cooling that lowered global temperatures by over 1°F. In contrast, although tree rings and instrumental temperatures agreed up to 1950, the instrumental temperature trend, as presented in NASA graphs, suggests a temperature plateau from 1950 to 1970 and little or no cooling. So, are these contrasting trends the result of an increased urban warming effect offsetting natural cooling?

clip_image008

After decades of cooling, tree ring data recorded a global warming trend but with temperatures just now reaching a warmth that approaches the 1930s and 40s. In contrast, instrumental data suggests global temperatures have risen by more than 1°F above the 1940s. Some suggest tree rings have suddenly become insensitive to recent warmth? But the different warming trends are again better explained by a growing loss of vegetation and increasing areas covered by asphalt affecting temperatures measured by thermometers compared with temperatures determined from tree ring data in natural habitats.

Humans are increasingly inhabiting urban environments with 66% of humans projected to inhabit urban areas by 2030. High population densities typically reduce cooling vegetation, reduce wetlands and soil moisture, and increase landscape areas covered by heat retaining pavements. Thus, we should expect trends biased from urbanized landscapes to continue to rise. But there is a real solution to this “urban climate crisis.” It requires increasing vegetation, creating more parks and greenbelts, restoring wetlands and streams, and reducing heat absorbing pavements and roofs. Reducing CO2 concentrations will not reduce stifling urban temperatures.

Jim Steele is the retired director of San Francisco State University’s Sierra Nevada Field Campus and authored Landscapes and Cycles: An Environmentalist’s Journey to Climate Skepticism.

Contact: naturalclimatechange@earthlink.net

0 0 votes
Article Rating
73 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Tom Gelsthorpe
May 17, 2019 6:26 am

A popular children’s story of my youth was “Mr. Flibbertyjib,” about a fussbudget who lived in a city and got annoyed about car horns, garbage trucks, etc. until he couldn’t sleep. He longed to move out to the country for peace and quiet.

When he got his wish, the silence drove him crazy until the occasional cricket or tree frog made him jump a foot. Mr. Flibbertyjib slept worse than before. He couldn’t wait to get back to the city. When he did, he felt grateful for familiar surroundings and slept like a baby.

The moral of the story is that people are always discontented, until they learn to be grateful. Joni Mitchell’s song, “Don’t it always seem to go that you don’t know what you’ve got til it’s gone?” is a similar theme.

I believe that much of today’s climate panic is Flibbertyjibism. It’s city people with a distorted view of the way things are supposed to be, without knowing how good they’ve got it. The urban heat island effect is too subtle for them. They’re jaded from crass entertainments, and crave melodrama — facts be damned. Next to nobody would notice a change in the weather, or climate, if it weren’t for the relentless propaganda.

I was a farmer for 34 years, and remain acutely attuned to weather, but city folk don’t believe a word I say.

Reply to  Tom Gelsthorpe
May 17, 2019 7:32 am

The incessant warming propaganda is starting to have serious side effects. The warmists and media need to start being held to account. A number of programs recently here in the UK about the increasing number of people having “eco anxiety”, with people saying they are overwhelmed by the “disaster”, refusing to have babies and all the rest of it.

Goldrider
Reply to  Tom Gelsthorpe
May 17, 2019 7:37 am

Maybe when they’re standing up to their butts in snow, they’ll finally admit we’re right.

accordionsrule
Reply to  Goldrider
May 17, 2019 9:31 am

I wish, but nope. GlobalWarming causes snow.

Bill Powers
Reply to  Goldrider
May 17, 2019 10:33 am

Gold pay close attention. Accordion is correct. Nothing will move them from their position because their motive is dishonest. They created a hobgoblin to scare the unsuspecting populace.

They will never admit defeat even in the face of overwhelming evidence to the contrary. They have control of the Propaganda Press so they can disappear past commentary and redirect the narrative however they deem necessary to keep the unsuspecting in fear. Newspeak. You can read about it in Orwell’s “1984”. Snow now stands as evidence of Warming. The novel use to be a dystopian cautionary tale. in another generation it will read like a current affairs documentary

Severian
Reply to  Tom Gelsthorpe
May 17, 2019 7:47 am

Sadly, urban folks get really distorted views about a lot of things. What happens outside the cities is reality, inside the cities is anything but “natural.” Robert Silverberg wrote an excellent science fiction book called “The World Inside” which highlighted the essential differences between urban and rural people and societies. One place this is highly apparent is in how city folk view food animals. Most farmers I know understand they are product, not pets.

As for noise, I grew up an an Air Force base, and was used to hearing aircraft running up engines and taking off and landing all night. When we moved off base, I had a hard time going to sleep for a couple of weeks, the lack of noise took a while to get used to. My father could sleep in a tent next to a taxiway with B-52s going back and forth.

