There is a new paper that is being somewhat uncritically accepted at the new Climate “Skepticism” site (and, yes, the inverted commas are necessary) and at Bishop Hill. It’s by someone called J. Ray Bates and claims to estimate climate sensitivity using two zone energy balance models. The paper essentially concludes that the method used
give low and tightly-constrained EfCS values, in the neighbourhood of 1°C
which is slightly bizarre.
Fortunately, Andrew Dessler has already provided a rebuttal to an earlier version of the paper. Essentially, the new paper is based very heavily on Lindzen & Choi (2011) which has been heavily criticised. Lindzen & Choi (2011) used sea surface temperatures and satellite measurements of the TOA flux to try and determine the feedback response. They essentially concluded that the non-Planck feedbacks were 0, or negative. However they only considered the tropics (20S – 20N) and then assumed that the non-Planck feedbacks everywhere else where 0.
The new paper tries to present a slightly more sophisticated model in which it considers two zones, but given what it assumes for the non-Planck feedbacks, it’s no surprise that it returns an effective climate sensitivity (EfCS) of about 1oC; that’s close to the value for the the non-feedback climate sensitivity. It’s fairly straightforward to illustrate why this result doesn’t make sense. Even though it is a two-zone model, it is still fundamentally an energy balance approach. You can write the basic energy balance formalism as
where 
Today 
which then gives an EfCS of about 2K. Of course, these numbers are ballpark figures and there are uncertainties that should also be considered, but it is still hard to justify an EfCS close to 1K. Essentially, we’ve already warmed by almost 1K, have yet to double atmospheric CO2, and still have a planetary energy imbalance that is probably greater than 0.5 Wm-2. How, then, can the EfCS be 1K? I’ve emailed the author to ask him this question. I’ve yet to get a response.
...and Then There's Physics 
ha thanks..
I just finished reading the paper. what a waste of time
Steven,
I thought I’d read it so you wouldn’t have to 🙂
I saw the link over at Lucias and read it. All the references to Lindzen were a dead giveaway.
Actually, for a while I thought LC referred to Lewis & Curry. It took me a while to realise that Lewis & Curry wasn’t published in 2011, and that this referred to Lindzen & Choi.
“Essentially, we’ve already warmed by almost 1K, have yet to double atmospheric CO2, and still have a planetary energy imbalance that is probably greater than 0.5 Wm-2. How, then, can the EfCS be 1K?”
Even reality disagreeing with contrarians doesn’t dent their confidence. And it takes some serious gall offering up Lindzen & Choi (2011) given all the inherent flaws in that paper, especially the cherry-picking of time periods to come up with their result.
Rob,
It is impressive, in a slightly depressing way.
That Climate “Skepticism” site seems really weird.. Including some fantastic claims on Michael Mann’s recent retirement from science.. I am starting to slowly suspect that there really is an infinite number of parallel universes, some leaking over every now and then…
The Climate “Skepticism” site is utterly bonkers, as far as I can tell.
I’ll have to edit the prepublication version to remove the watermark first (I find them distracting, along with the preprint formatting or lack thereof), but was wondering what weight a toy model is supposed to be assigned in estimating ECS. It’s 2016, after all, not 1996.
I’m not looking forward to slogging though 57 pages, but I just know my friendly neighborhood deniers will be all over this paper like flies on… honey. Not that they’ll have read any of it. They’ll just take their cues and talking points from WUWT or some other denier blog.
Magma,
A toy model where the uncertainties seem to come from the range of guesses, and that produces a result that seems completely at odds with reality should probably be given no weight.
“Actually, for a while I thought LC referred to Lewis & Curry. It took me a while to realise that Lewis & Curry wasn’t published in 2011, and that this referred to Lindzen & Choi.”
the climate feedback parameter is the giveaway
Dr. Dessler’s response is concise and would normally put an end to discussion.
Its short enough to exerpt..
Well, yes.
Indeed.
“I’m not looking forward to slogging though 57 pages, but I just know my friendly neighborhood deniers will be all over this paper like flies on… honey.”
if you read papers non linearly you’d be surprised how much faster it goes.
Read the Introduction.
Read the conclusion.
Since he has a couple parameters he wants to vary, then find those sections and when you see what he did, the game is over.
Steven,
Yes, that’s roughly what I do, well in terms of how I start. First work out what they’re trying to do, and then what they’ve conluded. Then you spend some time working out how they did it.
Over the past couple of years I/we have written to Prof. Bates on three occasions for explanation of his contrarian climate science views as expressed in The Irish Times and in public presentations. To date, other than two brief notes – one to say he had no time and one, last year, to say he would reply at some point – we have received nothing at all of any substance. Oh except to a colleague to whom Ray Bates gave a GWPF document as a reference.
For someone in line for global fame if he was to be correct in his ECS claim the lack of engagement does not say much for his confidence in his own modelling, despite his critiques of other climate models.
He does not seem to have learned much from the IPCC AR5 responses to his LC09&11 focussed comments either: 0-16 to 0-19 here:
Click to access Ch00_WG1AR5FOD_RevCommResponses_Final.pdf
Our state broadcaster here in Ireland repeatedly insists on having him on to talk climate. His main talking point is that climate change is “not a planetary emergency, it’s a very serious threat and it justifies action to cut back emissions but it is not a planetary emergency that would require countries like Ireland to seriously damage their national interest”. On the same programme on Dec 3 last year he also said “The climate models are at the moment giving a warming, a rate of warming that is about three times what has been observed in practice”, seemingly channelling Christy, as discussed here 2 posts back.
The real question is surely: how can an AGU journal, even an off-topic, open review one, come to publish an article that is so clearly flawed?
Paul,
I didn’t realise he was quite that prominent in Ireland. I haven’t had a response to me email yet and I suspect, given what you’ve said, I shouldn’t expect one.
I don’t know. It is very odd because you’d expect any relevant reviewer to question the suggestion that EfCS ~ 1K. In addition, his error analysis appears largely non-existent, so it’s also poor on many levels.
> the inverted commas are necessary
You could say “cliscept” – after all, the collective indeed shortens scepticism.
“cliscept” – yes, I like that.
aTTP,
I wouldn’t say he’s that prominent, it’s more that media editors here give him a platform apparently for purposes of ‘balanced debate’. Oddly, enough actual prominent climate scientists are not very interested in appearing on the same panel due to the false balance problem that our media here don’t seem to have heard about.
So they end up having Bates as ‘the scientist’ claiming “I follow the IPCC report and when I speak as a scientist I am not giving personal opinions” immediately after and before giving a slew of personal opinions. Opposite Bates they’ll have politicians and NGOs trying to counter Bates on climate sensitivity while supposedly the programme is discussing mitigation of Ireland’s relatively enormous ruminant emissions.
It’s all a bit silly.
When you make a calculation, you should make a sanity check, does the number make sense? I thought that was one of the first lessons for our students. In this case the number clearly does not make sense in the light of what is already known.
If you decide to continue, it is on you to explain the differences and motivate why your method is more accurate than previous work.
Climate Sensitivity (no quotes)
http://agupubs.onlinelibrary.wiley.com/agu/search/?journal-doi=all&q=Climate+Sensitivity
(citations) Publication Title
(8455) Journal of Geophysical Research: Atmospheres
(3901) Geophysical Research Letters
(2240) Journal of Geophysical Research: Oceans
(2067) Water Resources Research
(935) Global Biogeochemical Cycles
(874) Paleoceanography
(655) Journal of Geophysical Research: Biogeosciences
(522) Eos, Transactions American Geophysical Union
(493) Journal of Geophysical Research: Solid Earth
(442) Journal of Geophysical Research: Earth Surface
(342) Journal of Geophysical Research: Planets
(305) Reviews of Geophysics
(262) Journal of Advances in Modeling Earth Systems
(261) Geochemistry, Geophysics, Geosystems
(212) Journal of Geophysical Research: Space Physics
(96) Radio Science
(63) Journal of Geophysical Research
(58) Chinese Journal of Geophysics
(52) Tectonics
(41) Earth’s Future
(14) Earth and Space Science
(14) Space Weather
(3) Space Weather Quarterly
22307 Results
The other 13 papers in E&SS don’t appear to deal with climate sensitivity … oh wait …
“Climate Sensitivity”
http://agupubs.onlinelibrary.wiley.com/agu/search/?journal-doi=all&q=%22Climate+Sensitivity%22
(citations) Publication Title
(766) Journal of Geophysical Research: Atmospheres
(591) Geophysical Research Letters
(170) Paleoceanography
(135) Journal of Geophysical Research: Oceans
(112) Eos, Transactions American Geophysical Union
(79) Global Biogeochemical Cycles
(70) Journal of Advances in Modeling Earth Systems
(62) Reviews of Geophysics
(57) Water Resources Research
(25) Journal of Geophysical Research: Biogeosciences
(22) Geochemistry, Geophysics, Geosystems
(18) Journal of Geophysical Research: Earth Surface
(14) Journal of Geophysical Research: Solid Earth
(12) Earth’s Future
(9) Journal of Geophysical Research: Space Physics
(2) Journal of Geophysical Research: Planets
(2) Earth and Space Science
2146 Results
I give the abstract a rating of four (well EfCS is greater than zero, so maybe a three).
O’Corollary: Climate change is a planetary emergency that requires countries like Ireland to seriously damage their national interest; just cutting back emissions isn’t enough.
Well that much used Climate “Septicism” graphic has a rather rich contrarian history …
Dr. David R.B. Stockwell
http://scienceblogs.com/stoat/2015/11/23/cagw-rears-its-ugly-head/#comment-54729
Vinny –
==> …to seriously damage their national interest;
Do you self-describe as a skeptic?
To date, other than two brief notes – one to say he had no time and one, last year, to say he would reply at some point – we have received nothing at all of any substance — Paul Price
Dr. Bates may have to step up his pace for the ensuing Comment and Reply session. I’m also mildly curious as to who covered the ~US$2300 publication charge. I don’t imagine as an essentially retired adjunct professor that he has much in the way of research funding.
NYtimes article:
‘Dr. Lindzen acknowledged that the 2009 paper contained “some stupid mistakes” in his handling of the satellite data. “It was just embarrassing,” he said in an interview. “The technical details of satellite measurements are really sort of grotesque.”’
@-Vinny Burgoo
O’Synthesis;- Climate change is a planetary emergency that causes countries like Ireland serious damage to their national interest; It requires action to cut emissions.
I came across Bates as a medal winner at the EGU a few years ago. He gave a rather tedious lecture on how different definitions of “feedback” were used in the scientific literature.
IIRC he has also written a paper showing how a positive forcing can give rise to a cooling – you just have to have a sufficiently inhomogeneous forcing and local sensitivity. Split the globe into two halves, impose a small negative forcing in the region with a high sensitivity, and a larger positive forcing in the region with low sensitivity. Hey presto, the net forcing is positive but the net temperature response is negative.
At the time, he didn’t sound like a sceptic, just an irrelevance. But it was years ago (just checked – 2009). Perhaps if you ignore someone for long enough, they turn up their rants to 11.
James,
Some on social media seem to be suggesting that having won that medal that he is beyond reproach and must be taken seriously. What that mostly indicates is that they don’t really understand how these medals are awarded.
Professor Bates has “form”, as they say, and has published articles arguing against “alarmism”:, and for “pragmatism”. To give him his due, he is less strident or overtly political than the likes of Lindzen.
http://www.irishtimes.com/opinion/climate-action-preoccupation-with-level-of-farm-emissions-is-misguided-1.2268409
It has to be said, though, he has shown definite signs of going completely emeritus in recent years.
Off-topic
@ATTP: This is — of course — your blog and your rules and you are free to give whatever consideration you wish to the following comment.
Richard Tol and discussions related to his work or social media commentary have hijacked a number of posts here. I’ve made my personal opinion clear on the value of such threads; those of other readers’ likely differ. However his disgraceful, personal attacks on Lewandowski and Cook on Twitter today, and his gloating about how he enjoys winding up you and others at this site crossed a clear line.
I’ve previously compared Tol, not entirely seriously, with Luboš Motl. But the similarities are growing uncomfortably close. I respectfully suggest that until Dr. Tol apologizes to — in particular — John Cook, that you consider giving him an enforced hiatus at this site and allow him to work out his personal demons or obsessions elsewhere.
Lewandowsky, not Lewandowski. Typing in anger isn’t conducive to accurate spelling.
James Annan writes:
Isn’t this similar to what the Hansen et al ‘Ice melt’ paper proposes, but instead of inhomogeneous forcing Hansen et al propose inhomogeneous feedbacks where in the more extreme scenarios the net forcing is positive but the net temperature response is negative?
“‘Dr. Lindzen acknowledged that the 2009 paper contained “some stupid mistakes” in his handling of the satellite data. “It was just embarrassing,” he said in an interview. “The technical details of satellite measurements are really sort of grotesque.”’
grotesque captures it.
Re: Magma’s and toby’s observations – is this MR-itis epidemic endemic to academics?
Final thought – linking to the appropriate post in RT’s tweet might help illustrate the threadJAQ and his resulting jazz hands.
@wheelism: No, I suspect they are just more prone to writing and giving public presentations, and their utterances are given more attention. Certainly if you are an academic of a certain age and limited talent, energy or productivity, this is one of the few approaches guaranteed to draw a spotlight.
I’ve noticed this climate contrarian trait tends to emerge at late career stages or in retirement among a subset of engineers, chemists, geologists and even accountants. My completely non-peer-reviewed hypothesis is that some aging conservative males pick up a post-retirement hobby with a numerical/technical flavor (incompetent though they may be) where they can commune with like-minded cranky and increasingly conspiratorial fellow ‘skeptics’.
Attitude-wise it’s a combination of ‘get off my lawn’ and ‘the world is going to hell’ and a closing mind. I saw a similar process at work in an older relative who had been a very successful mechanical engineer in his working career, and it was truly painful to watch his facilities diminish. (In his case he had little interest in climate, and his hot-button issues took other forms.)
Whether this is something that will recur with a newer generation is an open question, but I suspect it’s a one-time event. Climate skeptics are not renewing their numbers.
linking to the appropriate post in RT’s tweet might help illustrate the threadJAQ and his resulting jazz hands — wheelism
I deliberately chose not to do so. DNFTT(ol).
Magma,
It is tempting and I do regard Tol’s behaviour with respect to John Cook and Stephan Lewandowky unbecoming of a serious, and senior, academic. I have, on occasion, deleted his comments if they are on a post in which he is not mentioned. I might start doing so again. I should probably check what he’s gone and said this time, but I’m quite enjoying my Saturday evening, so may check tomorrow.
Joshua: A climate sceptic? No, not for years. It’s a term that’s become meaningless except as a tribal cockade/stigma and I am, of course, above all that tribal stuff.
So what’s your beef, Vinny? Why do we see you here?
I engage in Socratic dialogue
You ask questions
He JAQs off
I asked you a straightforward question, Vinny. A straightforward answer would be appreciated.
Anders: “Essentially, we’ve already warmed by almost 1K, have yet to double atmospheric CO2, and still have a planetary energy imbalance that is probably greater than 0.5 Wm-2. How, then, can the EfCS be 1K? I’ve emailed the author to ask him this question. I’ve yet to get a response.”
How are we certain the pre-industrial energy imbalance was .1C? This necessary to complete the equation, right? Also, your assumption of 1K warming is not accepted by 97% IMO. And I would place the beyond a doubt consensus at > 0.6C. How do you know imbalance next year? Does not El Nino take away imbalance temporarily? Doesn’t anything places aerosols in the stratosphere reverse imbalance?
Uncertainty cuts both ways; it preserves unlikely models you like and unlikely papers you don’t like.
Ron,
We aren’t, but the values I’ve seen are between 0.08 and 0.15C. Also, if it were much higher, we’re probably expect to have seem some more signficant warming then.
I said “almost”.
Okay, using values that lie within the uncertainties, explain how we can have warmed as much as we have, emitted as much as we have, and still have a positive planetary energy imbalance. [Edit: I meant to add “and have an EfCS of ~1K.”]
Good answer, Vinny.
So then, how convinced are you that action stronger than just cutting emissions to address climate change will damage national interest?
Anders: “…the values I’ve seen are between 0.08 and 0.15C. Also, if it were much higher, we’re probably expect to have seem some more significant warming then.”
I submit the degree of imbalance at pre-industrial had to be significant ocean temperature was near a 1000-yr minimum, according to Oceans2K (2015). And, our thermometer record of SST for 160 years ago is non-global, (to be kind,) and highly susceptible to bias (to be blunt).
Anders:
I don’t think the percent those who are informed and only look at the land record, (considering worry of UHI, micro-site contamination and homogeneity adjustments,) without consideration to the divergence of the satellite records, is only ~97%. When need a survey on that.
Ron,
I don’t think your argument for why it should be large in pre-industrial times is very strong. If it were large we should have seen substantial warming in that era.
I’ve no idea what your issue is about my suggestion that we’ve almost warmed by about 1K. It seems pretty self-evident.
Actually, the Oceans2K paper had the SST at a 2000-yr minimum. The full paper pdf is hereas the last line item of 2015. The conclusion for the falling temp was geologic increase in frequency of major volcanic eruptions.
My argument is that we did have substantial warming compared with what would have been the low level of CO2 emissions pre-1950.
For the first 100 years (1850-1950) we only went from 280-310 ppm CO2.
Ron,
Going from 280ppm to 310ppm is still a change in forcing of about 0.5W/m^2.
Ron Graf
I haven’t checked this, but if correct, it would presumably reflect the gradual cooling expected toward the end of an interglacial. How would a slow (multimillennial) orbitally-forced process result in a significant radiative imbalance?
You’d expect the ocean to be cooler than it was 10ka because of orbital dynamics. What you wouldn’t expect is for it to start warming up again relatively fast.
> I asked you a straightforward question […]
Then it should not be too difficult to answer it yourself, BBD.
Why are you still here?
Ron Graf writes: “(considering worry of UHI, micro-site contamination and homogeneity adjustments,)”
Is a clearer signpost needed to warn of the futility of further discussion?
Willard
I remain curious about the exact nature of Vinny’s concerns. Among other things, of course.
By chance you said “among other things,” dear BBD, otherwise I’d worry that you might disappear if Vinny ever answer your question.
‘We’, BBD? Isn’t that a bit tribal?
(Too abstract for me, Joshua.)
Fair enough, Vinny
BBD:
The Oceans2K paper specifically eliminated orbital solar (M-cycle) and solar variance (Maunder Minumum, etc…) as the predominant cooling influence and came to volcanic by using computer model analysis. This added several years to the paper’s publishing as well as co-authors. Steve McIntyre criticized that it was an unnecessary delay in publishing, implying that if their results had gone to plan and produced a clear hockey stick it would have been published like five years ago.
Ron,
None of that suggests that there was a large planetary energy imbalance in the mid-1800s.
> Too abstract for me […]
Speaking of abstractions, Vinny, what’s that me?
> implying that if their results had gone to plan and produced a clear hockey stick
If Vinny appreciates counterfactual squirrels, AT’s the place to be today.
The paper’s press headline was the oceans warming in one centurty had erased 2000 years of cooling. But we are in a lull for major volcanoes for the last 200 years. We have had nothing like 1883 Krakatoa or 1815 Tambora (3X as large) since, well, back then.
Ron,
Again, this does not imply a large planetary energy imbalance in the mid-1800s. This is most likely because we’ve produced a change in external forcing of around 2.3W/m^2.
“Ron,
None of that suggests that there was a large planetary energy imbalance in the mid-1800s.”
What would be your type of evidence if not ocean temp?
Ron G
So if we start off with a cooler world, then subsequent CO2 (and other anthro) forcing is a bit more efficacious than we thought? Since it got us where we are with a bit of an initial handicap.
Ron,
1. In what way is temperature evidence for a planetary energy imbalance?
2. Sea level rise is better evidence for a planetary energy imbalance.
The subtle knife 🙂
By chance I looked up from the spade to the narrowing circle of daylight and decided to add a qualifier. It was, as you guessed, a close-run thing.
BBD:
My analysis is if the Ocean2K paper is correct, and I’m not saying I think highly of them, but let’s say they were correct, the volcanic aerosols from about 1200 to 1850, temporarily threw Earth into an imbalance forcing cooling. Once the aerosols clear and there is no more large volcanoes the Earth is left in an imbalance in the warming direction. I’m not saying it was huge. But if it was plausibly .4C rather than .1C then even if GISSTemp (GMST record) is correct then low ECS is possible. And, if the GMST record is biased high also then sensitivity is very very weak.
All I am saying again is that uncertainty cuts both ways. Nobody is happy.
Source. Data from Kopp et al. (2016)
Ron,
Even if you make it 0.4W/m^2 (which I think is implausibly high) you’d still have trouble trying to justify an EfCS of ~ 1K.
Anders:
1) Relative sea temperature to the atmosphere determines imbalance. If you would like to say the sea was cooling from the Holocene Optimum that is fine. That is not what the paper said.
2) Sea level has been rising since the YD 12K years ago. Polar ice melt is not a good proxy for imbalance (IMO) due to long periods and the slow pace of melt. Perhaps accelerations in SLR could be evidence but I don’t know of any undisputed evidence of that currently.
BBD, remember even if you chart is accurate it would be explained by the Oceans2K paper without GHG.
Ron,
No, that’s bizarre. The SST is a proxy for surface temperature.
The rate of sea level rise is likely an indicator of the planetary energy imbalance. That’s the point.
Ron Graf — as a believer in the ‘one true graph’ — may appreciate this; though it doesn’t exactly jibe with his current concerns (though they’re not even that current, having been aired last year at WUWT and over at Ed Hawkins’).
Ron G
I missed the bit where O2K says the C20th SLR has nothing to do with CO2.
Ron, your argument linking the PAGES to a putative reduced ECS is weakened by several bits of evidence:
1. analysis of millennial Arctic cooling [Kaufman et al, (2009) Science 325, 1236] indicates that this was orbitally-forced (and orbital forcing on the millennial timescale must have had a cooling contribution from which we cannot have “rebounded” from). Although PAGES interpreted their cooling data as being largely volcanically-forced it’s unreasonable to assume that some component of the apparent ocean cooling wasn’t orbitally-forced.
2. There were humungous volcanic eruptions in the early 1800’s but the ~50 years up to the Krakatoa eruption had low volcanic forcing and this is a long period in the context of the recovery of surface temperatures to the effects of volcanic forcings. We can see the effects of the late 19th century eruptions on the temperature record (e.g. in the NASA Giss and Hadcrut global temperature reconstructions). However if we use the pre-Krakatoa temperatures as a baseline for comparing with current temps, we’re still over 1 oC warmer now (i.e. comparing current temps to a time when we’d expect most of volcanic forcing to have been recovered from).
3. If we wish to consider negative volcanic forcing as potentially suppressing surface temperatures in the mid/late 19th century (as a baseline from which increased surface temps are measured) then we should also consider volcanic forcing in the current period. In fact negative volcanic forcing in the period ~1960-2000 is of the same order as the late 19th century negative forcing (see e.g. Fig 8.18 of IPCC AR5 assessment report chptr 8). Overall [and this is supported by attribution studies – see e.g. Lean and Rind (2008)] – the contributio of volcanic forcing to warming from the mid/late 19th century seems to be small…
Volcanoes: SIgl, et al (2015), a revision copy here with graphs on p.38, including PAGES2K temperature series and volcanic forcings.
Here’s a short version that describes the work.
Of course, volcanoes are spiky random noise overlaid on Milankovtich cooling, plus the CO2 drop into 1600AD, as per Ruddiman.
Oneilsinwisconsin, thank you for the graph. I think this supports the claim of increased eruptions from 1200-1850. And, further, it demonstrates a quiet period before and after looking at the histogram. The plot seem it should go down after 1850. I don’t get it.
Anders, I revise my claim on defining imbalance to: that century scale it can be inferred by the lag of the top 1000 meters of ocean adjustment to forcing. On the decade scale the imbalance is inferred by the atmosphere relative to the sea surface only.
“The rate of sea level rise is likely an indicator of the planetary energy imbalance. That’s the point.”
Perhaps, but it is more controlled by M-cycle, millennial scale, imbalance. It can also be affected by polar precipitation patterns and ocean currents, for example, as in west Antarctica. Glacial melt caused SLR is a poor proxy for imbalance unless you can also correctly model for ocean expansion due to warming. There is a lot of confounding variables to even be able to measure SLR due to continental subduction and from aquifer draining and movement from isostatic rebound, both on the coasts and on the sea beds.
> The paper’s press headline […]
No more “but the media” deflection, please.
Thank you for your concerns.
Ron Graf: ” …if your chart is accurate it would be explained by the Oceans2K paper without GHG.”
BBD: “I missed the bit where O2K says the C20th SLR has nothing to do with CO2.”
Correction: if your chart is accurate it could be explained by the Oceans2K paper without GHG. Think about it; the glacial trend would react to the same cooling that is forcing the SST to cool. Also, the cooler sea would contract, both effects to slow or half SLR temporarily, leaving a rebound in the pipeline.
“Ron Graf writes: “(considering worry of UHI, micro-site contamination and homogeneity adjustments,)”
Is a clearer signpost needed to warn of the futility of further discussion?”
###############
Ron’s “reasoning” goes like this ( from Lucia’s site) paraphrasing (too lazy to link)
The temp has increased 1C
Some people think 1/3 of the land record increase is UHI
Therefore .65C
###########
What he forgets is that the land is 1/3 of the total.
That said its a fair enough sensitivity analysis to do .
If the Land has warmed 1C and SST has warmed 1C.. and 50% of the land record is bogus
UHI…. then….. well you still end up with .83C warming or so.. hmm. so much for UHI importance, even THEORETICALLY…
Long ago I had to ask myself why was UHI important? And the answer was… Delta T !!!! is important for sensitivity.. And then when I didnt find any UHI.. I slapped my forehead and said..
oh shit mosher… even if UHI is big it doesnt change the sensitivity answers in an interesting way!
doh!
So the temperature record ceased to be a wedge issue for me.. and just became a technical past time. something cool to work on.
Re: Volcanoes; I’m having déjà vu all over again. If I’m correctly reading Fig. 4 d in McGregor et al. (2015), Robust global ocean cooling trend for the pre-industrial Common Era, the linear trend line indicates about 0.2 °C/millennium of SST cooling is attributable to volcanic forcing over the interval 800-1800 CE.
According to HadISST1, the warming trend is 4.0 °C/millennium over 1870-2015, 6.8 °C/millennium over 1950-2015 and 9.1 °C/millennium over 1970-2015.
I don’t know about anyone else, but over one whole order of magnitude difference pretty much sews up this one for me.
There is a few reasons I would say the land record is more important than the sea in proportion to its sample of the Earth. It shows more warming for one, and thus less measurement uncertainty component although still high. Land temps have a broader more dispersed sampling field (before Argo). Until late in the 20th century we weren’t looking for global sampling of either. But when trying to infer a global sea temp 150 years ago by simply having a sporadic bucket check on the usual seal lane its ripe for bias (intentional or not). The land record is also possibly biased but there is more potential for being able to investigate and uncover UHI because it may have a unique fingerprint in diurnal temperature range trend, etc…
If the land record is found to have been biased the sea record I think follows or else the models fall apart, and, well, that’s another story.
Steven Mosher,
And perhaps that SSTs are adjusted down. But Karl …
Speaking of SSTs, penny for your thoughts.
That’s all for today. I’ll let Anders go to bed.
Steven, I know I need to read more but I honestly hope to perhaps add insight to your long-sought goal of finding a novel investigation method to scope UHI and/ or micro-site.
As you admit, UHI is real acknowledged thing by the consensus. Your position is that it does not matter. That needs proof. And I don’t understand your’s or Menne’s. How can using the BE scalpel find a gradual low frequency effect as systematic UHI bias would be hypothesized to be?
Ron Graf writes:” I think this supports the claim of increased eruptions from 1200-1850.”
Ron, so your hypothesis is that the increased volcanic injections at 1450 caused almost immediate and sharp *warming* over the next 50 years and likewise the volcanic injections between 1600-1650 caused another bout of *warming* and (the only one you noticed) that more volcanic injections in the 1800’s caused yet another bout of warming?
You really should have followed the links John Mashey provided; especially PAGES MAGAZINE ∙ VOLUME 23 ∙ NO 2 ∙ December 2015
Now, what was that you were saying again?
That urban areas are warmer than rural ones is not disputed. That they are warming faster than rural ones is another question and the data from urban and rural weather stations casts doubt on the proposition.
Asserting that there is an UHI effect is not should not be asserting that there is a difference in trends except for those trying to pull the pea out of the cup without being seen.
Pingback: Cultura all'italiana e altro blog-crossing - Ocasapiens - Blog - Repubblica.it
[No need to be sorry – RonG. Just resubmit without the “but press release.” -W]
[No mediatic squirrel chase, please. -W]
I do hope you realize that your 100-200 year lag is simply not possible with volcanic injections. That was my point. They have an immediate effect that diminishes rapidly. From Hansen et al 1978
Note those are tropical tropospheric temperatures, the actual GMST drop was about half that.
Paul Price (@swimsure):
Media consumers can’t be expected to have comprehensive scientific literacy, but they do need some scientific meta-literacy:
Yeah, I know, in a perfect world… In any case, viewers ought to wonder how a talking head’s claim to scientific authority makes him an authority on their interests. Even if Ireland is a more unified country than the US is, one presumes that what the national interest is depends on which Irish person one asks.
[Playing the ref. -W]
BrandonRGates says: “McGregor et al. (2015), Robust global ocean cooling trend for the pre-industrial Common Era, the linear trend line indicates about 0.2 °C/millennium of SST cooling is attributable to volcanic forcing over the interval 800-1800 CE.
According to HadISST1, the warming trend is 4.0 °C/millennium over 1870-2015, 6.8 °C/millennium over 1950-2015 and 9.1 °C/millennium over 1970-2015.”
Comparing the last 100-year trend to the 2K by extrapolation ignores random fluctuation. This is similar to an investor’s fallacy that his stock wiped out a year’s worth of gains in one day’s loss. This tells us nothing about the normal trading range of stock which is critical for making any conclusions. McGregor et al could not make any observation of the common era frequency greater than their 200-year bins. And, this was then dampened by using proxies, (speculative ghosts of temp,) not live thermometers. On top of the proxy dampening and statistical dampening there is actual dampening by the ocean’s heat capacity and mixing.
There is land based evidence, ice cores and documentary history, of large fluctuation, (LIA, MWP, RWP, etc…).
Ron g.
Let’s recap because you failed to close on and issue you raised.
To recap.
1. Anders used 1c.
2. You argued that maybe this should be as low as .6
3. Your reason was. Uhi…bias..etc.
4. I point out that the land is 1/3 of the total.
Now.. Rather than closing on the issue as you should you shift to the sparse sample argument. Here is the problem with that. The sparse stations or sparse ocean temp records. .have uncertainty that goes two ways..hotter or cooler.
So anders suggested 1c
NOW you want to argue but uncertainty and suggest .6c.
But uncertainty has two sides. .which means.
1.4C is also a possibility.
Just start over and suggest .9c due to uhi.
Ron,
I’m no longer following what it is that you’re trying to suggest. The key point is that it is very difficult to see how the ECS could be as low as ~1K, given how much we’ve warmed, how much we’ve emitted, and that we still have a positive planetary energy imbalance. Do you disagree with this?
> Rather than closing on the issue as you should you shift to the sparse sample argument.
If only.
Eli says: “That urban areas are warmer than rural ones is not disputed. That they are warming faster than rural ones is another question and the data from urban and rural weather stations casts doubt on the proposition.”
I don’t know if anyone is claiming that urban areas are warming faster than rural. I’m not.
Are you claiming that all urban areas are just as urban as they were 100 years ago? Are even rural areas as rural as they were 100 years ago? If not, then the gradual transition placed systematic bias to the temperature record. There is no other conclusion to be drawn. Making things worse, once realizing the UHI issue the first remedy was to move urban stations to airports thought to be more rural (and adjusting for it). Then urban sprawl and airport growth simply repeated the skew of bias a second round.
Oneills, Hansen’s assertion was regarding atmosphere. The oceans lag the atmosphere. McGregor (2015) spent extra years doing modeling to isolate the cause of the cooling to primarily volcanic influence using GCM analysis. I am not saying McGregor is the last word on 2K reconstruction. I am glad we have common ground that pre-industrial sea temps were at a post ice-age low. Do you not believe there was variability of like .5C per inflection inside the common era? What do you believe the variability was and why?
Steven says: “…NOW you want to argue but uncertainty and suggest .6c….”
As in a courtroom the burden of proof lies with those who make the original assertion, (Anders’ post,) not the examiner (me). I should have the freedom to provide as many alternative explanations as I please. And, each alternative should be tested for validity independently. There is no need to couple them for consistency. A defense attorney could point out that the real perpetrator could be any one of three other people. He doesn’t need to choose one. Thus uncertainty here cuts against the one making the original claim.
My point of the sparseness of the 100-year-old sea surface record is that it would make it all the more easily biased AND uncertain. Thus, if the land record was proven to be biased by 30% the sea records would have to be assumed to be biased or the models are completely bogus. You might say that the models could be corrected with a higher SST ratio to LT, but monster uncertainty does not help those making a scientific claim.
Ron,
No, it does not. Science doesn’t work on the “innocent until proven guilty” idea, largely because we can’t define what innocent is. There is ample evidence to suggest that the surface has warmed by about 1K since the mid-1800s. If you want to sugget otherwise, the burden lies with you to show why it could be quite different to that. Simply peddling various possibilities isn’t really good enough.
Ron Graf writes: “Do you not believe there was variability of like .5C per inflection inside the common era? ” Huh? Have you actually considered the thermal capacity of atmosphere versus that of the oceans? The oceans are going to change far *less* than the atmosphere – not the other way around.
Anders says:
There are essentially three independent possibilities that could lead to 1K EfCS.
1) The surface thermometer record is biased and the satellite record, for example, is more representative of a high GHG emission trend.
2) There was a significant negative imbalance in the pre-industrial interval that acted to accelerate (sensitize) the hydrosphere rather than the record (if accurate) reflecting internal sensitivity to the weak positive forcing of GHG provides by itself, (1.1C/ doubling GHG).
3) A combination of 1 and 2.
Ron,
Okay, so if we’ve warmed much less than virtually all teams who study this think and if there was a substantial planetary energy imbalance in the mid-1800s, maybe then the EfCS could be 1K. Lots of ifs and how likely is this? I would say pretty unlikely, given that at least 5 teams who produce surface temperature datasets would need to be wrong, as would our estimates for sea level rise in the 1800s, and our estimates for changes in external forcing during that period too.
Oneils: “Huh? Have you actually considered the thermal capacity of atmosphere versus that of the oceans? The oceans are going to change far *less* than the atmosphere – not the other way around.”
My question was your opinion of paleo GMST fluctuation max and min amplitude in the last 2K years. GMST is defined at atmosphere at 2m over land plus SST. You’re overthinking the question.
Anders says: “…Science doesn’t work on the “innocent until proven guilty” idea, largely because we can’t define what innocent is.”
I never said innocent or guilty. I did not say liable or not liable, which is another test. You bring up a good point though; what is the test? I think the test should be commensurate to claim. If there is a prize or action compelled from the proof, thought, there must be a test. The burden lies with the entity making the claim for alteration the status quo.
For a drug company to market a drug in the USA they must prove safety and effectiveness beyond question. The drug maker could say there is a good theory that it should work and it even worked in mice. That doesn’t work the same reason government should not act cautiously before intervening in any markets.
@Ron:
For a fossil fuel company to market a pollutant they must prove safety and sustainability beyond question! Again, you’ve got all your analogies dead wrong (not surprisingly).
Btw, we know very well what the imbalance pre-industrial was. Go and read the literature before you make more ridiculous suggestions. You may also want to learn how the climate system actually works. You seem pretty clueless judged on your postings here.
Ron,
You really should learn how to close html tags 🙂
Scientists are not trying to market AGW, they are presenting evidence as to the consequences of continuing to emit CO2 into the atmosphere. We can choose to ignore this and demand stronger levels of proof. That won’t change what actually happens as we continue to emit CO2 into the atmosphere.
Ron – if you were discussing *atmospheric* fluctuations – well of course we’ve seen that size fluctuation – we’ve seen similar in our lifetimes. But that makes your earlier comment even more unintelligible than it was when I first read it. We have *not* seen a 100 or 200 year lag on the inflection – which would basically be either physically impossible or have to be so large as to wipe out all life.
Anders, it looks like this is the point of the string where we appeal to authority, (ours and the ones that agree with me). I must go because is I reply down this road my comments will evaporate from view anyway. And, belittling my HTML skill… That really hurts… My commas are worse.:)
. It’s by someone called J. Ray Bates . .
See http://www.egu.eu/awards-medals/vilhelm-bjerknes/2009/j-ray-bates/ for context.
Anon,
What context does that provide?
> As in a courtroom the burden of proof lies with those who make the original assertion […]
If we’re to take Popper seriously, RonG, that would rather be the burden of disproof. According to that viewpoint, there’s no such thing as a conclusive proof in science. It’s all about conjectures we try to refute.
That viewpoint is insufficient to explain how scientists proceed In Real Life, i.e.. IRL. IRL, they compare the evidence basis of competing hypotheses and go for the one which looks best substantiated. The best explanation (almost) always win, even in IRL, i.e. Ireland.
You can’t dismiss a hypothesis by mustering a zillion other alternatives. I say a zillion, but it could be more than that – our ignorance knows no bound. Only GROWWWWWTH is “more” infinite.
FWIW Ronny dear, center cities are far less dense than they were in at the turn of the 20th century and weather stations have been moved to outlying areas called airports. Your continued invocation of the urban heat island as a source of warming in the surface record is pretty much fantasy
So?
https://honors.agu.org/winners/richard-s-lindzen/
Peas in a pod, emeritus and wrong.
Oneils, I’m not sure what you are talking about but my question was simple. Even so, I will repeat it for you another way. What do you envision the Tmax-Tmin range is for the common era global surface to one decimal deg C ? While you are at it, how about also for the last 1Ka, 10Ka.
Eli, how do you define density? Cite a UHI study of a city that has trended down in raw observed temp. In Kim(2003) study of Seoul the trend over time was a growing deflection from rural.
Ron,
Do you understand the concept of “temperature anomalies”?
==> …Here is the problem with that. The sparse stations or sparse ocean temp records. .have uncertainty that goes two ways..hotter or cooler.
Only off you assume there isn’t a conspiracy to fool the public and line the pockets of capitalism-hating eco-zealots.
And btw, you have the burden of proof to show that a conspiracy isn’t possible.
Temp anomalies are the deflection from a historical base period. Climate science uses monthly anomalies summed over the year to produce yearly ones. Is this a test? My question to Oneills would be gross range Tmax-Tmin (annual) for 1K, 2K, 10K. Hope that clears it up. I think we can assume the intra-year range to be similar to the modern.
Ron,
Not really, because I don’t understand why you’re asking this question.
Anders, do you have neither knowledge nor interest in potential paleo natural variability range?
Ron,
Ahhh, your 1K, 2K, 10K were years, not temperatures?
I don’t have any specific knowledge. My understanding is that this variability is indeed of interest and is one reason we would like paleo reconstructions to have the highest resolution possible. However, I don think that there is much evidence for long timescale internally-driven variability, at least not in the last few thousand years.
Ron – now you’ve lost me completely. Suddenly, after more than two dozen posts, you start talking about Tmax and Tmin — though using them in a completely non-standard way. If you simply mean what are the highs and lows we’ve seen over that time, probably 1.5C for the last 1ky and 2ky periods and 2C for the last 10ky.
” If not, then the gradual transition placed systematic bias to the temperature record. There is no other conclusion to be drawn.”
1. It’s unclear and false that every urban location is warmer or warms faster than nearby rural stations.
2. It’s unclear and falso that all urban transitions are graduAl
3. You are assuming the uhi signal is large.
The question started from discussion of Ocean2K McGregor (2015) who reported a net drop of ~.7C in the mean sea temp mostly in the last 800 years. The range of the mean with 200-year bins smooths the trend and, as I mentioned before, the low accuracy of proxies smooths also dampens the anomaly.
The graph on climate audit’s post on this let’s one eyeball the graph here. Although the Y axis is in SI units rather than deg C the range can also be mathematically inferred by the following quote of one of the authors in their press release.
Let’s say the average slope for the anomalies of the last 100 years is 0.82C. McGregor reported in their supplemental section 4 that their average slope in the last 1000 years was 0.41. So you can see how they got their 20X modern slope. This, of course, would only be true if there were not reversals in the ten 100-year intervals. As I said in the stock chart example I gave earlier, one can’t extrapolate daily gains to mean anything about yearly gains.