Reply to  Severian
May 17, 2019 10:08 am

You and Tom Gelsthorpe make excellent points. In my pre-teens, we were living on the approach path to the runway used most often in the summer at Idlewild (now JFK). Not only did the 707s come low over our house, their landing lights illuminated my bedroom. I spent my early teens living in Manhattan – the good old days with everyone honking their horns all night long, and lots of sirens. I then moved to the countryside in New England. I guess I was lucky, because I never had any trouble sleeping in any of those places.

Loydo
Reply to  Severian
May 17, 2019 5:02 pm

“Most farmers I know understand they are product…”

And that excuses the horrors and medieval cruelty of factory farming? Reducing living creatures to mere “product” is a symptom of a very sick society. Do you think any other animal values it’s own life less than you do?

Farmer Ch E retired
Reply to  Loydo
May 18, 2019 4:30 am

Is it possible that farmers/ranchers treat animals better and more humane than their natural predators? We fed our black angus daily on snow-covered pastures in western MT and provides sheds with straw bedding for protection during winter blizzards. Not exactly factory farming but a different prospective. Natural predators (eagles, hawks, mountain lions, bears, orcas, seals, etc.) are not very humane but are very natural.

Dr Deanster
Reply to  Loydo
May 18, 2019 8:35 am

What vegan-terririst don’t seem to comprehend is that the meat producing industries are the buffer that protects the wild populations of animals from decimation. They farm beef, chicken, fish, etc, so we don’t go out and extinct the wild populations. The deer population in my state was decimated during the Great Depression because people could not afford to buy food.

As such, I’m sure the wild populations of animals would prefer that you take your sanctimonious political brand of environmentalism and stick it.

Vicus
Reply to  Loydo
May 20, 2019 12:08 pm

Ahh, your hyperbolic word usage matches your level of “Reeeeeeee”ing.

Steven Miller
Reply to  Tom Gelsthorpe
May 17, 2019 9:12 am

Tom Gelsthorpe… very good observation… right on the money.

commieBob
May 17, 2019 6:27 am

Although it’s wise to think globally, organisms are never affected by global averages. Never! Organisms only respond to local conditions. Always!

That reminds me of a joke.

One of the best-ever economist jokes has three econometricians out deer hunting. They encounter a deer, and the first econometrician takes his shot and misses one meter to the left. Then the second takes his shot and misses one meter to the right, whereupon the third begins jumping up and down and calls out excitedly, “We got it! We got it!” link

brians356
Reply to  commieBob
May 17, 2019 10:51 am

Well over half of US states’ hottest temperatures were recorded in or prior to 1940. Fact.

Reply to  commieBob
May 17, 2019 10:52 am

LOL. I like it and will use it

Steve O
Reply to  commieBob
May 17, 2019 12:37 pm

Like the drunk walking down the sidewalk next to a busy road. On AVERAGE, he’s okay.

griff
May 17, 2019 6:36 am

So… Alaska? Svalbard? I notice you don’t include the arctic in this.

and the USA is a small part of the world…

May 17, 2019 6:38 am

Why does the MSM never pick up on this kind of news? Lies of ommission are as bad as lies of commission.

Gary
Reply to  Tim Gorman
May 17, 2019 8:04 am

Actually, they’re worse because a lie of commission can be refuted to its face. Lies off omission fester in the dark and continue to deceive because they don’t get addressed. This is the trick of the MSM — leave out important context and allow the consumer to believe a fantasy.

May 17, 2019 6:56 am

Trying to distinguish a general trend from data with Urban Heat Island biases in some of the sites has apparently led to general “corrections” of the data, rather than examining the individual stations.
That sort of thing is probably why the satellite measurements have diverged from land station databases like GISSTEMP or HADCRUT.

a_scientist
Reply to  Tom Halla
May 17, 2019 7:10 am

Tom Halla, you are right, and this is further confirmed by the USCRN, with its pristine well sited measurements, that shows NO warming in the USA for the past 15 years of its existence.

Bindidon
Reply to  a_scientist
May 18, 2019 4:52 am

No he isn’t, nor are you.

Here is a CONUS comparison of 46 of the 71 well sited stations chosen by surfacestations.org with the full GHCN daily set (over 8000):
https://drive.google.com/file/d/1SNR72I3gN9nztqKfHi2MbMxCTvgxY2yb/view

and with a zoom into the sat era, with a UAH6.0 LT comparison:
https://drive.google.com/file/d/12Khxeii6he3PhW-xKJL-nhlJkA1frWmS/view

Maybe one day I process the CRN stuff too. But there will be few differences.

May 17, 2019 7:31 am

I live the majority of my entire existence in a temperature well above the global average. I could easily tolerate a change of a few degrees above this, and I would definitely be uncomfortable at a few degrees below this.

I’m good at about 75 F. air temp. I deal well with upper 70’s, 80’s, and tolerate 90 (F.)

Many people in the developed world live in climate controlled environments most of their lives, and have no relationship to the anything other than artificially maintained temperatures. Also, in these climate-controlled enclosures, many people live most of their lives in higher CO2 environments than the official measure.

Many people, thus, have no clue what they have been led to believe that they should be alarmed about. They are already living out of those idealized bounds, … every … single … day.

Latitude
Reply to  Robert Kernodle
May 17, 2019 9:59 am

80F on a cloudy windy day…and I’m cold

Reply to  Latitude
May 19, 2019 11:44 am

Not quite that bad for me; for I am further north of you, Lat; my former neighbor. Still if it gets much below 68F/20C, I don’t like it. It get hotter here, too; I think, because I am both further north and more inland. To all you Northerners/Yanks who are not acclimated to HHH like me or Lat; that’s why sudden drops to temps you call warm feel cold to us. We are adapted to losing heat somewhat steadily and our mechanisms to retain core temps can’t rapidly compensate.

Vicus
Reply to  Latitude
May 20, 2019 12:18 pm

75° and up for me. I relish in 90°. In a cold zone too.

On the outer Barcoo
Reply to  Robert Kernodle
May 17, 2019 11:35 am

As an exploration geologist, I’ve regularly travelled between sub-tropical Far North Queensland in Australia and the Yukon Territory in Canada. The air temperatures fluctuated between more than +40ºC (Australian summer) and -30ºC (Canadian winter … and I’ve lived to tell the tale.

Rhys Jaggar
Reply to  Robert Kernodle
May 18, 2019 5:51 am

Your body actually adapts to where you live.

I lived in Scotland for seven years before returning back to SE England where I grew up. I really noticed my body in effect semi hibernating in the long Scottish winter nights and when I came back south, despite being physically fitter than ever, I realised my body had adapted to the cooler Scottish temperatures so I sweated more despite being capable of running a marathon in around three hours or so. My body readapted over three or so years to the warmer south.

tom0mason
May 17, 2019 7:35 am

“Although it’s wise to think globally, organisms are never affected by global averages. Never! Organisms only respond to local conditions. Always! ”

Yes, yes yes!
And I’ve been saying as much for decades.
Climate, just like weather, is a local affect of many LOCAL parameters and variables.
All this talk of Global Climate is just BS . An notion from the faulty idea that if you average some parameters within chaotic system you can find a trend, that’s just not possible as the chaos is the signal that sets the trend. Go ask the butterflies.

Robert W Turner
May 17, 2019 7:43 am

Hmm, what should I wear today? I better check the global average temperature.

H.R.
Reply to  Robert W Turner
May 17, 2019 4:43 pm

Perfect.

cerescokid
May 17, 2019 7:44 am

Another great post.

“Despite urban areas representing less than 3% of the USA’s land surface, 82% of our weather stations are located in urbanized areas.” A bookmark in my brain for that one.

Jim, I do have a question about one statistic. Satellite data reveal maximum surface temperatures in forest areas are 36F cooler than in grassy areas.

Is that typo? I understand why they would be cooler but this is a very big difference.

Could you expand on that? Would other variables be involved?

Reply to  cerescokid
May 17, 2019 10:31 am

Cerescokid,

The comparison is from Fig 3 linked below in the MIldrexler 2011 paper (link to paper is embedded in the text “Satellite data reveals” in the article

http://landscapesandcycles.net/image/131656598_scaled_818x755.png

The temperatures are surface, or skin temperatures not air temperatures measured 6 feet above the ground. Skin temperatures are mostly proportional air temperatures but usually much warmer.

Forest skin temperatures have a fairly wide range as do grasslands and barren ground, so my 36F comparison is based on 20C differences seen in the graph, but there is wiggle room to choose a smaller or larger range

IAMPCBOB
May 17, 2019 7:53 am

Facts, logic and good old common sense are things foreign to the liberals, and the Left. All they can see, apparently, is the tip of their own noses. Common sense tells us that the cities, with all that concrete and brick work is GOING to retain more heat than the country side, but chicken little’s think it means the end of the world. Maybe they just need to drive OUTSIDE the city and see for themselves, that the world is doing just fine! WITHOUT their fear mongering! Just sayin…

Clyde Spencer
May 17, 2019 8:06 am

Jim
It has been a few years since I started following this topic. Early on, I did a search for historical temperature records. I ran across a California map that was apparently part of a PhD dissertation. What it basically showed was that there was a correlation between the average temperatures and population growth of counties. Notably, there was a cooling in the NE corner of the state. Alas, like so many things on the internet, it apparently has disappeared. Recent attempts to find it have failed.

Reply to  Clyde Spencer
May 17, 2019 10:42 am

Hi Clyde,

Perhaps you refer to a 1996 paper by past California state climateologist Goodridge. Anthony has posted that graph a few times.