To infer the degree of pre-industrial positive imbalance (not negative as I wrote before) we have few clues from the smoothed proxy studies about natural (pre-AGW) average century anomalies. Historical records such as Central Europe Temp CET and archaeological evidence suggest similar variability to the last century. But let’s just say it was half. Where did we start, on a high inflection or low? All indications are that it was a low. Again quoting the Ocean2K press release:
http://phys.org/news/2015-08-years-global-ocean-cooling-halted.html
The Ocean2K authors attribute the drop to unusually high volcanic activity. If this is true then in the absence of volcanic aerosols there is an imbalance equal to about half the range of drop from year 1250-1850, or 0.7/2 = .35C imbalance
Steven, here is a copy of my reply to Eli on this at the Blackboard:
Anomaly(urban) – Anomaly(rural) = UHI(u) + Micro-site(u)
Where rural is pristine natural setting for 50+ square Kilometers
The question is not whether the urban and rural trend is different. It’s what the difference in trend would have been if the station surroundings and instrument would have remained a constant. Rural stations today are still in developing areas. Each year dozens of stations cross over the threshold classification denoting rural from urban. The change happens continuously through the urbanization of billions of people over 150 years. Tons of new concrete and asphalt are put down — swamps drained. Rarely does this reverse. All non-porous surfaces and construction add to the effect.
Remember the climate change signal is un-noticiable, tenths of a degree. UHI can show deltas of 4C. Tokyo has seen up to 7C.
Ron Graf writes – “The Ocean2K authors attribute the drop to unusually high volcanic activity. If this is true then in the absence of volcanic aerosols there is an imbalance equal to about half the range of drop from year 1250-1850, or 0.7/2 = .35C imbalance”
What is the average time for EBM recovery per Ocean2K? Looking at Fig. 3c I estimate about 60 years for even the largest eruptions. Eruptions in the distant past aren’t having an effect today. Any imbalance circa 1850 was due to eruptions in the first half of the 19th century. Even if we assume that the ocean still had 0.35C of recovery in the pipeline circa 1850 wouldn’t this recovery have taken place by 1900? Of course volcanism didn’t end in 1850; during the industrial era we’ve had Krakatoa, Santa Maria, Agung, El Chichon, and Pinatubo.
Apparently it needs to be spelled out; episodic volcanic activity can lead to global cooling. Despite continued volcanic events the earth is now warming. This is not even taking into account anthropogenic aerosols.
Ron.
Tokyo was adjusted to remove uhi.
Here is the thing. If the bias is big we see it and adjust for it.
If the bias is small but abrupt we see it and adjust for it.
You are left with supposing that uhi is gradual but too small to be detected.
Thanks.
Like I said. If anders says 1c of warming you might get agreement if you said .9c after perfect uhi correction.
Ron: the thing is, the data from the pristine USCRN network does match the same temperature trend as that from the less-pristine data.
Which shows you that the adjustments are working correctly. And yet, the global version of those adjusted temperatures show warming. So the algorithm isn’t wrong from UHI effect, as far as we can tell. And we’ve put these through a lot of rigorous statistical tests.
It’s pretty much just an assertion that the temperature data is messed up. There’s literally zero evidence to support it.
See note on Alaska, “Over the past 60 years, the average temperature across Alaska has increased by approximately 3°F.[3] This increase is more than twice the warming seen in the rest of the United States. Warming in the winter has increased by an average of 6°F [3] and has led to changes in ecosystems, such as earlier breakup of river ice in the spring.”
Clearly, the population must have grown greatly since the last time we visited. 🙂
Ron Graf says: “Eli, how do you define density? Cite a UHI study of a city that has trended down in raw observed temp. In Kim(2003) study of Seoul the trend over time was a growing deflection from rural.”
If a city grows around a station this will likely produce a warming bias. But such stations are often relocated to other locations. In the Austrian network, for example, there was a cooling bias of 0.2°C because the stations are currently more rural than they were in the beginning.
http://variable-variability.blogspot.com/2016/03/cooling-moves-of-urban-stations.html
Ron Graf says: “1) The surface thermometer record is biased and the satellite record, for example, is more representative of a high GHG emission trend. ”
Personally, I do not put much faith in satellite estimates of temperature changes and you better not do so as well if you want to claim a major bias in the thermometer record that would justify a ECS of 1K.
The temperature trends of the tropospheric temperature (satellite) and the surface air temperature (thermometer) is very similar over the full period for which we have satellite data.
(You may have been confused by pictures shown at WUWT & Co. comparing only recent data where the deviations are larger because the tropospheric temperature responds more to El Nino; that is just looking at noise.)
The temperature trend should be a few percent stronger for the tropospheric temperatures, but that will not get you to an absurd value of 1K for ECS. If you want to go to such values you will have to claim that also the tropospheric temperatures are hugely wrong (and the lake and river freezing times, the lake temperature, the glacier melting, the snow cover decrease, well basically everything).
Again, the warming in Alaska is clearly a form of UHI – Ursine Heat island.
Polar bears (white) move back onto land and evolve back into dark-colored bears, lowering albedo and absorbing more energy. You heard it here first.
That makes as much sense as some of the other arguments 🙂
So the consensus argument against century frequency natural variability producing a low inflection in the mid 1800s and UHI biasing the weather stations and satellites being systematically biased instead is:
1) CET record indicating the LIA is wrong due to CET being a localized sampling and LIA being a European event only. Same for MWP and RWP. But a 1.5C change in Alaska is valid global sample. Got it.
2) McGregor’s years of work and delay in publishing Ocean2K in order to gain attribution to findings was wasted because there is no way their model fingerprint analysis of volcanic forcing was why the SST dropped .7C from 1250-1850. And, it is highly improbable that any other combination of temporary forcing could have caused the drop clearly indicated. The true cause of the drop had to be gradual orbital influence, contrary to the McGregor finding of 0-1000AD interval being a flat trend. And, therefore there could be only little imbalance in the oceans to affect a sensitivity for rebound at the start of the industrial era. Got it.
3) UHI is a real thing but it is handled by moving stations out of the city and making a correction. Not correcting for it would only be a problem in Tokyo. Australia stations are in less and less densely populated area (and they pulled up the concrete as they left), so all in all it balances out globally. Got it.
4) Satellites records, although run by independent teams that largely corroborate each other, one academic (ex-NASA analyst) and one private organization (NASA contractor), they are just wrong; systematically biased over 37 years. Got it.
5) And, foremost, there is no room for bias or politics in academic/government science. There are careful protocols to enforce neutrality while mandating adversarial validation of all findings. Well, I’m not sure you asserted this, but this assumption goes without saying, I guess.
6) There is 97% consensus on all these points. Okay, you got me. I need to read up.
Ron,
1. No idea what you’re getting at.
2. Again, not sure what you’re suggesting. Even if the cooling was partly volcanic, I still don’t see how this is evidence in favour of a large planetary energy imbalance in the mid-1800s.
3. Yes, that might be right.
4. No, I don’t think that has been suggested.
5. Again, there are all sorts of different possible biases. That’s why we don’t trust individual studies, or individual researchers.
6. Hmmm.
There are a couple of odd things with your comments on the McGregor paper, Ron. You’re surely incorrect to say that: “was why the SST dropped .7C from 1250-1850”. It’s more like a 0.2 oC drop isn’t it (their Figure 4 comparing the binned standardized 200 year values covering 1250 and 1850, respectively)? Or perhaps a 0.4 oC drop using the slope data in Supplementary section 4.
Their negative slope in this period is heavily weighted by the very large volcanic events in the early 19th century (see Fig 4 again). But (as also stated earlier on this thread) the sea surface temp is expected to recover from volcanic forcing-induced cooling quite quickly on a decadal timescale. So by the end of the 19th century (pre-Krakatoa) there isn’t expected to be much of a residual SST imbalance.
Although McGregor largely (and you exclusively) interpret the apparent SST cooling to episodic volcanic events, their own analysis indicates significant contributions from (especially) land-use changes (larger than the volcanic contribution according to their analysis in Supp Table S14), orbital and even greenhouse gases.
Since the overall suppression of SST in the period you mention is fairly small (~0.4 oC according to McGregor et al), some of this is almost certainly due to non-volcanic contributions (as McGregor’s analysis indicates), and the volcanic contribution would be expected to be largely dissipated during the relatively volcanic-benign 19th century half-century period pre-Krakatoa, it simply doesn’t seem plausible that there would be much of a negative SST temperature imbalance. And even if there was still 0.2 oC of negative temp imbalance that still leaves around 1 oC of warming in the both the NASA Giss and Hadcrut4 temp since late (pre-Krakatoa) 19th century.
1) Is there evidence of natural variability on the 100yr-1000yr frequency? Is it possible it is as much as 1.5C amplitude?
2) Is there a preponderance of evidence suggesting that the 16-17th century a minimum in atmospheric temp? Would not the SST lag this inflection by 100-200 years? Would this not put the 19th century ocean in a probable imbalance. This imbalance would be about equal to the average anomaly of years 1250-1850 as compared with 0-1250, about – 0.7C. Or, what is your correction in math?
3) Will definitely have to come back to this as I do not accept the logic given by S. Mosher.
4) Then why do you and Victor Venema have little confidence in satellites? I will excuse Steven Mosher since his work brings him in competition with them.
5) Good, maybe we can find common ground for area discipline methods allowing universal confidence to all scientifically submitted results in climate science. Save a lot of money and time.
6) Speak out against “consensus” claims you know untrue and I will do the same for “skeptical” claims.
Thanks for reply.
Ron, on your #4 above, it does surprise that you’re not a little more considerate of potential problems with the satellite MSU tropospheric data. The difficulties with converting this very messy primary data into a temperature (and temperature trend) is exemplified by the fact that the initiators of the methodology got this hopelessly wrong for around 15 years until major corrections in 2005. One only needs to look at the effects of recent updates in analyses [*] to see that the satellite data temperature data is still subject to potentially very large uncertainties in the analysis.
[*] e.g. the March update of RSS data (Version 4.0) resulted in a change in temperature trend from 0.08 to 0.13 oC per decade.
see: http://www.remss.com/blog/RSS-TMT-updated
April
http://berkeleyearth.org/temperature-reports/april-2016/
Land & Ocean
Temperature Anomaly 1.010C 0.061 C (+-) Relative to 1951-1980
Temperature Anomaly 1.372C 0.090 (+-) Relative to 1850-1900
Anomaly Rank since 1850 5 out of 1996 range: 4 – 7 Any 1 month period
Anomaly Rank since 1850 1 out of 167 range: 1 – 1 Same 1 month period
Ron,
1. If you mean internal, then I would say that that is essentially impossible.
2. I don’t think there needs to be a 100 – 200 year lag.
3. Okay.
4. What do you think Victor does?
5. Okay.
6. Personally, what I choose to speak out about does not depend on what others choose to speak out about. Maybe that’s just me, though.
Ron, why would SST lag “atmospheric temperature” by 200 years? By “atmospheric temperature” do you mean “surface temperature”? After all we don’t know the atmospheric temperatures 100’s of years ago! What we do have are measures of surface temperatures (composites of sea, and land from temperature proxies) . But where does the lag come from?
And you didn’t respond to my post a few stops above where I queried your value of 0.7. oC of apparent SST cooling. At most this is 0.4 oC. And not all of this is volcanically-forced according to the McGregor paper you’re sourcing your analysis from…
@ Steven
Yup, a scorcher. Must be all that internal variability.
Ron.
Lets Recap
” Also, your assumption of 1K warming is not accepted by 97% IMO. And I would place the beyond a doubt consensus at > 0.6C.”
1. please NOTE that you made a claim. You now have the burden of proof for that claim.
2. you made the same claim at Lucias
3. The claim is this. The record shows 1C of warming
BUT you think the real number is something like .6 -.65C
4. You claim .6-.65c BECAUSE of UHI and microsite
THAT is the claim YOU have to defend or amend.
There is a whole body of science that shows the UHI is nowhere NEAR .3-.35C
but YOU have made a claim that it is.
Against you claim we have this.
http://berkeleyearth.org/temperature-reports/april-2016/
The OCEAN ALONE with ZERO UHI has warmed around .9C
the OCEAN is 70% of the total.
The land has warmed about 1.5C
Even IF 30% of that were UHI… you’d STILL have 1C of warming
Now, Assume that ALL OF THE 1.5C land warming is UHI.
Calculate the global average..
done?
.63C
So your position REQUIRES YOU to demonstrate that ALL of the land warming is UHI
and Further you have to explain why the land doesnt warm as much as the SST
So you have two choices.
A) Admit that you OVERESTIMATED the effect UHI can have on the global reocord
B) forget you lost that fight and go attack the ocean record
You made the claim. .6C
you own that claim and defend
or
amend
I will not abuse you for amending your claim. I made the same mistake myself.
It was hard to admit I was wrong. But look, nobody on this site trawls through the
old climate audit posts to remind me how wrong I was about certain questions about
the temperature series. NOT ONE.
> The difficulties with converting this very messy primary [MSU tropospheric data] into a temperature (and temperature trend) is exemplified by the fact that the initiators of the methodology got this hopelessly wrong for around 15 years until major corrections in 2005.
You might also like:
Thank ClimateBall ™ God that we had Hans’ graph.
Ron Graf,
So here’s the thing about the difference between climate and the stock market. One is a deterministic system, the other is stochastic. Critical to making correct stock picks is knowing in advance the decisions that millions of fellow market players are going to do. Which is impossible. Over the long term, very few professional stock pickers beat the market index of choice, and practically zero amateur stock pickers do. The ones who beat market indices do so at the expense of the vast majority of losers, some of whom get completely wiped out.
Critical to your concerns about a major volcanic eruption saving us from (C)AGW/CC is that you do not know when the next one will happen. So your precious rational approach to somehow accounting for the possibility is for nought. You cannot give me the critical information we need on either the timing OR magnitude of the adjustment you insist is necessary.
You only think it is critical because you are foolish. You are banking on a ~1/200 chance over the next century to significantly slow down an upward trend which is guaranteed to happen unless GHG emissions are brought to zero. Your luck might be better if you parked all your money in a short position against US Treasuries for the next 85 years and left it there.
You don’t get to do something so thoroughly stupid with our planet.
Wake up.
“1) Is there evidence of natural variability on the 100yr-1000yr frequency? Is it possible it is as much as 1.5C amplitude?
2) Is there a preponderance of evidence suggesting that the 16-17th century a minimum in atmospheric temp? Would not the SST lag this inflection by 100-200 years? Would this not put the 19th century ocean in a probable imbalance. This imbalance would be about equal to the average anomaly of years 1250-1850 as compared with 0-1250, about – 0.7C. Or, what is your correction in math?
3) Will definitely have to come back to this as I do not accept the logic given by S. Mosher.
####################
well, logic is logic. I don’t own it.
WRT #1. Long ago on Climate Audit bender, me and others pondered the possibility of LTP
Basically, could a portion of warming be down to a “unknown” long term process?
Well, you cant rule it out.
For a while that argument made sense to me. it really did..
you cant rule it out. perfect rational philosophical skepticsm.
but.. then there is this.
I cant rule out unicorns.
The Appeal to LTP is really hanging on by the barest skeptical thread. IF you were a scientist and IF you were curious about how the earth works, AND you thought there was an LTP… you’d be spending your time looking for it. You know like Bigfoot.
I cant rule out Bigfoot. But merely speculating that it might be bigfoot, or but unicorns, or LTP, really isnt science. Now if you start looking for it, then you are at least doing something.
Take Judith’s Stadium Wave. She went looking. I count that as doing science.
Its simply not enough to speculate on what MIGHT BE.. you have to.. you know.. try to find big foot if you have an actual belief that it might be real.
So ya it might be LTP or Unicorns.. In the meatime we have a explanation that works pretty good.
ECS is between 1.5 and 4.5C. if you want to contribute work on that..
“Then why do you and Victor Venema have little confidence in satellites?”
Mears came out with TTT v4.0 versus TTT v3.3. Comparing those two suggests something.
RS&JC are still coming out with TLT v6.0b5 versus v5.6 almost FIVE years after starting on v6.0, Comparing those two (v6.0b5 versus v5.6) suggests something.
http://vortex.nsstc.uah.edu/data/msu/t2lt/readme.04Mar2015
“Update 8 Nov 2011 *************************
.
.
.
When the testing is completed we will be issuing version 6.0 of the
temperature products. We do not know when such testing will be completed.”
You can say that again (“We do not know when such testing will be completed.”).
When one compares the changes made between these versions versus any of the four SAT time series, one can generally state that … wait for it …
The originators of the satellite time series (Mears and RS&JC) exhibit ‘little confidence’ in their own creations. Why would anyone else.
Steven, I don’t mean to be giving you a hard time. I’m maintaining an non-emotional investment in any positions. BTW, Brandon, that’s also a good stock trading tip. Steven, I appreciate that you can modify views (and claims) with the entrance of new evidence. It’s helpful to often step back and reweigh. And you’re to be commended for being one of the most objective people around here. BTW, I think old CA posts are a great research tool, and sometimes it’s neat to see comments from 5 years ago. Sorry for citing your IPCC attack. It was several months ago.
First our UK friends might not know that on Bigfoot is a crypto-zoological creature, possibly a Gigantopithecus, that roams the United States. There are many absolute believers (mostly the witnesses) and skeptics. I am neutral and weighing any new evidence. I place the likelihood somewhere between extra-terrestrial visitation and ghosts, the the later being one in a billion. The chances of bias entering climate science based on the evidence I’ve seen is much much higher.
Currently the surface recorded indexes are at 3X the trend of the satellites over the last 20 years. (according to Nick Stokes yesterday). And, the satellites show more sensitivity to El Ninos than surface records. Two surface record professionals, Steven and Victor, claim, from evidence that I am unaware of, that the trend from the satellite record — and the balloon record too — are systematically >3X biased down. The evidence of unreliability I have heard so far only would account for poor precision, not poor accuracy. And the precision side-by-side with the each other and the surface record looks pretty good.
So who is the conspiracy theorists? Should each side start smearing the other? Should we investigate religious beliefs or can we get to the bottom of this scientifically? I am suspending my judgement as to which indexes are more representative of the GMST. Before somebody jumps in to inform me that the troposphere is apples and the surface is oranges let me say that Gavin Schmidt (God) says that the conversion is approximately 1:1 with any advantage to sensitivity given to the troposphere, especially over the ocean. BTW, this is not the link that was embarrassing to Steven. I think it was a WUWT post of his attacking the IPCC.
Okay, don’t throw eggs at me yet; I’m not done.
Ron… So, there’s a lot of difference in trends since 1996? So… You’re saying trends over statistically insufficient very very short time lines. Which by definition your graphs must always be extended to include 1998? Hmmm. 🙂 Funny how that is.
Now your beloved satellites are in error?
No wonder. You best start question that… How’s the ole RSS Satellite doing these days anyways? Spiking like everything else. How inconvenient. So inconvenient.
http://www.skepticalscience.com/trend.php
How was that so called documentary you went to? Morano’s latest bit of intellect? You seemed to think it would some how inform you of something.
I recommend that instead of back peddling, and quibbling over every changing minutia, you start looking at all the data objectively, and drop the need to include 1998 in all your so called evidence. Just a thought.
> So who is the conspiracy theorists? Should each side start smearing the other?
There’s no need to rip off your shirt while throwing rhetorical questions, RonG.
Please stick to something that looks more sciency.
As ever, thank you for your concerns.
[Playing the ref. -W]
“The evidence of unreliability I have heard so far only would account for poor precision, not poor accuracy. And the precision side-by-side with the each other and the surface record looks pretty good.”
“poor precision, not poor accuracy”
That doesn’t even make any sense. It’s accurate to 15-digits but we decided to round to the nearest degree centigrade. You can have high precision but poor accuracy, you can’t have high accuracy with poor precision, unless you decided to throw out perfectly good data via rounding to fewer digits. I took the ice core CO2 data and rounded to the nearest units of 1000 ppmv. I took the GMSL data and rounded to the nearest kilometer. The Earth’s temperature is 300 degrees centigrade (rounded to the closest hundred degrees).
High frequency? Yes. Low frequency? Not so much, specifically for UAH v6.0b5.
(1) You should plot all data (e. g. from 1979).
(2) You should plot RSS v4.0 TTT (not RSS v3.3 TLT as shown), as that is a reasonable proxy for what an RSS v4.0 TLT would look like.
(3) You should plot UAH v6.0b5 TLT (not UAH v5.6 TLT as shown).