I have saved it to my site in case it “disappeared” at http://landscapesandcycles.net/image/77144952.jpg

There is also a graph of California max and min temps in a 2014 paper showing regions where maximum temperatures had cooled since the 1930s.

http://landscapesandcycles.net/image/103207378_scaled_615x539.png

Clyde Spencer
Reply to  Jim Steele
May 17, 2019 11:25 am

Jim

The time is about right for the paper by Goodridge. However, what I had found was a color-coded California map showing average temperature changes over the preceding 100 years.

I did another internet search to see if I might get lucky, but came up empty handed. Should you be interested, I may still have a copy on an older computer that isn’t running. However, I’m not going to get to it right away. And, we would have to make arrangements for me to send it to you because I don’t have a WordPress account.

Reply to  Clyde Spencer
May 17, 2019 12:21 pm

Whenever you get the chance send it to naturalclimatechange@earthlink.net

Reply to  Jim Steele
May 17, 2019 2:44 pm

Jim,
Great article…..thanks.

I note that you illustrate the difference in late afternoon temperatures between rural and urban areas. I’m in agreement with this.

However, I would like to point out that the majority of the global warming contribution comes from night time/low temperatures. I believe from the increase in moisture in the air.
Your link to California temperatures that shows average mins and average max temps is an ideal example of this. Only the mins have increased across the entire state over the last century.

This is too be expected with more moisture in the air, which could be modulated in a state like California by warmer Pacific Ocean water temperatures.
With an increase in El Ninos during this time frame, this also would line up with cooler highs and warmer lows.

The Winter-only Arctic warming is the quintessential example of this as the Arctic, during the Winter is almost like one long night with constant darkness or very low angled sun. The added moisture in the air keeps the bottom from completely falling out under conditions(in the absence of air mass advection) that feature, favorable around the clock heat loss from the surface.

This is something completely different than the UHI that you describe and the distortions that the UHI effect causes (which I agree with) because of the very flawed locations for weather instruments/thermometers as comprehensively documented by Anthony with his past research.

Steve Skeptica
May 17, 2019 8:14 am

Not too long ago Tony Heller posted a historical graph of temperatures in a locality. It did seem like variablility decreased with time. Could changes in variability be a proxy for the impact of the heat island effect?

Rick
May 17, 2019 8:14 am

Opening your eyes, nose, and feeling the breeze will tell you more about the weather than any weather channel. Doing it daily will tell you more about the climate than any laboratory experiment or computer program which only work with tiny samples of our weather, much of which is suspect data.

CoRev
May 17, 2019 9:00 am

Jim, is there a decimal missing in this statement: “surface temperatures are 36°F cooler than in grassy areas, and grassy areas’ maximum surface temperatures can be 36°F cooler than the unvegetated surfaces of deserts and cities. “?

Reply to  CoRev
May 17, 2019 11:22 am

Nope

The comparison is from Fig 3 linked below in the MIldrexler 2011 paper (link to paper is embedded in the text “Satellite data reveals” in the article

http://landscapesandcycles.net/image/131656598_scaled_818x755.png

The temperatures are surface, or skin temperatures not air temperatures measured 6 feet above the ground. Skin temperatures are mostly proportional to air temperatures but usually much warmer.

Forest skin temperatures have a fairly wide range as do grasslands and barren ground, so my 36F comparison is based on 20C differences seen in the graph, but there is wiggle room to choose a smaller or larger range

CoRev
May 17, 2019 9:06 am

BTW, doesn’t this 82% of stations in urban/developed areas contradict Mosher’s recent article? https://wattsupwiththat.com/2019/05/07/willis-favorite-airport/

Reply to  CoRev
May 17, 2019 10:50 am

CoRev, The 82% number is from an article Mosher co-authored earlier. Mosher’s arecent rgument concerns how we define urban vs rural. My argument in this article is quibbling over the best definition of rural or urban is not helpful regards determining how landscape changes affect both rural and urban areas. I am suggesting tree rings from more pristine areas are the best measure of natural climate change, and tree ring temperate data diverges (cooler) from instrumental data that are mostly “urbanized to some degree .

Robert W Turner
May 17, 2019 9:18 am

All that wall space in major urban areas is an opportunity to grow things like lettuce and tomatoes. This would bring so many benefits: cooling urban daytime highs, reduce pollution, provide local food sources that are fresher and cheaper, aesthetics, reduced land use change from agriculture, and get urban dwellers outside more. Too bad environmental science has been hijacked by the mentals.

James McCartney
May 17, 2019 9:30 am

Quite apart from loss of vegetation and the presence of heat-absorbing structures, the amount of waste heat produced in cities must be a major factor here: by vehicles, by heating buildings in cold weather, and by air conditioning in hot weather. Anything actually burned in situ will also produce a local increase in carbon dioxide, although the local effect of this would be debatable.