(4) Pull off the Christy trick of starting the pinch point, at say, January 1979 (either raw data or corresponding trend lines), but regardless, show corresponding OLS trend lines on top of the raw data.
Not saying that the SAT is right on, just that the TLT/TTT time series are all over the place, relatively speaking, in their trendline estimates. Too few satellites not nearly enough overlap for reconciliation purposes.
300 centigrade should be 0 degrees centigrade (or 300 degrees kelvin).
Ron Graf,
A good stock trading tip is … not to.
That’s entirely backwards Ron – it’s worth getting to grips with the difference between accuracy and precision. Accuracy relates to the relationship between the determined value of a parameter and its real value. Precision relates to the statistical variability amongst a set of measurements of that parameter.
let’s say you have a thermometer that unbeknownst to you has a flaw. You measure the temperature of a cold room known to be at 4 oC. In 8 seperate temperature readings the thermometer reads 7.8, 8.1, 7.9, 8.0, 8.0, 7.9, 8.0 and 7.9 oC. That’s a rather precise temperature measure 7.95 oC +- a tad. However the temperature readings are not accurate since they’re 4 oC high in error.
Interestingly a similar thing applies to the satellite temperature measures. Oddly, Spencer and Christy’s breakthrough paper on MSU temperature was titled “Precise monitoring of global temperature trends from satellites” They claimed to be able to measure temperatures with a precision of 0.01 oC. However as we now know, however precise their measurements were, they were hopelessly inaccurate.
Oddly you have plenty of information to come to a similar conclusion yourself. Several of us have pointed out that the recent RSS TLT update has changed the warming trend from 0.08 oC per decade to 0.13 oC per decade. This clearly indicates a major element of inaccuracy somewhere (either with the original or updated values, however precisely those 0.08 and 0.13 values were measured at.
Given the excellent track record of RSS, I’m inclined to consider that their updated mid-tropospheric temperature trend is more accurate that the previous value…but we don’t know this for sure…
Ron, a couple of times on this thread you’ve asserted a SST cooling of 0.7 oC from the McGregor et al paper that you based your volcanic-induced temperature imbalance idea on ( e.g. “was why the SST dropped .7C from 1250-1850” ).
I don’t get that value from their paper at all. From their Fig 4d the temperature diffference between their binned values incorporating 1250 and 1850 AD is around 0.2 oC. if one uses their trend data in the Supplement (section 4), this might be 0.4 oC.
McGregor et al. isn’t the clearest paper in the world and is actually not terribly well explained either, but i don’t see how one can arrive at a 0.7 oC SST temp reduction over that period from their presentation. Could you clarify that please.
Likewise wouldn’t you agree that McGregor et al. determine that there are potentially rather large non-volcanic contributions to the apparent SST reduction in the pre-industrial milennium? That’s apparent from their Supplemental Table 14. Do you agree?
[Mindprobing. -W]
Two surface record professionals, Steven and Victor, claim, from evidence that I am unaware of, that the trend from the satellite record — and the balloon record too — are systematically >3X biased down.
No I did not. No Steven did not. [More mindprobing. -W]
If you go back to my comment, you will see that I argued that you would need to argue that the satellite record is also wrong. The trend over the full satellite period is about the same for satellites and surface. Your assumption that there is a huge warming bias in the surface temperature record, thus implies that there is also a huge warming bias in the satellite record.
Your factor 3 comes from cherry picking a shorter period. Something which has an especially large impact for the very noisy tropospheric/satellite temperatures. A bottom-of-the-barrel deceptive strategy.
Ron.
You have run away from your obligations.
You raised the issue that anders 1c was not the consensus.
You claimed .6c
You have run away from the choices to
Defend or amend your claim.
Now you have run to sat records.
“Two surface record professionals, Steven and Victor, claim, from evidence that I am unaware of, that the trend from the satellite record — and the balloon record too — are systematically >3X biased down. ”
Look at this site . See my post on satellites.
I don’t claim what you think
Here is Steven’s post on the surface and satellite discrepancy.
Steven, I wrote a lengthy response to you last night with clear logic supported by multiple studies for my position on UHI. I see it’s gone now but I will reconstruct it tonight. The 1C I used came from the post above by Anders. That’s what he used.
Victor, I pointed to Nick Stokes’s recent post to show the 3X delta trend from surface trend for the last 20 of a 37-year satellite era to highlight the degree of divergence. 20 years is a significant time. I know the trend delta is not 3X for 37 years but it is not 1X either. If they did match you would have an important support. You and Steve should put together a team to get to the bottom of the problem with RSS and UHI. Just remember to includes some RSS and UHI expert defenders to supply balance so that the results are accepted universally. 😉
I am well aware of the difference between random error that produces poor precision, like a shotgun pattern, and systematic error that produces poor accuracy, like a miscalibrated gun sight. For the satellites to show good correlation in high frequency but low correlation in low it would reason that either satellites or surface recordings are systematically askew. Any thoughts?
Ethan Allen, I would welcome your chart. I just used what was now available with Woodfortrees interactive. No Tricks. Honest.
Ron,
A couple of points.
1. That you can find studies that support your position does not mean that your position is supportable.
2. Steven has already explained that the SST have already gone up by about 0.9C. We also have good evidence that the land surface warms faster than the sea surface. Given this, it becomes incredibly difficult to justify that UHI has had a big influence on overall global warming.
There’s more, but I have to go and finish dinner.
Chris, sorry for the delayed response and also to anyone else. The – 0.7C 1250-1850 anomaly approximation comes from several independent supports.
1) A 1.4C amplitude for a 800-yr scale is very plausible, as acknowledged by ONeillsinWisconson, who also placed the 2K plausible amplitude at 2C, both of which I find plausible. For an amplitude of 1.4 the half wave would be 0.7C
2) For the Little Ice Age to be the LIA it would have required a noticeable drop in temp, on the order of 1C.
3) Specific eruption are documented to have produced winter type conditions in the summer. This would have required a 20-30C temporary drop that would have logrhythmically dissipated over at least a decade or two.
4) I agree with ONeillsinWisconsin when he says that McGregor’s claim to be putting 80% of the anomaly eggs into the volcanic basket seems implausible, even if “confirmed by GCM fingerprint”. It’s also likely that the Maunder Minimum solar inactivity contributed to the temporary forcing. This several hundred years of minor but steady loss of energy combined with the staccato of severe volcanoes, along with weakening M-Cycle orbital NH insolation could have combined to create the LIA. The center of the inflection minimum for surface temp for the LIA was around 1650. Since the ocean temp always lags the surface it is plausible to move the deep ocean temp minimum to 1850. Remember, the deep oceans are lagging the warming now, which is what we call “the pipeline” and why we are interested in what “equilibrium” temp with the deep ocean is.
I stand by an up to a -0.7C to -0.4C, being the most plausible range for 1850 imbalance. I would dispute -0.1C as being plausible.
How is this an imbalance?
If the ocean is cooler than equilibrium it will result in a positive imbalance that will gradually trend toward higher GMST to correct the imbalance. You know…
Ron,
But you don’t know that it is colder than equilibrium, or – if it is – what the equilibrium value is.
And by what mechanism were the oceans ‘cooling’?
Ron Graf:
Over on the Skeptical Science trend calculator[1], the trends from 1996-2017 in C/century are:
RSS 4.0 (ttt) — 1.43 +- 1.60
UAH 5.6 (tlt) — 1.58 +- 1.60
GISS — 1.85 +- 1.02
Note two things:
1) the large difference between UAH 5.6 here (1.58 C/cen) and UAH 6.0 as reported by Stokes (0.53 C/cen). This is from the difference in versions; the difference in trend from changes in methodology. UAH is still undergoing large changes in their reported data. That should give you some pause about the accuracy of the data. Seriously, how could it not?
2) Look at the error bars on all of these over these timespans. The trends with uncertainties overlap just fine. The end result is: “no, the 20-year trends are not statistically significant enough to say that the surface and lower troposphere are divergent”. You have to look at longer trends, because these trends + uncertainties overlap just fine.
And when you use longer trends, they still overlap pretty well, because the satellite trends rise even as the uncertainties drop.
Ron Graf:
Well, let’s go to the source. What does Carl Mears (a senior scientist at RSS) have to say about the satellite temperatures? “I would have to say that the surface data seems like its more accurate.”[2] That’s his position in many publications and interviews; I recommend watching ref [3], in which Mears discusses how the discrepancies between surface and satellite temperatures and models. Or you can read about it on his blog, [4].
Skeptical Science also gives a pretty good overall perspective here: [5]. Key takeaway: “the known uncertainties in the satellite trends, as estimated by the record provider, are five times the uncertainties in the thermometer record trend”.
I’m sorry, but your claims here are not supported by evidence.
[1] http://skepticalscience.com/trend.php
[2] http://www.yaleclimateconnections.org/2016/03/experts-opt-for-surface-rather-than-satellite-temp-data/
[3] https://climatecrocks.com/2016/01/26/carl-mears-on-climate-data-vs-models/
[4] http://www.remss.com/blog/recent-slowing-rise-global-temperatures
[5] http://www.skepticalscience.com/Satellite-record-vs-thermometers.htm
> I see it’s gone now […]
It was the third time you tried to inject something you were specifically told not to, RonG. Your “I see it’s gone now” could be considered your fourth time. To be on the safe side, you might try to stay away from smileys and predicates about mental states like “honest.”
***
Everyone else,
If you could stick to sciency points, that’d be great.
No, RonG’s mind, morals and overall manners don’t count as sciency points.
***
> That you can find studies that support your position does not mean that your position is supportable.
Perhaps you should rephrase that, AT.
Ideally a scientific position should be based on considering all the available evidence, not just a subset that supports the position you would like to hold. That you can find some studies that appear to support what you’re presenting, doesn’t mean that what you’re presenting is somehow likely.
Ron,
I await your post. However, merely “pointing” at papers and making vague arguments about what UHI MIGHT BE… still avoids your obligations.
1. Anders asserted the consensus is about 1C since 1850.
2. You asserted that the consensus was .6C.
3. You use this lower number here and at Lucia;s because you argue that there is a large
perhaps 1/3 of the warming signal in the land that is UHI and Microsite.
4. We dont even have to discuss the merits of your claim to see that it cannot be Correct
I will show that now:
A) The Ocean warms at .9C it is 70% of total
B) The land warms at 1.5C it is 30% of the total
c) simple area averaging gets you to 1.1C
D) Now I will do several “parametric” verssions
D1) All of the 1.5C is UHI !! This yeilds an average of .63C for the globe
D2) 50% of the warming on land is UHI. This yeilds an average of .88C for the globe
D3) 33% of the warming on land is UHI This yeilds an average of .96C for the globe
On Lucia’s you asserted that some studies suggest a that 1/3 of the warming on land is
UHI and Microsite. This would give you D3. That is why I generously suggested that you
should have argued that .9C was a better number.
Next, we have good physical reasons ( heat capacity) that tell us the land must warm more
than the SST. This effectively eliminates the possibility of UHI biases greater than 50%
Now, you bring up the satellite.
The Difference between the Land (RSS) and the Land (BE) is roughly .11C per decade
Now, some have suggested that this is due to UHI.
There is a problem with that.
A) The Spatial pattern does not support this as the difference between satellites and surface
is greatest at both poles.
B) The discrepency between satillite and surface changes dramatically when the sensors
were changed ( may 1998)
Something is up with that..
What this suggests is that we cannot take the difference between satellites and surface as simply a UHI micro bias effect. The science isnt settled. Is some portion of the difference due to
UHI and microsite? I cant rule that out, but I do know that I cannot extrapolate the difference to the past outside the satellite time period.
Sorting out the differences between satellites and the surface is difficult work.
Look at it this way. Since the early 1930,s scientists ( ok data scientists) have been working
on methods and data for the surface. That’s 80 years of attention to bias, methods, corrections.
It is now so well understood that its simply a matter of engineering that even a brain dead English major can do. not rocket science. look ma no Phd!. Dozens of folks have taken their best shot at improving it. The numbers dont change much.
On the other hand we have 2-3 groups working on Sat data. Their code has only been available for a short time. Not many eyes have looked at the methods. the methods are all AD HOC.
Over the past 10 years their answers have gone up and down by huge amounts.
This doesnt mean I reject their data. I dont. I want to understand why the data differ.
Full satellite period (1979-2016)
UAHv5.6 TLT Trend: 0.142 ±0.063 °C/decade (2σ)
RSSv4.0 TTT Trend: 0.168 ±0.061 °C/decade (2σ)
GISTEMP Trend: 0.165 ±0.041 °C/decade (2σ)
HadCRUT4 krig v2 Trend: 0.178 ±0.041 °C/decade (2σ)
Ron Graf, you claim that the surface temperature increase is only 60% of what science says.
If that were true over the satellite period, the trend would be
TGraf = 60% * Treal = 0.6*0.178 = 0.1068 °C/decade
Even generously taking the largest HadCRUT4 trend, this is way below the tropospheric temperature trends of the satellite datasets. Thus you claim that the satellite temperatures are wrong. Any suggestions what John Christy and Roy Spencer are doing wrong?
VV,
I get the following for just the RSS and UAH time series …
Monthly Data Trends 1979-01 thru 2016-04
RSS TLT v3.3, 1.297 °C/century
UAH TLT v5.6, 1.492 °C/century
RSS TTT v3.3, 1.211 °C/century (linear combination of TMT and TLS)
UAH TLT v6.0b5, 1.192 °C/century
RSS TTT v4.0, 1.755 °C/century
UAH TTT v6.0b5,-0.011 °C/century (TTT = UAH TTP = linear combination of TLT, TMT and TLS)
That last one, UAH TTT (TTP), is a doozie at -0.011°C/century versus RSS TTT v4.0 at 1.755 °C/century
I get your two satellite numbers if I drop the 2016-01 thru 2016-04 months of 2016.
Ethan Allen, I used the SkS trend calculator.
http://skepticalscience.com/trend.php
Your RSSv4.0 TTT Trend is basically the same. Could be that SkS does not use the last few months.
I see no reason to use an outdated RSS version when they just published a paper explaining that they found new problems and gave an update. Nor do I see any reason to use an unpublished UAH version that does not take these problems into account yet.
Steve Mosher writes:
A point I’ve made many times – anyone can realistically duplicate the surface datasets within a few hours to within the a few hundredths of a degree. The satellites series? Good grief. We probably have to start with something like this:
The above taken from, ALGORITHM THEORETICAL BASIS DOCUMENT (ATBD)FOR THE ADEOS/AMSR SEA ICE ALGORITHM, Josefino C. Comiso.
That’s the starting point. It gets worse from there. As John Mashey quoted Richard Lindzen upthread, ““The technical details of satellite measurements are really sort of grotesque.”’
Victor, Ethan, You got to supply the bowl of cherries by choosing the 37-year span and global surface indexes. So I get to choose the better of each pair.
HadCRUT4 krig v2 Trend: 0.178 ±0.041 °C/decade (2σ)
UAH TLT v6.0b5, 1.192 °C/century
1.19/1.78 * 1C = 0.66C Okay you got me. But my claim was simply that 0.6 was in the good plausibility range.
I don’t know that TTT should be following the trend of the TLT or surface.
Victor, I think Steven is looking for a partner to investigate UAH 6.0. Please do, but do us all the courtesy of reporting what you found.
Comment on UHI coming…
Steven, you and I started our discussion on UHI the beginning of January at Lucia’s. Your first informative comment was here on Jan 8 where you stated in part:
So you started off like me doing your own investigation into the source documents and not relying on pundits and web hosts. And, I was eager to be enlightened, ready to here the first major sensible refutation of a meme of the WUWT crowd led by Watts, who is seen as not even a lukewarmer, which I am certainly one, as were you. The you continued (in part):
continuing…
“Comment on UHI coming…”
Ron. Your comment needs to address the following.
1. You claimed .6C of warming since 1850
2. You claimed this is the result of UHI and microsite.
3. to get to .6C you need ALL OF THE LAND WARMING to be the result of UHI/microsite.
I am not too very interested in discussing papers that establish
A) UHI for a city
B) UHI for a region
C) UHI for a country
D) UHI for a continent
Why?
Because your claim of .6C REQUIRES that Every continent, country, region and site
shows NO warming. we know that is not true.
So you can either
1. SPECIFICLY DEFEND your ,6C estimate
2. Amend it ( I’ll suggest that .9C could be discussed )
3. Run away and attack the STT.
The other way to look at it is this.
In the ECS formula there are three key variables. Delta C, Delta Forcing, Delta Q
When I looked at that a long time ago.. it was clear where the real uncertainty lay.
The frickin SCIENCE told me the biggest uncertainties were in forcings.
So, if you want to attack.. point your sword there dont bang your head on the UHI wall.
The same goes for delta temperature. The biggest uncertainties are not in the Land
The fricking math tells you the weak points; SST and Forcings.
Instead You give arguments on UHI which at best hits 30% of one of the parameters.
And its been studied to death. That said, when you finish beating your head against that wall
I will empathize.
Totally confused, but not giving up, I went to the papers. UHI has been known as long as AGW, over 100 years, starting with urban London. Oke, who started studying at UHI back the 1970s to present started by driving his car from the countryside into the city on a controlled altitude with a thermometer out the window. Today, many people can do this experiment with their car thermometer. Steven Mosher responded to this evidence with this:
“Observation. When you drive across a city you will perceive that it is warmer than the country. NOTE, you don’t stop at every discrete point to check whether it is IN FACT warmer at that precise location.. you just note as you drive and record temps along ONE transect that it is warmer..[Oke is devastated.] Prediction: If you pick a point outside the city and record temperatures and you pick a point inside the city and record temperatures… Observation number 1,, says you should expect these to to be different. Guess what? Still, even THIS misses the point. because the question isn’t :do transects taken across a random city show a urban rural difference for a particular time. [Did everyone understand this. Good.] The question is. is global Tave Biased by UHI. You might think that evidence from a few transects answers this question. it doesn’t. [Move along. Nothing to see here.]
continuing…
This is all confusing. So, Eli tries to help cheer me up the next day:
Eli Rabett (Comment #142654)
January 9th, 2016 at 8:54 pm
Yes there are urban (even suburban) heat islands. The question is whether trends in temperature observations are affected by those urban heat islands. Everybunny here knows that but the confusion contaminates the comments. Another fun thing is that urban stations systematically were moved to less dense parts of the urban areas during the 20th century and the population of core cities decreased as cities spread. Unadjusted one would expect trends in urban areas to be lowered by both effects.
OK, UHI is real. So forget about Oke fooling himself like Feynman warned. Eli says there is suburban heat islands? I knew this from the reading. But could there be rural heat islands? Is the idea of islands misleading? Population density is used to assume UHI, but when people depopulate an area do they decrease the UHI? Or, do they leave behind what causes UHI? What about micro-site, where an air conditioner condenser is installed by the HVAC crew and nobody notices that it’s 15 feet away from the sensor — for years. I think Anthony Watts discovered this phenomena. It has been corrected now mostly by moving the stations. But does moving stations to “more rural” airports in the 1960s and 70s correct the problem or re-initiate it? Cities tend to grow toward the airport. Airports grow in every direction. What cause UHI? Is there any of that at airports? Do people live at airports? How is population density measured, by residents or visitors? So many questions. So how big is UHI? If it’s tiny then Steven is right, it can’t affect Tavg.
Kim(2002) Studied UHI in Seoul compared to the close-by rural station on the same latitude. Kim broke down associations and patterns relating to time of day, humidity, season, precipitation and wind, the latter two wiping UHI out, (which provides a second control beside from the rural station). Kim found a 0.54C increase in the mean UHI events over a 24-year period, with an event being defined as an increase over rural of more than 1C for 24 hours, with an average maximum UHI increasing from a whopping 4.0C to 4.5C in the winter over the interval. Jongtanom (2011) saw and average maximum UHI of ~ 2.5C in three cities in Thailand.
I confronted Steven with these papers and he acknowledged they good papers. Kim(2002) was one of favorites and they were not even overstating UHI. Tokyo, Steven said, I believe, has UHI peaking at an impossible 6-7C. Steven said BE finds no need to adjust for UHI normally except for cities as large as Tokyo. I’m not sure how many.
So, is there any way to know if UHI is a factor outside of looking at temperature deflections from control stations? The answer is Yes! DTR.
Hamdi (2010) states:
So, how was UHI dismissed by the climate establishment? A far as I can tell they looked at the land record and looked at the sea record and the models and saw a match. To me this is pretty confident way of doing things. Some might call it circular validation. What if the models are wrong? What if the SST record is biased? Or both? But, there is not any way to verify UHI from AGW. WRONG!
Hamdi confirmed a fingerprint of 2.5X difference in trend of TMax to TMin can be compared to rural deltas nearby to find the net UHI component of TMin trend to AGW’s component. Furthermore this can be checked by comparing trend of windy or rainy days trend of TMin to that of clear, still days for a comparison of controls. If the self-control shows less trend than the rural control then we have evidence of rural UHI. This is probably something that Karl would never do giving him a thousand years. But why somebody hasn’t done this can only be that there is no desire to reveal it (IMO). Are you going to tell me to do it, Steven? If I’m crazy please explain how.
ATTP says: “Ron,
But you don’t know that it is colder than equilibrium, or – if it is – what the equilibrium value is.”
Good question. Actual equilibrium continually changes depending on TOA radiative flux relative to ocean temp. The longer the time interval you are referring to the deeper ocean layer that is included to calculate your equilibrium. For instantaneous equilibrium you are using the ocean skin. For a 1000-yr equilibrium you are talking 1-3km deep. I honestly don’t know or if anyone knows how deep. The depth depends on mixing rate and convection, which are dynamic steady-states. So good question.
Tokyo UHI? Big deal, who cares? Lake Suwa is 190 km from Tokyo and it has it’s own story to tell:
You don’t need a global dataset of thermometer readings, or a GCM ensemble, or a satellite temperature series to read the story that both paleo data and phenological records are telling. In fact, if the thermometers *didn’t* say it was warming at a record pace we’d suspect they must be broken – cuz everything we observe in nature tells us it *is* warming at unprecedented rates.
Ron,
You are still failing to defend your claim that overall warming is probably about 0.6C. Let me remind you of Steven’s comment and ask you to directly address this.
“Victor, Ethan, You got to supply the bowl of cherries by choosing the 37-year span and global surface indexes.
So I get to choose the better of each pair.”
Actually not true. I picked all available satellite data which currently has a common base line of 448 months.
If there were even more data I’d have included that also. Some people say use a minimum of 30-years (360-months), I prefer 40-years (480 months), so technically, 32 more months to go.
I now report all relevant 447-month surface and satellite trends (NOAA and HadCRUT have not reported 2016-04 yet), SAT, TLT and TTT.
(01) UAH TTT v6.0b5 (-0.025 °C/century) That one is the kicker. UAH sez TTT is cooling (while at the same time, both TLT and TMT are warming), at least that is what Monkers would say.
(02) UAH TLT v6.0b5 (1.175 °C/century)
(03) UAH TLT v5.6 (1.476 °C/century)
(04) RSS TTT v3.3 (1.195 °C/century)
(05) RSS TLT v3.3 (1.283 °C/century)
(06) RSS TTT v4.0 (1.755 °C/century)
(07) NOAA (1.594 °C/century)
(08) BEST2 (1.612 °C/century)
(09) HadCRUT (1.688 °C/century)
(10) GISS (1.706 °C/century)
(11) BEST1 (1.787 °C/century)
(12) C&W (1.828 °C/century)
The combination you chose was (02)/(12), the lowest satellite value with the highest SAT value!
I’m not including (01) because UAH has totally hosed that one to date (due mainly to their low balling TLT).
You also ‘think’ that ‘ratio’ has some utility and take the ‘liberty’ of extrapolating it back in time 4X.
Throw out the, as yet, unpublished UAH v6.0b5 time series and using the newly published RSS v4.0 time series (as VV suggests) and one gets a mean ratio of 0.95 (n = 12 combinations). That’s close enough to one to call it one.
If you don’t like them apples, I’d very kindly suggest you publish you own results in E&E (which is about the only place that publishes such trash sans an offshore (China or India) pay-to-play predatory journal).
Ron:
But no published sources apparently.
So when you say:
You were just making it up. [Mod: redacted]
That’s lots of words coming from another conversation, RonG. With lots of questions, too many questions, notwithstanding a whole paragraph about the “establishment,” a paragraph which conspires to be quite rhetorical. A link with a short description may have sufficed.
Too many words and too many questions for not answering a simple question.
There’s a lot of arithmetic about temperature here, which really should be about heat. And if it’s about heat, none of that arithmetic concerning matters such as ocean/land, surface/satellite and UHI splits is valid. ATTP, I think you should pull the lever on the red chair.
Folk who think their back of the envelope is superior to a physically-based earth system model have another think coming. Those models transport energy pretty well. It’s the analysis that we subject them to that I don’t always agree with, but their capacity to represent the climate system is impressive.
d[Anomaly(urban) – Anomaly(rural)]/dt =0
Talking trends.
Ron… You need to show that UHI affects the measurement of global temperature. So far, all you’ve done is hand wave, and point at small regions.
So.. without even, debating your reference to Kim 2002;
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0450%282002%29041%3C0651%3AMUHIII%3E2.0.CO%3B2
You need to consider that Seoul is only 605.2km^2, and (giving you the benefit of the doubt) the earth’s land surface is 148,940,000km^2… So how much error would this UHI add to our measurements? 4.06X10^-6… Or you have claimed UHI affects readings by .000406%. This is the part in the conversation where you admit you are talking about something unrelated to global warming.
What everyone keeps repeating at you is that you need to look at all the data, and not select locations, and specific time lines (which must always include 1998).
If you consider the the globe for your data you get something like this;
Click to access 2327-4581-1-104.pdf
(Steve Mosher… Great Paper dude… the links in the references work!)
Let me urge people not to respond to Ron about satellites or his recounting of past conversations.
He has an obligation to defend or amend his claim of .6c.
First.
Then we can discuss the ins and outs.
Look ron. I asked did ran a straightforward question about his mail to santer. He answered it. I’ll ask ron to do the same.
Then he can question me endlessly if he likes..
Attp said 1c
Ron said 6c.
He attributes the difference to uhi.
It’s a simple mistake.
Let’s clear it up and proceed.
Thanks oilman.
Thanks Steven for your patience. It takes a bit free time to formulate a substantive response that will move mutual understanding forward. [Snip. -W]
Ethan, regarding satellites trend, I forgot to add to my equation that the consensus factor for theoretical sensitivity of TLT to surface is a factor of 1.1 to 1.2. Thus 0.66/1.1 = 0.6C
Steve, this is my original claim in my first comment:
The point was questioning certainty of Anders’ claim that EfCS was more than 1.1C
1) Anders used 0.1C pre-industrial imbalance. Since then the legitimate question as to the definition of imbalance has come up. In a EfCS context I would presume that it is the mean surface temperature at which the average TOA radiative flux for the past 500 years would create an equilibrium. And then realizing that the earths average mean surface temperature would remain constant at a fixed TOA flux if the ocean heat content and surface were in a hydrodynamic steady-state. And, whereas I demonstrated peer reviewed evidence that the GMST for the 800 year before 1850 deflected 0.7-0.4C I presume that should be a more accurate range in a multi-century context then 0.1-0.08.
2) I said that 97% of climate scientists do not agree with the 1K warming. I did not say I was 97% sure that 1K was wrong. But I do present a plausible argument of those with skepticism.
3) High uncertainty cuts both ways and I am all for working to reduce it. The IPCC still has the ECS range from 1.5-4.5C. I believe that Lewis and Curry’s best estimate is 1.8, Linzen and Choi, 1.2.
continuing…
Anders, when you say there is a 0.5C imbalance now I presume you are talking on a centennial scale and acknowledge that El Nino’s and volcanoes fluctuate imbalance from the mean by increasing/decreasing OLR. My argument would be that if we started with a 0.7C imbalance and we have a 0.9C imbalance now then this paints a very different picture about EfCS. Particularly if the land record is biased.
anoilman says: “UHI affects readings by .000406%. This is the part in the conversation where you admit you are talking about something unrelated to global warming.”
I think you are misunderstanding UHI, which is known (by Hansen and Karl) to have biased the measurements of land stations by a significant amount. Actions and adjustments were made and are still made to correct for it. If you are saying that cities temperature is a very tiny footprint compared to the globe I agree with you. That is why the point of the allegation of sampling bias.
Eli says: “d[Anomaly(urban) – Anomaly(rural)]/dt =0”
I understand that this is what the Mosher investigation came to in his Jan 7 comment I quoted and apparently confirmed by Karl with USCRN. I am not 100% sure that I accept it since it is not accompanied by logic or a revelation into the mechanisms. And, Mosher admitted his own surprise at the equation above. Therefore, critical thinking demands looking into possible confounding influences that could trick the result.
continuing…
> I do present a plausible argument of those with skepticism.
Focus on your plausibility argument, RonG. No need to rehearse this thread, and please refrain from more “97%” provocation. The latter rests on an appeal to ignorance, by the way: search for “survey” on this page to see why.
UHI and micro-site (MS) are real — complete with continual and reproducible evidence. That being the case one would naturally think UHI / MS would bias the land record by the amount of city/suburban stations component to rural. There was an assumption that the rural stations are pristine and thus they could be used as the control to investigate this strong hypothesis. Surprisingly, bias was not found. Rural followed same trend as city. And then the rural control is compared to the oceans which is thought to be a known to verify their trend. But my critical thinking side flashes a red light there. What if the oceans are biased? Do we have other controls? I found through investigation that we do.
1) Windy and rainy days erase UHI / MS.
2) UHI / MS leave a fingerprint of diverging trends of TMin from TMax by a factor of 2.5
Has either one of these been used to look for UHI / MS in “rural” and “suburban?”
Steven, how can we eliminate the possibility that UHI / MS is not in rural record? Couldn’t we use 1) and 2) independently and, better, in combination, to confirm?
BTW, it took me a while to find where found the consensus peer-reviewed literature that showed an inferred UHI from the DTR trend. It was by Vose’s work. here is the comment from last Jan where I cited the work and came with a back of envelope calculation of 0.35C.
Despite Parker(2010) and McGregor(2015) not making skeptical claims about the “consensus” does not mean there work is unfruitful to skeptics.
–end
Thank you, RonG.
In case anyone missed it, if the satellites are more accurate than surface record a 0.6C GMST or 40% less warming has likely occurred. Supported such a likelihood with independent evidence is the possibility of a 0.35C bias in the land record.
.
This was all inspired, I must admit by the excitement of the news that Anthony Watts is trying to get a paper published that supports this. I hope we get to see it and also see more debate.
Thanks Steve for your hard work on this problem and for informing me. I hope to read your old work and to see some new I can read.
Thank you Willard.
Ron
Nice try at walking from the .6C argument
“2) I said that 97% of climate scientists do not agree with the 1K warming. I did not say I was 97% sure that 1K was wrong. But I do present a plausible argument of those with skepticism.”
here is what you wrote at Lucia’s on the same subject
@Eli: “we are at 1C above 1880. People keep forgetting that.”
Your reply:.
“If the ground record is contaminated by 1/3 with UHI (as some like Watts presume,) and the satellite and balloon records are more representative of trend (as Christy, Spencer and Mears would claim,) then we have had perhaps as little as 0.65C warming since pre-industrial.”
See that?
On Lucia’s you responded to the 1C warming by appealing to Watts… Claiming 1/3 UHI
and then concluding .65C
Here you repsonded to ATTP claim of 1C by saying .6C and then playing the same UHI card.
I challenged you to defend the math of 1/3 UHI getting you to .65C over at Lucia’s and you
did the same dance. avoid the question
Over here you have a similar claim. 6C.. and a similar dance.
Here is the simple fact. The unanimous verdict on global warming since 1850 is 1C or more.
Not consensus.
UNANIMOUS.
here is another clue. Studies of cities, of regions, of 600 sites, do not and never will constitute
a global record. When someone publishes a record with less than 1C warming, you will have an argument. until then you have zero.
“In case anyone missed it, if the satellites are more accurate than surface record a 0.6C GMST or 40% less warming has likely occurred. Supported such a likelihood with independent evidence is the possibility of a 0.35C bias in the land record.”
1. There is little to no difference between the SST record and the RSS record for the ocean
2. The land is 30% of the earth for the last time. The difference in land records, is on the
order of .1C per decade. If you extrapolate that backwards 16 decades you
get…… Zero warming for the land. which is physically wrong.
3. there is no reason on its face for the satellite records to follow the land records.
.
This was all inspired, I must admit by the excitement of the news that Anthony Watts is trying to get a paper published that supports this. I hope we get to see it and also see more debate.
1. his paper doesnt support it. read the paper.
2. Like I’ve pointed out ( and as anthony discovered) “microsite” bias stopped for some strange reason.
Ron Graf writes: “This was all inspired, I must admit by the excitement of the news that Anthony Watts is trying to get a paper published that supports this.”
Do you mean the original – Stop the presses! 2012 manuscript or its metamorphosis into a 2015 AGU poster?
Victor Venema covered them both:
2012- Blog review of the Watts et al. (2012) manuscript on surface temperature trends
2015 – Anthony Watts at AGU2015: Comparison of Temperature Trends Using an Unperturbed Subset of The U.S. Historical Climatology Network
Co-author Evan Jones makes numerous appearances in the comments to VV’s 2015 post. And also of interest on the same topic is the comment section to this 2014 post at Stoat’s, The Fruitarians Are Lazy with Evan Jones discussing the 2012 debacle, and their new ‘heat sink’ hypothesis with VV among others.
Actually Ron.. the beginning of the Berkeley paper discusses all the different surface measurements and how each of the different systems has identified and compensated for UHI (if at all). So… its all known and understood.
For their part the Berkley paper states that the urban areas account for about 1% of the earth’s land mass. Meaning… a tiny error could be introduced because of it. Its even less if you include ocean surface temperatures. They then go on to measure it as a non-existent issue.
At some point this is just quibbling over minutia of no discernible impact to what we know.
“UHI and micro-site (MS) are real — complete with continual and reproducible evidence. That being the case one would naturally think UHI / MS would bias the land record by the amount of city/suburban stations component to rural. ”
1 yes that is the natural hypothesis.
Jones tested this
Hansen tested this
Peterson tested This
Zeke, Nick Stokes, and I tested this
berkeley Earth tested this.
In all but one case no UHI was DETECTABLE in the global record. In the one case (zeke et al)
where we did find a signal it was in raw data. Zeke went on to publish what happens when you
adjust.. the signal vanishes.
So you natural hypothesis has been tested. repeatedly. Look, If UHI was so easy to find in the global record, ask yourself why no skeptic has take the few hours it would take to demonstrate it?
2. Assuming that the UHI amount is proportional to city/sub versus rural is very simplistic.
A) It assume a constant “urbanization” in the record. urban growth hasnt worked that way
Ever notice how nobody attributes the “warming in the 30’s and 40’s” to UHI?
you know the warming the GCMs cant capture?
B) the urban and suburban landscape is NOT UNIFORM. Please read Oke on LCZ.
C) Some rural locations are warmer than city locations, especially if the land cover is bare soil
D) Studies of UHI ( of cities) Tend to focus on deriving PEAK UHI figures, not average
“There was an assumption that the rural stations are pristine and thus they could be used as the control to investigate this strong hypothesis. Surprisingly, bias was not found. Rural followed same trend as city. And then the rural control is compared to the oceans which is thought to be a known to verify their trend. But my critical thinking side flashes a red light there. What if the oceans are biased? ”
1.Ah yes. Your hypothesis about finding it in global was busted, so you turned to attack the oceans. I thought of that too years ago, and Dr feynman kicked my ass. The simplest explanation
was I was wrong about UHI infecting the global record
2. But, I thought, what if there, is another record… a really LONG record
http://onlinelibrary.wiley.com/doi/10.1002/grl.50425/abstract
######################
Do we have other controls? I found through investigation that we do.
1) Windy and rainy days erase UHI / MS.
2) UHI / MS leave a fingerprint of diverging trends of TMin from TMax by a factor of 2.5
Has either one of these been used to look for UHI / MS in “rural” and “suburban?”
A) yes windy and rainy days modulate UHI. That is WHY you can find 7C at tokyo.
The study is DESIGNED to isolate and find the WORST CASE UHI. As I have tried to
explain to skeptics for a long time these mitigating facts result in a muted UHI signal.
B) The problem with DTR is that multiple issues are in play. In the end, what matters
is the global average
Steven, how can we eliminate the possibility that UHI / MS is not in rural record? Couldn’t we use 1) and 2) independently and, better, in combination, to confirm?
urban heat island in the rural record?
thats a good one.
Here is the bottomline
The hypothesis was that UHI/MS would show up in the global record.
1. That was tested multiple times: no cookie.
2. The result was Bolstered by comparisons with STT
3. then people attacked SST
4. The result was backed up by re analysis using no temperature data..
5. And now your suggestion is that rural isnt rural.
And you give Homework on top of that.
Look. as someone who was very skeptical of the work that hansen and jones did, I actually went out and edumucated myself all over again. I had the benefit of being a former math and physics student ( yes I started college as math and physics ), but it was not that hard to get up to speed. Just time on my butt reading readin reading. At this stage, I am not disposed to look for things that skeptics themselves should be looking for. because there will always be another assignment.
My hourly rate varies. we can work something out
==> “I challenged you to defend the math of 1/3 UHI getting you to .65C ….”
Rather remarkable that as many times as the question was asked, it still wasn’t answered.
Downright Tolian.
Tropospheric temperature trends:history of an ongoing controversy (2011)
Click to access ThorneEtAl.WIREs2010.pdf
“It is concluded that there is no reasonable evidence of a fundamental disagreement between tropospheric temperature trends from models and observations when uncertainties in both are treated comprehensively.”
Revisiting the controversial issue of tropical tropospheric temperature trends (2013)
http://onlinelibrary.wiley.com/doi/10.1002/grl.50465/epdf
“Using these approaches, it is shown that within observational uncertainty, the 5–95 percentile range of temperature trends from both coupled-ocean and atmosphere-only models are consistent with the analyzed observations at all but the upper most tropospheric level(150 hPa), and models with ultra-high horizontal resolution( 0.5ı 0.5ı) perform particularly well.”
Tropics != Globe
To anyone who may be listening …
This rehashing of well worn contrarian talking points is getting VERY OLD!!!
Let’s treat RonG as the messenger he claims to be. No more piling on, please.
anoilman says: “For their part the Berkley paper states that the urban areas account for about 1% of the earth’s land mass…”
If the one percent one samples is contaminated with un-natural heat bias the calculation for the whole globe will be thrown off. That is what the UHI point in this thread is about.
Steve, did I miss Watts (2016)? Can you provide a link? I was not assigning homework. I’m simply asking questions. I am on your side and promote your answers to others if they make sense. You’ll not have to repeat yourself if you gave clear explanations for the following:
1) Since you point to an example where bare land causes UHI, and whereas agriculture is more likely to clear tress/vegetation in fertile areas than plant and green arid landscapes? True/False
2) Once lands are developed they stay developed even if they may become depopulated. Buildings roads and parking lots remain. Although development is uneven in each location as a function of time, the globe as a whole is gradually cultivation, road building, skyscraper constructing and airport expanding. True/False
3) The weather station record was created to give information to city populations and farmers, not to give controlled representative sampling of earth temperature. True/False
4) That UHI and MS existed to the same extend in 1850 globally as it does today makes no sense. True/False
5) Since the temp trend is the same in both rural and suburb and city locations the logical conclusion would be that UHI-MS is universal in the land record, not non-existent. True/False
6) The unexpected result is thus that UHI is found in rural records. This unexpected hypothesis can be tested by isolating wind/rain temp records from clear/dry for trends in temp and more importantly, DTR. Because the latter is a fingerprint for UHI. True/False?
7) Number 6 has not been investigated. True/False.
8) Rather than assign you work I offer to help recruit for you and support such work. True/False
BTW, I admit one day I said 0.6C the next I said it could be 0.65C. [But IPCC. -W]
Ron,
This is just horribly confused
An imbalance has units of W/m^2. At a fixed TOA flux, the mesn surface temperature would NOT remain constant; we’d be accruing energy and temperatures go up.
Ron,
I’ll add that you still haven’t answered Steven’s question. I think you should either defend your clsim that the change in global surface temperature could be as low as 0.6C, or withdraw that claim.
“If the one percent one samples is contaminated with un-natural heat bias the calculation for the whole globe will be thrown off. That is what the UHI point in this thread is about.”
It’s called inverse area weighting. Get a clue. There’s plenty of UHI literature out there. All your many questions have already been answered in the peer reviewed literature. It’s called prior art. Seek and ye shall find.
You really should try to come up with an original thought or three..