On the outer Barcoo.
Reply to  James McCartney
May 17, 2019 11:47 am

The use of electric appliances also generates heat: light bulbs grow hot, toasters and ovens grow hot, as do vacuum cleaners, computers, televisions, air conditioners, fans, etc., etc.. Then there are car, truck, bus and train engines. And living humans also radiate body heat, so it’s not surprising that the greater the concentration of human beings, the greater the UHI effect.

May 17, 2019 9:41 am

I remain dubious of tree-mometers, however thank you Jim Steele good post.
Funny after 40 years and ever increasing hysteria orchestrated by our bureaus, we are still at the start point.
Is it warming?
Or Cooling?
We really cannot say, “It depends” to quote Bob Carter, but to witness the current pseudo-mass hysteria over assumed warming, reinforces my cynicism.
We remain at the “Number of angels on the head of a pin” stage after wasting billions of dollars.
Our claimed signal(the warming) is less than the uncertainties in our records.
Historically the proper answer may be “Duh”.
Urbanization has two obvious consequences, we have the majority of our voting population cut of from the weather and concentrated pockets of waste heat.
But who needs logic?
Emoting ..feels so good?

Reply to  John Robertson
May 17, 2019 10:18 am

Looking at the politics of climate change, with the liberal – left -Progressive folks totally committed to it, it might be valuable to remember the map of the US showing where the Clinton voters were (cities) and where the Trump voters were (non-cities). The Clinton voters – liberal – leftist – Progressive mainly live in cities, and may not have much grasp of nature and agriculture. I’m seeing a correlation there, but, of course, it may not be the causative factor.

Editor
Reply to  John Robertson
May 17, 2019 10:32 am

I remain dubious of tree-mometers“. Me too. I seriously doubt that they can distinguish between temperature and precipitation. I therefore doubt that any meaningful temperature-oriented conclusions can be drawn from them.

Reply to  John Robertson
May 17, 2019 2:37 pm

I like the research interest and debate around climate and weather. Its the perpetuating link with CO2 and our energy expenditure that p155es me off. Cars,planes or crickets….those early innovators must be turning in there Grave.

SH is hardly local, yet temperature appeared to take a holiday for 30 yrs.

https://reality348.wordpress.com

john
May 17, 2019 10:08 am

NYC will be having crisis after crisis now.

https://www.zerohedge.com/news/2019-05-17/new-york-ditches-gas-pipeline-proposal-despite-soaring-electricity-bills

I was dispatched to NYC in 05 to repair their grid when a heatwave fried it. No breeze at all either.

All the grandiose plans for offshore wind will be a disaster, no wind during heateaves and of course the chance or three of annual hurricanes that will destroy them.

Best of luck!

May 17, 2019 10:14 am

Here in the midwest there are a number of changes that make it difficult for me to assume a constant environment. One, is that much of the “hedge rows” put in after the dust bowl have been removed to allow bigger fields for bigger equipment. This not only removes cooling vegetation but also removes any resistance to the wind at the surface (hopefully this won’t result in another dustbowl). Two, is that many, many rural roads that used to be gravel or even dirt have been paved with asphalt. Other roads have been widened through rural areas to accommodate traffic between urban areas (multilane interstates and highways). This simply has to result in a temperature gradient that increased surface winds would spread over a wider area.

I’m sure warmista’s would say the changes to temperatures are minimal. However, when forecasts of one-hundredth’s or one-thousandth’s are being tossed around, you are talking in the realm of minimal.

Robr
May 17, 2019 10:45 am

Interesting thoughts, and UHI doubtlessly creates an upward bias. The Science on UHI is more demonstrable and settled that the Greenhouse Effect.

With these thoughts in mind, shouldn’t we expect to see a divergence between terrestrial and ocean surface temperatures?

To expand on the concept: A random selection of temps from 30 cities experiencing substantial growth should be compared to randomly selected ocean locations of identical latitude.

I’m sure Dr’s Spencer and Christy could compile this information in short order.

J Mac
May 17, 2019 1:52 pm

Jim Steele,
Another thoughtful presentation – Thanks!

Loydo
May 17, 2019 3:53 pm

This post sets out to disinform: “some scientists and several politicians claim we are facing a climate crisis.”
No. The vast majority of scientists and nearly every politician outside the US.

“This prompts critical thinkers to ask, “have warmer urbanized landscapes biased the globally averaged temperature?””

But from the link provided the answer is no:
“The small size, and its negative sign,
supports the key conclusion of prior groups that urban warming does not unduly bias
estimates of recent global temperature change.”

Completely refuting the lie: “Thus, we should expect trends biased from urbanized landscapes to continue to rise.

And so with the cited evidence showing the opposite of the supposition the author bizarrely but predictably concludes therefore: “Reducing CO2 concentrations will not reduce stifling urban temperatures.”

Despite the obvious contradiction and the maligned use of tree ring data, the angry mob wave their pitchforks in furious, credulous agreement. SMFH.

Reply to  Loydo
May 18, 2019 8:05 am

No. The vast majority of scientists and nearly every politician outside the US.