Anders says: “An imbalance has units of W/m^2. At a fixed TOA flux, the mesn surface temperature would NOT remain constant; we’d be accruing energy and temperatures go up.”
Actually imbalance is a thought experiment created to simplify the effects of a relative change in GMST. To define it one needs to choose a baseline of what they consider “normal” and choose a set of parameters that they believe have changed since “normal.” I took the liberty of simplifying the simplification by converting the results of the imbalance to temperature after reaching that thought experiment endpoint of a unchanging world fixed at current forcing reaching equilibrium.
Ron Graf,
???
It’s real easy. When energy in doesn’t equal energy out, we have an imbalance.
Ron-
==>> ” BTW, I admit one day I said 0.6C the next I said it could be 0.65C. ”
I gather that is your answer to the ‘oft repeated question?
Ron,
Brandon has already responded, but you’re quite simply wrong. An imbalance is the difference between the amount of energy coming in and the amount going out. It does not require a baseline or any sense of what is normal. You still haven’t defended your 0.6/0.65.
@RonGraf
“Since the temp trend is the same in both rural and suburb and city locations the logical conclusion would be that UHI-MS is universal in the land record, not non-existent. True/False”
If the trend exists at all locations why would that indicate bias everywhere instead of reality?
> I gather that is your answer to the ‘oft repeated question?
RonG can correct me if I’m wrong, but I rather think he claims that his number is a possibility which he tries to accrue into a plausibility by citing references and asking “what’s up with” questions.
As far as I can see, AT and the Mosphit are telling him that with more plausible power comes greater calculational responsibility: not only does he need to handwave to this or that paper and ask “but what about” this or that dataset, he needs to come up with an alternative way to account for the data. If there’s a reason why we think it’s 1k, there should be accounting reasons why it’s any other number too.
Numbers seldom stand alone.
There are side issues like “but we haven’t checked for A – indeed we did” and “but here’s what says B – no here’s what B says” kind of things, but these are more related to credibility. I advise against epiloguing too much on them, if only because that usually leads to a world with more ripped off shirts.
BrandonRGate says: “It’s real easy. When energy in doesn’t equal energy out, we have an imbalance.”
Although balance is indeed energy in equal to energy out, that is non-informative in a chaotic system in which equilibrium is never reached and ideal equilibrium is a matter of conjecture and thought experiment. For example, a medium volcano like Mt. Pinatubo produced enough and reflective aerosol to likely curtailed the incoming short wavelength sunlight into a temporary balance with outgoing longwave. I will not recite the list of variables affecting ISR and OLR, but it’s long. The primary variable we can measure and reasonably project is ocean heat content. So in an attempt to entertain a discussion on what “normal” represents, I simply showed that OHC was at an abnormally low level in 1850 in a 800-yr running mean. And it takes about 800 years to change 90% of the OHC [rough guess].
Speaking of rough guesses, that is only what one can do without clearer data and validated models. I repeated my original claim a couple of comments ago that I was stating that an ECS of ~1K was not outside of the distribution of possibilities. IMO I presented good circumstantial arguments supporting my claim [But teh modulz -W], as was attempted on the prior post. Uncertainty cuts both ways.
Ethan Allen, I was correcting anoilman’s point that the city heat was not enough to increase GMST. I was agreeing with him an saying that is a good reason why we should not use such temperatures in our labeling of GMST. It is a case of poor sampling. Weather stations were never intended to be data for global sampling. I hope that is clear.
Joshua and Anders, regarding the “oft repeated question.” I will let the reader decide if my circumstantial evidence to refute Anders’ claim that 1K is implausible, which is what my original comment claim was. (Others have attempted to twist my claim into other things.) I am not claiming evidence shows likelihood of 1K ECS. I think everyone that follows climate science long enough adopts their own range of plausibility for ECS. The ones that do not change it may be politically safe but logic says that education should continually fluctuate the estimation. After all, science should be continually bringing new evidence.
Pbjamm says:“If the trend exists at all locations why would that indicate bias everywhere instead of reality?”
Again this comes back to sampling. The assumption by Steven Mosher and the “consensus” is that rural stations were pristine from contamination and thus were representative samples of reality. So when rural behaved the same way as urban their conclusion was to say contamination did not exist. I submit that the existence of contamination was never a question. Hansen, Easterling and Karl acknowledge the contamination. Oke, Kim, Hamdi and others attempted to quantify it. Steve Mosher admits that on peak days Tokyo is 7C higher than rural stations at the same elevation and latitude.
The only way that consensus argument can be true that there is not a bias to the land station record is to claim that cities, suburbs and rural towns have not grown in area, density or technologically. I say this is absurd. Population has grown geometrically. We know that residences produce much of UHI / MS by escaping furnace heat in winter, and to a lesser extent HVAC condensers in the summer. The summer also has more daylight which to build and trap heat in roads, lots and structures, which are growing in number continually globally.
I am admittedly am relatively new to the CS scene. Anyone that has studies of UHI – MS has been analyzed with controls against the possibility of rural contamination please post them.
There can be a TOA imbalance for a decade or more without a change in GMST. The 1-2% of the energy imbalance that in the long run goes to the atmosphere can easily hide elsewhere for a while. However, this chart https://www.nodc.noaa.gov/OC5/3M_HEAT_CONTENT/heat_content2000m.png
represents about 80% of the total energy imbalance, and this heat can’t hide anywhere else.
Since 2005 the energy content of the 0-2000m ocean layer has increased by 12.2*10^22 J.
That alone equals 0,76 W per m2 of the Earth’s surface. Assuming that this represents 80% of the total, suggests a total energy imbalance of 0,95W/m2
The 2005-20015 average TOA imbalance of all CMIP5 rcp4.5-8.5 models is 0.87 W/m2. If real world imbalance follows that of the models, can We then assume that the ECS is similar? About 3, or..??
> I repeated my original claim a couple of comments ago that I was stating that an ECS of ~1K was not outside of the distribution of possibilities. IMO I presented good circumstantial arguments supporting my claim [But teh modulz -W], as was attempted on the prior post. Uncertainty cuts both ways.
I’ve edited your “but teh modulz” tu quoque, RonG. I’ll also correct your appeal to mere possibility. The circumstancial evidence you bring argues for something stronger than being “not outside the distribution of possibilities.” For instance:
That counterfactual assumes a plausibility, not a mere possibility. There are more than 20 occurences of “plausib” on this page, many from your own comments. When you speak of “circumstancial evidence,” you appeal to plausibility more than to some lack of impossibility.
In other words, you can’t backtrack to “but it’s not impossible” when your interlocutors question your “but it’s plausible.”
You need to choose between whataboutism, which is common but not that interesting (I may be biased here, as I take no interest in following Tony’s), or an alternative hypothesis, which requires you to show your work.
Hope this helps.
Ron,
I really do think you need to correct this
This is simply wrong. If this is what you think an imbalance is, then you’re not defining it in any standard way and it’s not really possible to have a serious discussion if you simply redefine well-defined terms to mean something completely different.
Olof,
You’re quite right that variability means that a given TOA imbalance doesn’t translate into a certain amount of surface warming over short timescales; as you say, we can sustain an imbalance without associated surface warming for short periods of time.
I think most would agree that 3 is reasonable ballpark figure, with a likely range of 2-4C, or maybe 1.5-4.5C. What can say about the models is that a period of a decade or so with slower surface warming does not suddenly invalidate the models since the variability in the models is not necessarily in phase with what happened in reality and a model ensemble mean will likely average over the variability to produce something more indicative of the underlying forced component.
What you’ve said is maybe a bit too strong, but if the real world imbalance follows that of the models that is certainly another factor indicating that they do a reasonable job of representing the real world,
Ron –
You say: “The only way that consensus argument can be true that there is not a bias to the land station record…”
I’m not sure what you mean there by the “consensus argument.” Maybe you could clarify? Are you saying that the “consensus argument” is that there is no UHI effect? My understanding is that the “consensus argument” rests on a view that the impact of UHI is likely within a certain range.
Building from that understanding, I’m struggling to understand here, and my confusion may just be an inherent limitation imposed by my own deficiencies….but my understanding is that Steven’s questions suggest that even if there were a UHI effect that is larger than what his analysis reveals, there would still be a mathematical question as to the plausibility of your argument about the magnitude of the impact of UHI on the overall record. If I understand the convo (which I acknowledge is not particularly…er…plausible)… then you haven’t directly addressed that point.
Doesn’t the proportion of sea vs. ocean in the overall record place a limitation on the potential effect of UHI?
From Steven –
–snip–
Even IF 30% of that were UHI… you’d STILL have 1C of warming.
Now, Assume that ALL OF THE 1.5C land warming is UHI.
Calculate the global average..
done?
.63C
–snip–
So does the “consensus argument” assume that rural stations are pristine, or that the rural stations are pristine enough to ensure that if UHI biases the record, the bias is by less than somewhere between 30% and 100%? Does it assume that there is no bias, or that the bias is likely bounded within a range?
OK. I’ll bud out now.
sorry….land vs. ocean, not sea vs. ocean…
So…All stations show warming and are therefor contaminated by multiple independent micro forces all working in the same direction. Is not “the whole world is warming” also an adequate explanation for the warming? As for you repeated return to Tokyo, so what? It is hotter in the cities, there is a famous and great song about this, but cities are part of the world as much as farms are. If they are getting warmer that is relevant. The question is what % of stations are in the cities and therefor how much of an influence does that extra warming have on the overall trend? If I understand Steven Mosher correctly the answer is that the diff is negligible and easily adjusted for if you are so inclined. I feel like we are going in Tolian circles again.
Anders, you are correct that I oversimplified and should have supplied my definitions not conflicting with terms of art defined differently. I think people understood that I did not want to keep converting to W/m2 to deg C cluttering with math. You created this blog to bring the discussion to a wider audience, after all.
I supplied my use of imbalance in terms of loading the starting assumption and current assumptions about TOA radiative flux in relation to ocean heat content. In energy terms (standard use) your claimed 0.1 W/m2 (actually translates to 0.024K) as the mean imbalance in some unspecified interval of time around 1850. I say this is implausible if the ocean heat content was at a 5000-yr low and experienced a minimum inflection dip in GMST just 200 years prior. If EfCS is hypothetically 1K and the rise of GMST was only 0.6C (60% of EfCS) our current imbalance could be the same as it was in 1850, the warming just cancelling out the increase in GHG forcing.
The equation then becomes \lambda = 3.7/ [ – (0.7-0.7-2.3)/ 0.6] = 0.97K EfCS
Have the authors responded to your request for a similar explanation?
Ron G,
It is true our weather recording system was not originally designed for climatology. That’s what makes it fun. Lots of smart people have spent lots of time pouring over the data to see what useful information can be extracted from it. They invented lots of adjustments to make sense of the data.
What’s the alternative? Throw your hands in the air and give up? That’s not how a true scientist thinks. One never has a perfect measurement. In every experiment one always has sources of measurement error, often from sources one did not consider before hand. This is standard procedure.
Steve Mosher from what I understand says the BEST only adjusts for Tokyo (and maybe a few other cities). Adjustments for UHI have changed through the years. Hansen tripled the adjustment at one point. I believe GISS operates under protocols set by Karl currently. There is very little adjustment. Also, when stations are moved from urban to rural there is an adjustment called a break point to cool the record. All the breakpoints cool while UHI warms it gradually is the common allegation. As I said I am relatively new and have job outside climate science, (as Willard put I’m just a messenger). The proposal to study UHI / MS using DTR and meteorological ques is my idea. I don’t know if others have had it. I credit Steven Mosher for giving me the tools to come with it by challenging me to do research on UHI in January.
Gator, I agree that is why I used Ocean2K and Vose for my purposes. The satellite record does not have an alleged systematic forms of bias, a constant miscalculation. I have only heard allegations of poor precision but it tracks other indexes in high frequency very well.
Correction: breakpoints when stations are moved out of a city do not cool the record. They supposedly cancel cooling bias that would be transferred in the absence of an adjustment.
Ron,
I still don’t think your explanation makes much sense. If we had a large positive planetary energy imbalance, we should have seen some indication of that in the sea level rise. It’s not there, as far as I can see.
From what I understand sea level rise had a pause or reversal in the LIA, which supports a cooling. Otherwise SLR has risen continuously at Holocene equilibrium, with glaciers in decline since Ice Age. Let’s not get too off on SLR. Do you accept LIA or cooling per Ocean2K McGregor et al (2015)?
Ron,
No, because it is pretty fundamental to the existence of a substantial planetary energy imbalance. I’m still not convinced that you understand what a planetary energy imbalance is. For example,
If it is cooling, how can the planetary energy imbalance be positive?
When the oceans are colder less OLR radiates. This puts TOA out of balance with incoming. I can’t believe you made me explain that. You do agree with that?
No, because it doesn’t necessarily follow. If we’re cooling, then we should be losing energy. That it gets colder does not imply that we should end up with a positive planetary energy imbalance.
Chaos does not violate the 1st law of thermodynamics in a deterministic physical system, Ron Graf. “Ideal equilbrium” is a strawman. Try dynamic steady state instead.
Climate is mainly about chaotic attractors.
“Steve Mosher from what I understand says the BEST only adjusts for Tokyo (and maybe a few other cities). ”
1. Wrong.
2. The adjustment code works for any city. even cities that have negative UHI ( yup some do, there are even papers on it )
“Adjustments for UHI have changed through the years. Hansen tripled the adjustment at one point. I believe GISS operates under protocols set by Karl currently. There is very little adjustment.”
1. hansen has changed his method a couple times.
2. he doesnt use “Karl” protocals
3. he uses nighlights. his last paper did some severe testing of it. no change
” Also, when stations are moved from urban to rural there is an adjustment called a break point to cool the record. ”
1. Wrong.
2. First we Break series where the data ( metadata) and data stream indicate significant break.
This can happen for many reasons.
3. Next, stations are given a quality rating based on their divergence from the neighbors
4. lastly, quality adjustments are iterated to minimize the error of prediction.
“All the breakpoints cool while UHI warms it gradually is the common allegation. As I said I am relatively new and have job outside climate science, (as Willard put I’m just a messenger). The proposal to study UHI / MS using DTR and meteorological ques is my idea. I don’t know if others have had it. I credit Steven Mosher for giving me the tools to come with it by challenging me to do research on UHI in January.”
you have to be much more clear what you mean by “studying” UHI/MS by using DTR.
having a hypothesis helps.
Ron.
I keep answering your questions and you wont answer my simple one.
nevertheless I will continue. Readers are free to draw any conclusion they want
“If the one percent one samples is contaminated with un-natural heat bias the calculation for the whole globe will be thrown off. That is what the UHI point in this thread is about.”
1. The point of testing the whole globe is to see if the bias is actually large enough to rise above
the noise floor. We tested that. We could not detect a bias. That is DIFFERENT THAN claiming
there is no bias.
2. the WHOLE POINT of this UHI thread is your claim. I will repeat it.
A) Anders is wrong when he claims 1C of warming, because UHI may drive this down to
.6C
3. You have yet to defend you claim against my observation that you can only get to .6C
by assuming ALL of the land warming is UHI. simple math.
“Steve, did I miss Watts (2016)? Can you provide a link? I was not assigning homework. I’m simply asking questions. I am on your side and promote your answers to others if they make sense. You’ll not have to repeat yourself if you gave clear explanations for the following:”
yes you missed it, This July will be 4 years waiting. the first publication was in 2012.
That was pulled back because of a botched job. I asked for the data and he refused.
next he discussed it at AGU in 2015.
Read victor on it
http://variable-variability.blogspot.com/2015/12/anthony-watts-agu2015-surface-stations.html
Let me tell you the history behind micro site. While I was researching I found CRN.
I told anthony about it. They I directed him to LeRoy’s rating system. CRN1-5
Then I did a quick study and found nothing. Anthony wanted more data. Then
he wrote his 2010 paper and found nothing.
Then we took his ratings and wrote our paper and found nothing.
Over the years there has been a debate about what the 1-5 stood for. I talked to Leroy’s
partner. They had done one small field test to estimate the microsite effect.
they dound the AVERAGE BIAS to be .1C thats right .1C
The ratings 1C-5C basically represent the worst bias seen at a single point in time
during there short test. They didnt test all the various levels.. Basically the 1-5 scale
is untested and uncalibrated. Its a guide. A guess, at the RANGE of variance you might
see. So a CRN 5 will have a average bias of .1C, but some hour of some day of some month
you might see a 5C SPIKE.
In short, you like anthony have concluded the bias has to be big. but the rating system was NEVER TESTED… that means , you guessed it, the rating system maybe wrong. That is bad stations according to rating, might not actually be bad.
“1) Since you point to an example where bare land causes UHI, and whereas agriculture is more likely to clear tress/vegetation in fertile areas than plant and green arid landscapes? True/False”
A) My example was that bare land, bare rural land, can be warmer than urban land,
In Portland ( see Sailors study ) the urban areas with trees were actually cooler than the
rural.
B) WRt your question study LCZ.
“2) Once lands are developed they stay developed even if they may become depopulated. Buildings roads and parking lots remain. Although development is uneven in each location as a function of time, the globe as a whole is gradually cultivation, road building, skyscraper constructing and airport expanding. True/False”
But to answer your question, read my paper where I detail how our rural locations havent changed.
“3) The weather station record was created to give information to city populations and farmers, not to give controlled representative sampling of earth temperature. True/False”
A) The motivation of record keepers is mixed.
B) You can contruct a stratified sample from the convience sample if you like.
C) 90+ % of the variance in monthly averages is explained by two variables
latitude and alitude. THIS FACT ALONE shows you how little room there is for
UHI to play.
4) That UHI and MS existed to the same extend in 1850 globally as it does today makes no sense. True/False
A) open question, but most likely true
5) Since the temp trend is the same in both rural and suburb and city locations the logical conclusion would be that UHI-MS is universal in the land record, not non-existent. True/False
B) false. First no one claims it is non existent in the record. It is there in the raw record.
Second, Even then it is at the birder of detectability. basically you have a lot of noise
in trends and a small bias is easily swamped. And as you see adjustments can remove it.
“6) The unexpected result is thus that UHI is found in rural records. This unexpected hypothesis can be tested by isolating wind/rain temp records from clear/dry for trends in temp and more importantly, DTR. Because the latter is a fingerprint for UHI. True/False?”
A) UHI is detected because of the DIFFERENCE between rural and urban
B) wind rain records. You typically dont have this data, but its clear that they moderate UHI
this is WHY you can look at the worst days and find 3C, but for the rest of the year you
find nothing.
C) DTR is not a fingerprint of UHI. UHI can cause a change in DTR but every change in DTR
is not due to UHI. logic fail.
7) Number 6 has not been investigated. True/False.
number 6 hasnt even been clearly articulated. First we identify UHI as the Difference
between urban and rural. So, its kind of silly to say that there is UHI in rural.
UHI = U-R
8) Rather than assign you work I offer to help recruit for you and support such work. True/False
Thanks but I don;t need cheerleaders
BTW, I admit one day I said 0.6C the next I said it could be 0.65C. [But IPCC. -W]
That good
Now follow this
1. ELi said 1C
2. You said .65C BECAUSE 1/3 of the land record could be UHI
3. You came here and said .6C
4. Lets work with .65C
5. SST warms by .9C it is 70% of the total .9 * .7 = .63
6 SAT warms by 1,5C.
7. You claim 1/3 might be UHI.
8 1.5 * .33 = .5c
9. 1.5 – .5 = 1C
10 Under your argument the land has warmed by 1C, .5C is due to UHI
11. The land is 30% of the total
12 1C * .3 = .33C
13 .63C(SST area averaged) + .33C(SAT area average ) = .96C of warming
14. .96C does not equal .65C
So when anders said 1C, you could have done the math right and asserted :
Given estimates for UHI, the actual number could be .96C
And then, you and I could have a discussion about your 1/3 claim.
heck I even offered you the option of backing off to .9C or attacking SST
I have no problem discussing the possibility of 1/3 of the land record being UHI.
Look Victor and I and zeke discuss anthiinys work and mckittricks work.
They maybe wrong but they are not lunatics. so we can have that discussion.
But your claim would require 100% of the SAT record to be UHI.
Thats a physical non starter.
Steve, I will try again because you are not addressing what I consider to be the central points of my statements and instead focusing of my grammar.
1) You still give no clue as to how much your overall adjustment for UHI is. What you are saying now, that your code catches it in any city, even ones with negative UHI, which I think you mean warming the record to make up for when you think a city had more UHI in the past. Negative UHI relative to rural is yet another hint that UHI is far more widespread then assumed. In the past you have stated that BEST does not adjust for UHI. I will look your quote saying that if you dispute it. If you meant something else you were misleading us. You said you only adjust for the biggest cities.