The majority of people in Medieval Europe believed in witches and fairies too.

May 17, 2019 5:03 pm

Loydo, Seems like it is you who set out to disinform.

Please read the whole article more carefully and put on your thinking cap.

I acknowledged “some climate researchers suggest warmer urban temperatures do not bias the globally averaged warming trend.” You simply gave an example of what I had written and from a link I gave you. But oddly you suggest there is a lie.

I continued, “They argue warming trends in rural areas are similar to urbanized areas. So, they theorize a warmer global temperature is simply the result of a stronger greenhouse effect. However, such studies failed to analyze how changes in vegetation and wetness can similarly raise temperatures in both rural and urban areas.”

Studies using rural vs urban comparisons do not account for landscape changes that affect temperarturs in both places. The tree ring data reveals the true difference.

Loydo
Reply to  Jim Steele
May 17, 2019 8:14 pm

“I acknowledged…”. No you didn’t. You only mentioned that in passing so as to refute it.

Its either an error or you are trying to disinform. How else do you get from the reference you cited 4 times:
“urban warming does not unduly bias estimates of recent global temperature change”,
to:
“Thus, we should expect trends biased from urbanized landscapes to continue to rise”?

And then you double down:
“the different warming trends are again better explained by a growing loss of vegetation and increasing areas covered by asphalt affecting temperatures measured by thermometers compared with temperatures determined from tree ring data in natural habitats”.
In other words its gotta be “urban warming”.

All that is apart from this howler:
“…organisms are never affected by global averages. Never!”

Never! (Unless the local conditions are experiencing average increases. Or unless you’re in higher latitudes or at higher elevations in which case the effects are even more pronounced.)

Or, if you try and dismiss AGW as merely being an “urban climate crisis.””Thus, the difference between instrumental and tree ring temperature trends can illustrate to what degree landscapes changes have biased natural temperature trends. And those trends are strikingly different!” –
How do explain that the Arctic, with zero urbanisation and zero land-use change, is warming three times faster than anywhere else?
wberkeleyearth.org/wp-content/uploads/2018/01/AnnualMap2017.png

“Never!”

Farmer Ch E retired
Reply to  Loydo
May 18, 2019 4:45 am

Maybe the arctic warming is due to greenhouse gas – namely water vapor which increases nighttime minimum temperatures.

Reply to  Loydo
May 18, 2019 6:44 am

loydo in your frenzy to dishonestly smear this article you prove to be a very bad reader, as well as a few other things

The article discussed the bias introduced into the global average by landscape changes such as urbanization. It does not as you dishonestly suggest “dismiss AGW as merely being an “urban climate crisis.”

Again, read carefully Lloyd. I wrote “(Arctic warming also biases the global average, but that dynamic must await a future article.)”. So stay tuned. I have a future newspaper article sowing why Arctic warming is best explained by natural oscillations, not by CO2.

The global averaged temperature is a chimera of many things. Only a complete fool would think only CO2 causes warming.

Loydo
Reply to  Jim Steele
May 18, 2019 6:01 pm

“Only a complete fool would think only CO2 causes warming.”

And only a disingenuous propagandist puts up this kind of straw man.

Reply to  Loydo
May 18, 2019 6:25 pm

“How do explain that the Arctic, with zero urbanisation and zero land-use change, is warming three times faster than anywhere else?”

The data I’ve seen about the warmer waters entering the Arctic would explain much concerning that issue.

May 17, 2019 6:11 pm

Tom Gels Thorpe, May 17, is 100 % right. Hence the difference between
city folk and country folk.

We have just this problem in Australia, the country folk used to have a
political party called “The Country Party”, but some years ago it found that
with the development of mining, the money for re-elections was going to the
city, so they decided to change their name to “The National Party”.

They entered into a somewhat uneasy coalition with the larger city
based Liberal Party (Conservative) but the country folk now complain that
they are neglected as the money being in the city, their “Representatives “”
are far more interested in where the big mining interests are, and what they
want

The result is of course that small parties such as the “”Shooters and Fishers
Party “” are emerging, and at State level do quite well. Today being a
Federal electron day it will be interesting to see what results.

The ex leader of the Nationals Barnaby Joyce is fighting against this City base thinking , so as its almost a certainty that the Labour Party will win, expect a
split between the two Conservative parties.

MJE VK5ELL

May 17, 2019 7:02 pm

Jim,
I enjoyed this article and it makes some great points.
On the tree rings, however I have always been a skeptic, especially in recent years.

With the massive atmospheric fertilization from CO2 thru the law of photosynthesis, which effects woody stemmed plants/trees the most and is greatly accelerating growth rates of all trees around the planet, how does one separate out, the tiny effect of a slight increase in global warming from, what must be an order of magnitude greater effect from the 25% increase in CO2?

It’s tough enough to discern between weather elements that contributed to past tree growth/rings(too cold, too warm, too wet, too dry) but when the trees are growing 20% or more faster, it would seem impossible to detect these signals when imposed over a long period of time(individual years can tell us if their was some sort of weather adversity).
It’s also interesting to note in this study that urban trees are growing faster than rural trees, while of course all trees are growing around 20% faster since the 1960’s.

https://www.sciencedaily.com/releases/2017/11/171113111037.htm

So my point is, if all trees are growing 20% faster(from the beneficial CO2) and trees in urban areas grow even faster than that because of the additional beneficial warmth, how can you tell, when it comes to ALL trees, how much of the 20% was from beneficial CO2 and how much was from the beneficial, slight global warming?

Tree rings don’t leave a specific signature that I know which tell us that X amount came from more CO2 and Y amount came from temperature. If X is contributing 10 times more than Y, changes is Y are going to be overwhelmed by the value of X.
Hundreds of tree studies indicate different increased growing rates for different tree species and even individual studies for the same tree species can vary by more than how much a 1 degree global temperature increase would effect that species.

So its not like we can assign an accurate value to how much extra growth the added CO2 is causing and use the difference between that and the actual growth to assume temperature because the margin of error for the CO2 growth rate is many times greater than what the effect on growth of the average temperature over a long period would cause.

I’m happy to learn more about tree ring studies from somebody that knows more about them than me.

Reply to  Mike Maguire
May 17, 2019 10:43 pm

Mike,

Indeed tree ring reconstructions should be viewed with prudent skepticism. However I believe tree ring science has evolved and generally provides reliable trends. Tree ring series from arid regions are generally not used because indeed growth there is dominated by a response to precipitation. For temperature indicators, tree rings are collected from regions where temperature is the dominant limiting factor such as near tree line.

Furthermore there has been an evolution towards using Maximum Latewood Density (MXD) instead of tree ring width. MXD analyses correlated best with instrumental data and anatomical and physiological studies suggest why. A single tree ring will undergo a transition from early wood to latewood each year. Early wood tracheids are larger and less dense. Earlywood tracheids’ main function is to conduct water/sap up into the canopy. When water is more abundant the pressure increases the size of the new cells. So early wood is much more affected by precipitation. The seasonal transition to latewood creates smaller denser cells. The function of latewood is to strengthen support by investing more in cell wall thickness with smaller cells and thus a greater density. Cell wall thickness is driven mostly by temperature and not preciptation, so latewood density is most sensitive temperature

The MXD reconstructions correlate much better with nearby instrumental temperatures, but there is still a divergence problem with the global average in the most recent decades. As SmartRock relates below, the tree ring researchers addressed the divergence problem stating , “Modern warming is poorly represented in the new record questioning the ability of MXD to capture very warm temperatures” Such a statement appears to be an attempt to appease the global warming alarmists and avoid an outright confrontation over which method reconstructs temperature trends best, because it also calls into question the what factors are affecting the hockey-stick fabrications.

Reply to  Jim Steele
May 18, 2019 4:54 am

Thanks Jim. That tells me some things that I didn’t know before about tree rings.

May 17, 2019 7:17 pm

This is interesting. Jim’s fourth figure (the tree-ring stuff) doesn’t have a caption to tell you what the blue and gold lines mean, so I followed his link to the paper (Schneider et al 2015), which basically takes tree ring numbers and pushes them through a whole load of statistics that I don’t understand. Jim’s figure is their figure 3(b). The blue line is is Schneider et al and the gold is from one of the Briffa et al 2001 papers from the climate factory in Norwich.

What’s interesting is that Schneider et al figure 3(c) is an expanded version of their own plot of temperature anomaly (summer months only) from 1900 to about 2002, with the CRUTEM “instrumental” record superimposed. They both track well from 1900 to about 1980, with a broad peak of about +0.3°C between 1930 and 1950, then go down to zero (the anomaly is from 1961 to 1990, so it has to average zero over that period) and then the tree ring goes up to about +0.2°C by 2000, while the “instrumental” plot does a perfect hockey stick up to +1.0°C.

What this tells me, as a casual observer, is that EITHER tree ring data do not reflect temperature trends at all, OR that the instrumental record has been stepped on to produce the desired result of post-1980 extreme warming. They simply cannot have it both ways.

But they try. They ignore it, and only make this absurdly disingenuous statement: “Modern warming is poorly represented in the new record questioning the ability of MXD to capture very warm temperatures”

They can’t very well say their methodology is useless (if so, why publish it?) and of course they can’t say that CRUTEM data don’t reflect reality, because really bad things would happen to them if they hinted at that.

Climate science – the oxymoron of the century.

May 17, 2019 8:16 pm

For a couple of days after most everyone else had left the thread, the poster known as Bendidon and I carried on the conversation. He insisted that gridding had to be done to prevent the overwhelming preponderance of stations in the US, Northern Europe, and Southeast Oz from biasing the global average. As evidence in favor of his position, he posted two sets of graphs, one with gridding and one without, and both of the sets calculating the global anomalies with and without US data, ostensibly to show how gridding corrects for the US overload.