2) How can you say you adjust for UHI when you claim it is so tiny as to be insignificant? You said you could not find it in your experiments. You said Karl CRN experiments disproved UHI. If what you really meant is that the bias trend from rural contamination is the same rate found in city contamination why don’t you say I am correct in hypotheses that UHI has universally contaminated the record? Your logic does not seem sound.
3) If your claim is as Eli’s equation made clear, that the temperature trends are identical from pristine rural to urban, and we know that cities have UHI bias, the Occam’s Razor conclusion would be that rural stations are biased but still in their infancy in building urban sized UHI. Development is universal in rate in all geographies from the global perspective. A conclusion that the zero differential exonerates urban is unsound logic.
4) I concede that I do not know the code used for Hadley versus GISS except in general terms. I did see in Parker(2010) that the handling has varied wildly through the years. I believe that the USHCN code was controlled by Karl in 2010. If you say Hansen is now that does not change my point. Picking nits.
5) I true pristine environment would be rural stations on windy and rainy days. One could use this as a baseline for trend comparison of temp trend as well as DTR trend. The later would be an important metric on its own in independently diagnosing AGW is there is a signal. If there is not a signal then all DTR trend is attributable to UHI / MS. In either case, once the baseline for true pristine DTR trend one can use the differential from that trend as a direct proxy for UHI / MS. Do I need to patent this or is it too late now?
Did you read Anthony’s paper from this year? Is it published yet?
Anders, let me have another go at this. Climate Equilibrium, thermal stasis is never achieved in a chaotic system, except in transient instances, thus a radiative imbalance exists. To define a radiative imbalance in terms of energy one must choose the time interval to investigate. in order to diagnose an imbalance for that period one can look at the flux in the primary energy reservoir for earth, which is the oceans. Ocean heat content OHC is a reservoir for energy to store or dispense depending on whether the imbalance is positive or negative, respectively, over a given interval. When the OHC increases the OLR also increases as to attempt to cancel the imbalance. However this adjustment is a very dampened by the fact that OHC is so large.
ISR and OLR Assumptions:
1) The Holocene Thermal Optimum occurred 6k to 8k years ago.
2) The HTO was mainly driven by a peak in ISR about 9k-10k years ago from orbital influence and decrease in albedo which was then lagged by OHC for ~2k years.
2) GMST has fluctuated, but on a downward trend, since the HTO due to orbital cycle that had reached it’s optimum about 10k – 12k years ago.
4) For the 1250-1850 interval OHC dropped reflected by a proxy temp drop of ~0.7C relative to a previous 1.2k year baseline.
5) This from relative drop one can infer a negative imbalance driven by a drop in ISR.
6) The drop in ISR was likely due mainly to transient influences, including increased volcanic activity and decreased solar activity and to a lesser degree the orbital degradation of the aging interglacial.
7) The drop is confirmed in Central Europe Temperature records CET, ice cores, glacial expansion and sea level drop.
8) GMST reached a statistical minimum at ~1650, implying that OHC was dispensing more energy through OLR than incoming ISR in a negative radiative imbalance.
9) Although the GMST began a slow rise there was likely still more OLR flux than ISR but the later increasing, narrowing the imbalance resulted in higher GMST. This would account for the continued decline of the Ocean 2K proxies up until the 20th century.
10) Plotting radiative imbalance from 1250-1920 would likely show a statistical cross-over from negative to positive sometime from 1850-1920, making your assertions in the post of a small, 0.1K radiative imbalance plausible.
11) The radiative imbalance reached in 1850 was with low ISR and low OLR relative to the historical baseline from year 0-1000AD.
12) Although the equation in the post would for EfCS would thus become 1.3C with a 0.6C GST rise this would falsely assume that all of the rise was due to GHG forcing. Since the a return of the solar and volcanic activity profile of the 1st millennium AD would infer a 0.4C-0.5C rise as well, leaving off 0.2C to account for aging of orbital influence.
13) If any of these assumptions or the ones about UHI are even half correct a 1C ECS is very plausible.
Ron Graf,
*IF* billions of invisible farting unicorns exist, anything is plausible.
Ron,
What you seem to be arguing is that you can construct an argument in which an ECS of 1K is somehow possible. Yes, even the IPCC doesn’t rule it out. However, being able to construct a scenario in which it might be possible, doesn’t make it likely. I also don’t really agree with your scenario since it seems to be require that the planetary energy imbalance in the mid-1800s was much bigger than I’ve seen suggested before and that the warming we’ve experience is considerably smaller than seems likely. So maybe not imposssible, but not really probable.
Global temperature anomalies are area weighted. A major trend in the 20th century has been the emptying of the countryside in North America, Russia and Europe if not elsewhere where population has moved from rural areas to urban and suburban ones. Thus, if there is an UHI effect, there is also an anti-UHI which area weighted is larger. Moreover, within urban areas, peak central city density has decreased almost without exception (Tokyo probably being one of them).
@Steven Mosher : “So a CRN 5 will have a average bias of .1C, but some hour of some day of some month you might see a 5C SPIKE.”
RG is focusing on the anomaly rather than the trend. Seems a great deal like the obsession with 1998.
@RG : “Steve, I will try again because you are not addressing what I consider to be the central points of my statements and instead focusing of my grammar.”
If that is your take away from Mosher’s comment then I fear this thread is bound for Nowhere.
Eli, my point was that even if the populations leave the countryside they leave their buildings, roads, parking lots and drainage systems. All these things retain heat or reduce evaporation cooling that has steadily grown over the 160 years. You make my point though that UHI calculation is confounded by counting population density. They count by city lights in aerial night-time imagery. The land changes that were made is what causes UHI, lingering human bodies.
Anders, your insistence of my making my intuitive argument mathematically was helpful in me realizing that there are two separate imbalances; radiative and planetary. The later is controlled by the ocean storage of heat content which builds or diminishes as a function of time and net radiatve imbalance in that given time. I was confusing the two metrics as one. Thus, even though your equation numbers are likely correct for delta radiative imbalance and hindered OLR due to GHG, your equation assumption that all of the delta temperature in the 160 years has no delta ISR component is a faulty assumption. The paleo record implies that ISR was in negative anomaly in the 1250-1850 800-year interval as compared with its prior 1200-year interval. This implies a likely rebound in ISR which is not in your equation. The quantity of the rebound is implied by the negative anomaly in SST, which is a rough proxy for OHC, of about 0.7C according to McGregor (2015).
In addition to my demonstration that one cannot attribute all of the 160-year rise in GMST to GHG forcing, we have high uncertainty as the the rise itself. One component of the uncertainty is sparse sampling in the early half and another component is poorly quantified UHI / MS contamination in the entire land surface record. UHI is not only supported by studies but also circumstantially evidenced by the divergence of the satellite record TLT trend to GMST trend.
Ron,
What is ISR?
Incoming shortwave radiation
Ron,
The change in solar forcing is probably small. Since 1750, it is estimated at +0.05W/m^2. It could be slightly larger sine 1850, but it’s still going to be at least 10 times smaller than the change in anthropogenic forcing. It doesn’t really change anything.
Ron –
I know I said I’d but out…but I never live up to that claim..
I am curious to know whether, at least, you have addressed this:
–snip–
You have yet to defend you claim against my observation that you can only get to .6C
by assuming ALL of the land warming is UHI. simple math.
–snip–
I assume that you have and I’m just not smart or attentive or skilled enough to see that you did. At any rate, is there any way that you could answer that point in a very simplified form?
Actually, I wrote bud out, before writing but out, but what I meant was butt out. So maybe I’m off the hook for not living up to what I said I’d do. :0)
Anders, what kind of proxy gives evidence specifically for the solar forcing that reaches the surface? My ISR included only that radiation making it to the surface after reflected radiation from volcanic aerosols.
Secondly, you have yet to dispute there was a negative OHC anomaly. How could this be accounted for if ISR was constant over the 1250-1850 interval? The only alternative to account for drop in OHC would be increase in OLR. There is no hypothetical mechanism for that I am aware of.
Ron,
My info is from the AR5 radiative forcing diagram. Change in solar forcing was +0.05W/m^2 since 1750.
What do you mean by an OHC anomaly. Relative to what?
Ron
“Steve, I will try again because you are not addressing what I consider to be the central points of my statements and instead focusing of my grammar.”
1. Ron. I have asked you repeatedly to defend or amend your claim of .6C.
2. that defense would include math.
3. I have answer all of you questions. you have yet to answer that question.
4. please answer the question
1) You still give no clue as to how much your overall adjustment for UHI is. What you are saying now, that your code catches it in any city, even ones with negative UHI, which I think you mean warming the record to make up for when you think a city had more UHI in the past. Negative UHI relative to rural is yet another hint that UHI is far more widespread then assumed. In the past you have stated that BEST does not adjust for UHI. I will look your quote saying that if you dispute it. If you meant something else you were misleading us. You said you only adjust for the biggest cities.
1. There is no overall adjustment for UHI. I find it silly that you slag me for not answering
a question you never asked while refusing to show your math on .6C
2. The code doesnt catch any city. it detects divergences both positive an negative,
3. Some cities ( in rare cases) cool relative to their rural neighbors. Typically in
desert areas. GIYF
4. You didnt even do a lierature search on negative UHI.
5. We dont adjust for UHI EXPLICITLY, if a divergent trend is found it is handled regardless
of the cause.
6. you can check but I usually say EXPLICITLY, meaning there is no code that says… here is
a city.. adjust. rather at the end of the process we can look and see that cities are usually
downweighted.
2) How can you say you adjust for UHI when you claim it is so tiny as to be insignificant? You said you could not find it in your experiments. You said Karl CRN experiments disproved UHI. If what you really meant is that the bias trend from rural contamination is the same rate found in city contamination why don’t you say I am correct in hypotheses that UHI has universally contaminated the record? Your logic does not seem sound.
1. UHI can be large at a single site ( take reno as an example) but with area averaging
it becomes small. THAT is the whole fricking point. You dont get area averaging.
That is why you made the .6C blunder.
2. no experiments disprove UHI. I have repeated said, every researcher on this says that
UHI is real. It just doesnt impact the global average in a material fashion.
3. UHI cannot infect a site that has no human settlements, No pavement, no people.
Our definition of rural was this “zero evidence” of human changes to the environment.
4. When we compare tokyo to its rural neigbors we can see a difference. you want to argue
that they are BOTH infected. The evidence points to one being infected. the one
with the higher temperature.
3) If your claim is as Eli’s equation made clear, that the temperature trends are identical from pristine rural to urban, and we know that cities have UHI bias, the Occam’s Razor conclusion would be that rural stations are biased but still in their infancy in building urban sized UHI. Development is universal in rate in all geographies from the global perspective. A conclusion that the zero differential exonerates urban is unsound logic.
1. Development is NOT unversal in all geographies.
2. In our study, 15,000 of the 36,000 sites were classified as Very rural.
A) by Modis there was ZERO evidence of human development.
B) in 1900 the “population” was less than 1 person per sq km. ( its really zero)
C) in 2005 the population was less than 3 people per sq km ( its really zero)
3. 21,000 sites were urban.
A) by modis there was at least some evidence of human development within 10km of site
B) the 1900 population was 31 people per sq km ( median)
C) 1/3 of the sites we rated as urban had populations of less than 10 people per sq km
( hansen would call these rural)
D) by 2005 the median population of urban sites increased to 130 people per sq km
E) tokoyo by example has 6000 people per sq km
4. The logic is pretty sound. We selected sites that are very rural. Pristine. No human built enviromentwithin 10km of the site location ( And yes we checked out to 20km), We then cross checked
using historical population . NO people in the past and no people today. They are not
as you speculate with zero evidence.. “in their infancy of UHI” In our entire database there
were 15,000 sites ( out of 36K) meeting this criteria. The UHI hypothesis is that BUILDINGS AND POPULATION cause UHI. So we picked 15,000 places with no signs of building and no fricking people. And Now you want to argue there is some magical rural UHI that must be infecting them? How exactly does one “know” this? by comparing the rural to WHAT?
the Ocean? they warm slightly faster than the ocean as expected by physics..
Against these 15K we compare 21K sites that started as a combination of
rural ( population less than 10per sq km) and expanded to urban densities,
We find NO DIFFERENCE.
5. The UHI hypothesis is this: buildings and people warm the local air more than it would
be otherwise. The science shows this. Compare Reno to the areas outside the city.
The city warms faster. So, you hypothesize that the GLOBAL average must show this
this effect. After all the global average includes cites. Well we tested that. The hypothesis
is wrong. NOW, you want to claim that UHI must infect rural stations as well. ie rural must be urban. That is you want to argue that the 15K pristine sites we used, sites with no signs of
building, sites with no people, suffer from Ghost UHI.
4) I concede that I do not know the code used for Hadley versus GISS except in general terms. I did see in Parker(2010) that the handling has varied wildly through the years. I believe that the USHCN code was controlled by Karl in 2010. If you say Hansen is now that does not change my point. Picking nits.
A) you have zero clue and should stop while you are behind
5) I true pristine environment would be rural stations on windy and rainy days. One could use this as a baseline for trend comparison of temp trend as well as DTR trend. The later would be an important metric on its own in independently diagnosing AGW is there is a signal. If there is not a signal then all DTR trend is attributable to UHI / MS. In either case, once the baseline for true pristine DTR trend one can use the differential from that trend as a direct proxy for UHI / MS. Do I need to patent this or is it too late now?
A) You cant build a record from only windy and rainy days. lack of data and what the hell
would you say about desert sites. Long ago I decided the best thing to do was to look
at desert locations. No buildings, no people, no water. Guess what? it matches what you
you find in other locations. Further we are looking for trends so we need complete records.
B) DTR trend changes for a host of reasons, like cloudiness, like global warming.
C) because DTR can change as a function of cloudiness you cannot say its soley a function of UHI. basic logic
Did you read Anthony’s paper from this year? Is it published yet?
Did you read victors blog?
##################################
At this point Ron you have one choice.
Answer my ONE FRICKING QUESTION.
1. How, mathematically, do you get to .6C of global warming because of UHI?
Show your calculation.
Given .9C of warming in the SST record.. How do you get to .6C of global warming because of UHI?
Ron’s latest hypothesis Rural must also be infected by UHI/MS
How do we test that? Well, I can see some sense in the MS hypothesis. So we can test this
by looking at sites that are Rural and free from MS.
How do we do that?
1. First we consult the definition of MS. And there we have Leroys criteria for site selection.
This is the criteria used to select CRN stations. These sites are selected because they have
NONE of the known causes of UHI ( pavement, buildings, people) and because the sites
have none of the microsite issues like shading trees, local heat sources, heat sinks,.
here is a picture:
https://i.guim.co.uk/img/media/3f82f741e75a6a07b833e60550578f55d1f040ba/0_13_1022_613/master/1022.jpg?w=620&q=55&auto=format&usm=12&fit=max&s=5c78c23f78528b4b7c97fa9b03d4c0ec
No city. no UHI.
No air conditioners, no parking lot, no burn barrel, no aircraft exhaust,
https://www.ncdc.noaa.gov/crn/documentation.html
Click to access CRNFY02SiteSelectionTask.pdf
Click to access X040_d0.pdf
These 110 stations ( there are actually 230– but some have been discontinued)
can serve as a reference. hence the name. Climate reference network.
The sites are selected to have No known causes of UHI ( buildings, pavement people)
and no know causes of bias at the site, heat sinks, heat sources, shading, etc.
So lets test Ron’s hypothesis that other stations ( rural and urban) will suffer from a bias
that maybe due to UHI or MS or a combination thereof.
Do we have any evidence to support this?
Lets look at the 10 years of data we do have. from 2005 to the end of 2015.
Do we see any difference between Pristine and all the other junk
http://www.ncdc.noaa.gov/temp-and-precip/national-temperature-index/time-series?datasets%5B%5D=uscrn&datasets%5B%5D=climdiv¶meter=anom-tavg&time_scale=12mo&begyear=2005&endyear=2015&month=12
NOPE. In fact the Pristine show a tiny bit more warming ( in the noise)
1. Do the actual data support Rons hypothesis that rural must suffer from UHI and MS?
Nope.
2. 50 years from now when we still show no difference what will the Ghost of ron argue?
“maybe the pristine are corrupt too!”
People love Feynman. People love Popper. If you have a hypothesis and the data says its wrong, then its wrong
Hypothesis A: Urban warms more than rural. IF you only look at rural sites the global warming
we see will be decreased.
Test: Build a Rural only network. test rural versus urban. Rural will be less than urban.
Result: There is no statistically significant difference. ( maybe as much as .14C of UHI per
century… or .21C Bias in the Land record)
Well, that would be the end of it, but not for folks who need to hang onto a hypothesis.
Hypothesis B: Maybe Rural Suffers from sudden infant UHI and Microsite
Test: Build a Pristine Gold standard reference network free of UHI and Microsite
Result: 10 years of data shows NO DIFFERENCE between Pristine and all the “bad sites”
Hypothesis C: Maybe Unicorns are farting on the pristine sites.
Test: wait 50 years and monitor for unicorns. Meanwhile, do nothing about global warming.
Anders: “What do you mean by an OHC anomaly. Relative to what?”
As I explained, McGregor and other proxy evidence suggests that the interval from 0-1250 had a higher GMST than the interval from 1250-1850, a negative anomaly. If there is no proposed mechanism for a higher OLR the only other thing that could cause GMST lowering is non-reflected ISR lowering. This does have several proposed mechanisms including volcanic maximum, solar minimum and waning orbital. The first two are variable and the implied evidence, lower OHC, would presume a rebound at some point unless the ISR was further diminished by increased albedo from advancing ice sheets, (the ice age attractor). I think coal heat and petrol saved us from that.
If the UHI effect was driving global warming wouldn’t this be obvious from the distribution of warming with most warming concentrated in areas of high population?
Instead we get this:
A general warming pattern globally but with more warming in the Arctic than the equator. I wouldn’t have thought there was much UHI effect in the Arctic.
Ron,
What is your source for the 0.7oC drop in temperature between 1250 and 1850? My understanding is the total global drop in temperature is ~0.7oC from the middle to late Holocene (~5000 years ago).
Steven Mosher says:
My hypothesis is not the urban heat more than rural, it’s that rural becomes suburban and in some cases urban. This may not happen to a particular test station you are looking at but statistically over the globe over the 160-year interval there was huge change.
Now, to test this hypothesis one can test the data in the manner the stupid old me came up with, which is to use all the days where UHI was meteorologically nullified by wind and rain and compare it to the days that it wasn’t for DTR trend and Tavg trend.
[But conspiracy. -W] The way science advances is for investigators to create experiments that are reproducible by skeptics. Or, if a grand multi-year engineering project is required to produce the data you invite skeptic red teams in. If Anthony Watts’ team was part of CRN the experiment would have advanced science perhaps. [But Einstein. -W]
What is wrong with my proposed experiment using the meteorological record?
Is Anthony’s paper getting published? Do you know? I believe his claim is that his paper shows UHI bias exists.
Hydrologist: “If the UHI effect was driving global warming wouldn’t this be obvious from the distribution of warming with most warming concentrated in areas of high population…”
You are making the same mistake that anoilman did. We are not talking about UHI warming the surface of the planet. The debate is whether UHI warms where we have measure temperature on land for the last 160 years and thus providing a biased sampling of the earth land surface. And, although the oceans are twice the area, their sampling has been very sparse up until the Argo buoy program started 10 years to get an accurate global sampling of surface and depths to 2000 meters.
Hydologist: “Ron, What is your source for the 0.7oC drop in temperature between 1250 and 1850? My understanding is the total global drop in temperature is ~0.7oC from the middle to late Holocene (~5000 years ago).”
There are multiple studies of ice cores, tree proxies, marine biological based proxies (like those compiled in Ocean2K). See the links near the beginning of the post. Here is an image of the Ocean2K published chart. https://climateaudit.files.wordpress.com/2015/09/ocean2k_recon.png
Rn, it’s tedious to have to say this yet again, but McGregor et al (2015) show no such thing. Their analysis indicates a sea surface (SST) temperature change in the period 801-1800 of -0.4 oC (Suppl. Table S7). If you’re using their data to get the 1250-1850 temp change this would be around -0.3 oC.