What I thought was interesting was that with the gridding, he was able to remove the entire US data set from the chart and see no difference in the trend at all. Is it right that a calculation should allow one to remove practically an entire continent’s worth of data without having any significant effect on the average anomalies and trends?

These were the trends of the two sets of graphs.

No gridding, no US data: 0.15C/decade
Gridding, with or without US data: 0.10C/decade
No gridding, with US data: 0.08C/decade

My simple calculation gave a trend of 0.09C/decade.

What did the gridding do, though? Seems it smoothed the data so much that you could take away the entire US data set, which caused the ungridded trends to be different by a factor of two, and have the trends remain indistinguishable.

Is that really what we want?

Paramenter
May 18, 2019 1:23 pm

Hey Jim,

Great article. In terms of warming are you actually saying that mild naturally occurring warming trend was amplified by UHI bias? Or in fact we cannot even confirm with certainty any ‘global’ warming trend as observed in historical records warming trend is purely artifact of urbanization?

1sky1
May 18, 2019 2:52 pm

Progressive land-use changes outside of urban areas can have a profound effect upon the apparent “trend” in long records from nominally non-urban stations. That is the reason why linear trends vary so widely in any given region. Those trends cannot be assumed to be representative of truly pristine locations, which constitute the vast majority of the globe, yet form only a tiny fraction of of the global data base.

Given that high trend-variability and paucity of truly pristine sites, it’s easy to construct severely misleading time-histories of spatial averages. Through ad hoc algorithms and various other devices. the strong cooling that took place in the third quarter of the last century has been virtually eliminated in all of the widely trumpeted “global” indices. Although there should be many reservations about the efficacy of tree-ring records as temperature indicators, the presentation here is noteworthy in clearly indicating that we are not in any uncharted territory, but only now are reaching the high yearly temperatures experienced many decades ago. A very similar result can be obtained from carefully vetted station records throughout the globe.

Bindidon
May 19, 2019 5:04 pm

A further answer to James Schrumpf (hopefully the last one)

*
“What I thought was interesting was that with the gridding, he was able to remove the entire US data set from the chart and see no difference in the trend at all.”

You seem to have, maybe intentionally, left out the central point of the discussion.

Thus for the readers who might eventually visit this head post, I think it would be honest to show them again.

1. Regardless which GHCN global data set we consider, the US stations account for about 50% of the total stock. This is of course the case for the data set I was using (GHCN daily).

2. If we distribute all stations worldwide over a latitude/longitude grid, e.g. of 2.5 ° size , i.e. 77,000 km² (I prefer it because it has the same size as UAH’s grid) we see this:
https://drive.google.com/file/d/17ZgjmYUL43320EoLQ5bL0Hs3aYwas-gt/view

Even this picture still does not reflect reality: within the 100 most populated grid cells of the 413 having more stations than the average (14), 96 are in CONUS; the most populated one in the US has… 363!

The top 96 CONUS grid cells totalise about 16,000 stations; this no more than 4.5 % of the Globe’s land surfaces.

The effect is that if you generate a monthly time series out of all the worldwide station data, these 4.5 % will account for as much as 50 % of Globe’s land.

This is bare nonsense, not only at the global level, but even within CONUS: here you have 50 % of the grid cells with 16,000 stations, and 50 % of the grid cells containing the rest, i.e. 2,000 stations.

This evidently calls for a more equitable distribution of station measurement data, with the goal of avoiding that the Globe’s average temperature looks like CONUS’s backyard.

Let us look at two graphs showing this – not with data in anomaly form, depending on the baseline selection strategy chosen, but rather with data in absolute form.

The red plot in thre graphs shows The Globe, the other one the Globe minus CONUS.

3.1 Without gridding:
https://drive.google.com/file/d/1U0b9TfcSxUDp7EiDfDK85-8f7gMYCmwI/view

Linear estimates in °C / decade
– Globe: 0.07
– Globe minus CONUS: 0.27

3.2 With gridding:
https://drive.google.com/file/d/15aZe2L45P9AL2AjcM-prh509xrKuLKvz/view

Linear estimates in °C / decade
– Globe: 0.11
– Globe minus CONUS: 0.11

4. Why are the estimates in the gridding case so similar?

Simply because now, we compare two averages where 200 CONUS grid cells (at best 6 % of the Globe’s land) compete with 2000 grid cells (94 % of the Globe’s land).

*
“Is it right that a calculation should allow one to remove practically an entire continent’s worth of data without having any significant effect on the average anomalies and trends?”

An entire continent ? Really, James Schrumpf?

Don’t you suffer a bit of a little mental disease we Europeans use to call “americanocentrism” ?

Regards
J.-P. D.