BTW, this isn’t so different from interpretations of NH paleoteperature reconstructions which indicates around -0.3 – -0.4 oC surface temperature change in the period around 1000 to 1800 AD, (with much of this occurring by 1100-1200 AD consistent with “recovery” from the so-called MWP) [e.g. Fig 3 of Mann et al. (2008) Proc.Natl. Acad. Sci. 105, 13252] . Relative to the pre-MWP baseline, the cooling around 1800 is unlikely to be more than around 0.2 oC and some of this must be orbitally-forced. Any residual “out-of-equilibrium” negative temperature anomaly of the sort you are basing your argumentation must be small (not much more than 0.1 oC according to published evidence. This still leaves well over 1 oC of warming since the late 19th century (if one factors in the net cooling contribution of man-made aerosols, this warming would have been larger).
Interestingly the McGregor paper that you keep misinterpreting, if correct, would indicate that the MWP didn’t exist on a global scale and would support use of the potentially more accurate term “Medieval Climate Anomaly”.
Ron,
Yes, but this does not imply a substantial positive energy imbalance in the mid-1800s. It just means that it cooled relative to 0-1250.
OK, I can see where you’ve been fooled Ron. Rather than looking at the actual McGregor et al (2005) paper yourself, you’ve sourced “an image of the Ocean2K published chart” from some dodgy website.
Whoever produced the graph you linked to has removed the Y-axis label and given the graph a title “Compiled sea surface temperatures”. This gives the impression that the graph is plotting reconstructed sea surface temperatures. It ISN’T.
Look at McGregor et al 2015 Figure 2a which is the same graph as the one from your link, but with the Y-axis label included. This label says “Standardized anomaly s.d. units”. They’re NOT temperatures. Whoever made the graph you linked to has constructed a tedious misrepresentation.
If you read McGregor et al (2015) you will see that they find it very difficult to convert the standardized anomaly in s.d. units to temperature anomalies for a number of reasons. Based on their linear fit to their reconstructed standardized anomaly data the temperature change from 1250 AD to 1850 AD would be around -0.3 oC. NOT 0.7 oC.
Ron
My hypothesis is not the urban heat more than rural, it’s that rural becomes suburban and in some cases urban. This may not happen to a particular test station you are looking at but statistically over the globe over the 160-year interval there was huge change.”
1. WHERE IS THE ANSWER TO MY QUESTION. HOW DID YOU CALCULATE .6C?
2. We found 15K stations where there is no evidence of people building or living.
not in the past. not today.
3. We compared them to 21K stations of rural becoming suburban, suburban becoming
urban and urban becoming more urban.
4, We compared those that had NO CHANGE with those that became more urban
5. NO FRICKING DIFFERENCE
“Now, to test this hypothesis one can test the data in the manner the stupid old me came up with, which is to use all the days where UHI was meteorologically nullified by wind and rain and compare it to the days that it wasn’t for DTR trend and Tavg trend.”
1. No you cant test it that way because
A) you have NO GOOD RAIN RECORDS for all the sites
B) you have no good wind records for all the sites
C) there is NO FIXED windspeed at which UHI is eliminated.
D0 EVEN AT PRISTINE SITES wind and rain will effect the temperature.
2. DTR is NOT A FINGERPRINT OF UHI
“Why Karl had to build a multi-million dollar science experiment that he controls to back up his pre-conceived ideas for UHI invested in USHCN is obvious. The way science advances is for investigators to create experiments that are reproducible by skeptics. Or, if a grand multi-year engineering project is required to produce the data you invite skeptic red teams in. If Anthony Watts’ team was part of CRN the experiment would have advanced science perhaps. If Einstein had claimed General Relativity was proven in his fenced off lab but the equipment is all broken down now his name would be as well known as the guys with the cold fusion experiments.”
1. Anthony bought off on CRN once I showed it to him
2. he himself has proposed that it be used as the record.
3. The need for CRN was identified back in the 90s by internal critcism of the
system
4. Skeptics WILL NEVER JOIN ANY EFFORT to validate the science.
What is wrong with my proposed experiment using the meteorological record?
1. You havent defined any terms in an objective manner
2. Your understanding of DTR is wrong from the start.
3. the data from CRN already suggests you are wrong. we have been doing the
test since 2002!!!
4. You have not proposed a data source
5. You have not proposed a test statistic.
6. You are proposing a huge amount of work for ME when you cant even do me the f*cking
common curtosy of answering ONE SIMPLE QUESTION. How did you calculate .6C?
Here is what I suspect. I suspect you will not be interested in the answer. i suspect that if the answer doesnt go your way, you will have more homework, more cheerleading, more doubt.
And I suspect that after me working 6 months .. you will still not have the honesty to answer
a simple question; how did you calculate .6C
So.. lets play a game. the earth has warmed 1.5C over land. Per Anthony’s opinion
based on 2% of the spatial data ( the US) based on 25% of the time (from 79 on) 1/3
of the warming is fake. lets extrapolate that!! lets say the US is the world. lets say
1979 to present stands for the whole record.. That means the “real” warming over land
is 1C !! 1/3 bias!!!
Whats the global average?
.96C
Not .6C
Not .65C
.96C
Is Anthony’s paper getting published? Do you know? I believe his claim is that his paper shows UHI bias exists.
Did you read victors Blog?
Ron, It would be enormously helpful if you would show the data sources and answer direct questions when asked. You’ve been asked five or six times over five days where you got your 0.7 oC SST temperature change over 1250-1850. That particular issue could have been sorted out immediately if you’d had the courtesy to give a direct answer when first asked. If you have to be so evasive in order to “protect” your ideas then you might question whether those ideas have much merit. And why not simply answer Steve Mosher’s question that he keeps asking you about your 0.6 oC temp rise?
It’s also not a great idea to obtain “information” from sources that misrepresent other peoples data.
“What is wrong with my proposed experiment using the meteorological record?”
1. the requirement to look at only rainy days, biases your spatial sampling and temporal sampling.
A) for example.. if we expect to see GW to show up in winter seasons where it snows
and doesnt rain, you’ll not have data from high latitude places in cold seasons
B) Your test will give preference to locations ( like the tropics) where we see more rain
but also where we expect global warming to be less .. see polar ampflication.
There is one possible way to look at it. using reanalysis data you could categorize the world according to average monthly windspeed ( but DIRECTION really matters for UHI) and average
rainy days… and then see if there was any systematic difference, but the confounding of the
effects of GW with location makes it highly suspect.
Most rainy city in the US.
Buffalo.
167 rainy days per year.
http://berkeleyearth.lbl.gov/locations/42.59N-78.55W
Trend?
Higher than the global average
windiest city in the US
Dodge city
http://berkeleyearth.lbl.gov/stations/162128
Trend slightly lower than the Global.
Here is the other problem with selecting for Wind and rain.
Spatially this tends to give you places near to coasts.. Think the UK, think, british columbia,
Labrador etc.. As we have noted Ron… The Land generally warms more quickly than the Ocean
And we know that Coastal locations will see less warming than inland locations.
So being by a coast can mean More rain and more wind ( think San Francisco, think chicago)
And ALSO lower trends that are closer to the SST series..
So by using wind and rain to select stations… You introduce a spatial bias and select stations
that are closer to coasts and will as a result show warming more in line with the 1C of the ocean.
Further, being by a coast ALSO impacts your DTR. One effect we found updating our
series was this: Distance from coast did impact noise in Tmin and Tmax
But you can find inland cities with a lot of rain
halle has 266 rain days per year
http://berkeleyearth.lbl.gov/locations/50.63N-11.41E
Opps.. its average is inline with the global average
Koln 263 rainy days a year .. located away from a coast.. so thats good
http://berkeleyearth.lbl.gov/stations/14365
Opps it warms More than the global average
Tampere
246 rainy days
http://berkeleyearth.lbl.gov/stations/175445
Opps it warms in line with the global average.
So ya.. In these cities were it rains 2/3s of the year……
The average matches the global average.
So there… Wasted some time..
Hows my answer coming Ron
Chris: “Rn, it’s tedious to have to say this yet again, but McGregor et al (2015) show no such thing. Their analysis indicates a sea surface (SST) temperature change in the period 801-1800 of -0.4 oC (Suppl. Table S7). If you’re using their data to get the 1250-1850 temp change this would be around -0.3 oC.”
Chris, I do have copies of the paper and supplemental and referenced the links in my original comment. You are correct that McGregor chose to use SD units for the y axis, very unnecessarily according to Steve McIntyre, because the raw data was in deg C. So the conversion back and forth was another degradation of the data along with unnecessarily binning the data. But I think Steve was attacking the form and missing the substance, as are you. The comprehensive composite of over 50 independent proxy studies produced the graph shape you saw should have had deg C on the y axis IMO. [And conspiracy. -W]
[But tone. -W] I admit I inferred the -0.7 C/600yrs (not 800) by combining the visual of the chart with the values given in supplemental table S7 -0.41 deg C/1000yrs as the slope from 801-1800 and -0.31 deg/1000 years from 0-2000. As it can be seen on the chart that drop for 2000 years substantially occurred in the 600 years between 1100-1700. I think -0.7C was actually conservative.
Ron,
How is -0.7C conservative if the largest trend is -0.41C/1000yrs?
Anders says:“Yes, but this does not imply a substantial positive energy imbalance in the mid-1800s. It just means that it cooled relative to 0-1250.”
You are correct. I was confusing radiative imbalance with anomaly in paleo-climate trends. Looking at the chart shows an inflection or bottom near 1700. So somewhere beteen 1700 and 1900 the radiative imbalane crossed over from negative to positive (on 50-yr average scale or so). So a 0.1K imbalance for 1850 in very plausible. My corrected point is that a rebound was likely implied by the 600-year drop prior to 1850. So assuming that all the rise since then was GHG forcing was the fundemental flaw of your post, (aside from the UHI likely contamination of land record). https://climateaudit.files.wordpress.com/2015/09/ocean2k_recon.png
BTW, whoever mentioned the Medieval Warming Period not showing up in the chart please note that they chose 200-yr bins and stuck one on either side of the 1100 MWP peak to flatten in it out.
Ron,
The climate is not a bouncy ball. Please explain in what way it can rebound?
The supplemental S7 has an anomally slope for 1-2000CE of -0.31 C/1000 year. I think if one looks at the chart one can see that only 1/3 of the 2000-year interval produced 90% of the cooling. Thus one can multiply -.31 C by 3 to gain the average slope of the most severe drop, the 600 years between 1250-1850. So I amend my estimate to -0.9C Happy?
Ron,
But that’s only 600 years.
Replacing my last comment. If 1-2000 had a slope of -.31C per 1000 years that is -.62 C for the 2000 years. Looking at the chart one can see some rebound after 1700, so truncating that from being figured in the mean slope would produce about -0.7C, occurring in the said 600-years just prior to 1850.
Rebound is what happens when temporarily suppressed incoming shortwave returns to normal when solar activity resumes and volcanic subsides, as I explained before.
Ron,
How does 0.6C in 2000 years become 0.7C in 600 years?
In other words, it’s all forced. The change in solar + volcanic is probably an order of magnitude smaller than the change in anthropogenic forcing. Hence, any potential rebound can’t be more than 10% of the observed warming.
Is there evidence for sea level change sufficient to bolster an argument for radiative imbalance?
BBD: No. Sea level change is 1.1mm/yr through human history.
Steven, I said that as solid testable evidence is provided, because those are the ground rules of science, I will modify my analysis accordingly. I am not religiously invested in any opinion. I am “Mr. objectivity”, as are you, which I admire. I will read Victor’s blog as you assigned.
Something simple maybe you can answer, does Hadley’s CRUTEM4 start at the same place as BE in 1850? If not, why not and how much? How do you account?
Again, looking at landtemp charts, I see BE seems to start lower and I wonder if that is part of your warming. Also, I notice that the lion’s share of warming has occurred since Dec 1978, the start of the satellite record. The satellite warming trend since then is about 1.2 C/ 100 years. And, correcting for the TLT to GMST factor that would be 1.1C/ 100 years. And, if 0.5C of that was going to occur naturally by less volcanic activity and more solar activity then we have 0.6C of AGW and ~1K ECS.
I’m not claiming that is my most plausible estimate but it is not implausible as claimed by Anders’ post.
As far as UHI, why can’t we use the wind and rain data that is not compromised by other confounding aspects? I’m not saying you need to do this. They probably spent millions on CRN and millions more to maintain it. Why not establish a committee made of green team and red team to find if there are signals in the data?
You say now DTR is not a fingerprint of UHI. I think you are just upset. I will let you back off that one. What I think you possibly meant is that DTR is confounded by AGW and other signals. Perhaps, but I think they could be refined and separated.
Sorry for any grief, Steven, nothing I am claiming says your work was faulty but sometimes you are claiming that UHI does not exist, DTR correlation to UHI does not exist, and sometimes you claim UHI is adjusted for. Eli, needs to stop by and give us a comforting quote we can agree on, I think.
Anders, “Ron, How does 0.6C in 2000 years become 0.7C in 600 years?”
The 2000 years includes the first 200 years of a rebound. Take off the rebound and you get 0.7, which all occurs in 600 years on the chart.
We have little evidence of how much volcanic and solar radiation occurred but we have a lot of evidence that something occurred with the ISR and there is no explanation of an increase in OLR to cool.
Kemp et al. (2011) Climate related sea-level variations over the past two millennia:
Lambeck et al. (2014) Sea level and global ice volumes from the Last Glacial Maximum to the Holocene:
Ron – I pointed out to you upthread it was punctuated (episodic) volcanic eruptions. The recovery time for even the largest eruptions was no more than 60 years per the Oceans2K EBM model. Any imbalance circa 1850 was due to eruptions in the first half of that century and has little to do with 1200, or 1300 or 1400 etc. And that imbalance due to volcanic eruptions would have been non-existent by the turn of the century – except that volcanic eruptions don’t end in the pre-industrial era. Since 1850 we’ve had Krakatoa, Santa Maria, Agung, El Chichon and Pinatubo. Your hypothesis is built on ocean cooling due to volcanic eruptions, but does *not* take into account the effects of recent eruptions.
> I’m not claiming that is my most plausible estimate but it is not implausible […]
That’s one way to say your estimate is plausible, RonG. Even if it’s not your most plausible estimate, you still find it plausible. Which means, if you accept what I said earlier, that you need to show your work.
What would be your most plausible estimate, BTW?
Ron.
Answer my question.
“You say now DTR is not a fingerprint of UHI. I think you are just upset. I will let you back off that one. What I think you possibly meant is that DTR is confounded by AGW and other signals. Perhaps, but I think they could be refined and separated.”
Uhi can cause a change in dtr.
But a change in dtr does not entail the cause was uhi.
That is why I have told you multiple times your idea lacks
Merit.
Steve, harsh tone does not dissuade or persuade. I’m interested in sound logical supported arguments. Maybe we break down our questions and acknowledge answers. Also answering questions with openness and sincerity goes a long way to bring acceptance.
I hear you saying that UHI is real, not real, adjusted for, not adjusted for. I will read over your linked pages and try to understand how that can be.
You acknowledge to some degree the proposition is illogical. UHI peaks at 7C in Tokyo and studies reveal a trend in temp and an associated decreased diurnal temperature range that diverges from a rural control in Kim(2002) and Hamdi(2010) and others. But Karl creates a project to disprove this and that becomes the final word. And you say Anthony buys into it?
You yourself endeavored to do a study somewhat similar to what I proposed. I have a copy of your draft. I’m not sure why you wanted to do it, or if you ever did it. Seeing your draft is what got me started in January to realize wind and rain nullify UHI and present a mode of segregation to create a control. It was shortly after that I found DTR divergence from the rural control was noted by Kim(2002). Then Hamdi(2010) focusing in on divergence of DTR trend and quantified it in a longer interval as 2.5X to the rural control. I find it hard to accept that all the wind and rainfall records are lost. If we still have them and it turns out we have very little relative warming on windy and rainy days how could that be explained in an AGW context?
Ron.
“Steven, I said that as solid testable evidence is provided, because those are the ground rules of science, I will modify my analysis accordingly. I am not religiously invested in any opinion. I am “Mr. objectivity”, as are you, which I admire. I will read Victor’s blog as you assigned.”
1. You can’t even answer how you got a global temp of .6 assuming .33% uhi. Why would I expect you to be objective.
2. Note you say that you would modify your analysis not your beliefs. You can’t even show how you get .6c.
Willard: “What would be your most plausible estimate, BTW?”
A year ago I gave a gut guess of ECS = 1.5 here. That was just when Stevens’ aerosol paper came out if I recall. Although I think it is likely a tad higher, maybe 1.8, I also think it is the wrong metric. The critical and relevant time period is the next 70-100 years, with make TCR the more sensible metric, which is about 70% of EfCS in 70 years. We need methods for attacking uncertainty better than just waiting,
Earliest date for IJIS sea ice extent to drop below 11,000,000 km2
1. May 20, 2016
2. May 29, 2015
3. June 3, 2011
Average date
2000s: June 15
1990s: June 24
1980s: July 3
Obviously this can all be attributed to UHI.
Ron… Your link to temperature trends compares CRUTEM4 Mean, to BEST lower 95% confidence, to RSS global land mean.
Perhaps you meant to compare BEST mean to the other Means? Like so;
http://woodfortrees.org/plot/crutem4vgl/to:2016/mean:40/plot/best/mean:40/from:1850/plot/rss-land/mean:40
BBD: I was referring to Jerry Metrovica’s work on ancient texts.
http://www.sciencedirect.com/science/article/pii/S0012821X06000021
That is behind a paywall, but his latest isn’t;
http://advances.sciencemag.org/content/1/11/e1500679.full
Ron
How did you derive .6C for global warming under you assumption of 1/3 UHI?
I believe you owe us all an answer.
“Something simple maybe you can answer, does Hadley’s CRUTEM4 start at the same place as BE in 1850? If not, why not and how much? How do you account?”
1. Start at the same Place? No in 1850 we have more stations than they do.
2. How much? Go get the hancrut data. Go get our data. Do the math, it is slightly
harder than explaining how you got .6C. How can you in all fairness give me homework
when we and CRU have supplied the data for you to do your own damn science.
How do I account? There is nothing to account for. You have two different methods
using two different datasets. In our original paper the difference was on the order of .25C
Accounting for this would require an entire study. Since we use different data we would
have to rerun our method using only their data, which is already homogenized.
but their method can never be run with our data since we can use fragments. There
are a good number of fragmentary records in the early part of the record, stations that go
from the 1840s to 1860s for example. We can use them, CRU cannot. So how do
we account? One siimple way was to see which method worked better given the same
data: First A cartoon version for people who cant explain how they got .6C
Click to access visualizing-the-average-robert-rohde.pdf
Next a test
Click to access robert-rohde-memo.pdf
As you can see
1. in 1850 we have more stations ( and a cooler record than CRU by about .25C
2. CRU error is substantial in 1850
So How do I account? A) we use more stations B) our method is superior with sparse
stations. STILL the difference between BE and CRU at 1850 is minor and our uncertainties
overlap.
There I answered questions for YOU that you could have answered for yourself, had you
taken the time to read rather than Give me homework, And further, you could have answered
for yourself because we gave you the data. We gave you the power. We gave you the power
to do for yourself, because we want you to check our work and find our mistakes. That is what
objective people do. Contrast that with your behavior. Everyone on this thread can see that
you refuse to show your work. How did you get .6C
Ron
1. How did you get .6C? assuming 1/3 of UHI?
‘Again, looking at landtemp charts, I see BE seems to start lower and I wonder if that is part of your warming. ”
A) as explained above, we use more stations that CRU and cover more land.
B) In our methods test we show how CRU errors are larger given the same data inputs.
C) the uncertainties overlap
D) rather than Wonder and speculate, read the science.
“Also, I notice that the lion’s share of warming has occurred since Dec 1978, the start of the satellite record. The satellite warming trend since then is about 1.2 C/ 100 years. And, correcting for the TLT to GMST factor that would be 1.1C/ 100 years.”
A) you assume the satilite trends are correct
B) you neglect the structural uncertainty in satellites
C) you ignore that the satellites match the land and ocean through 1998
D) you ignore that the highest discrpenacy between satellites and land happens
in areas of northern latitude
E) you seem utterly unaware ( see UAH caveats) That there temperature product
ASSUMES constant emissivity for the earth. This is aproblem if snow cover changes
as it has
F) You ignore that the biggest differences occur after the change to AMSU
G) You assume that there is an agreed upon “factor” for going from GMST to TLT.
:And, if 0.5C of that was going to occur naturally by less volcanic activity and more solar activity then we have 0.6C of AGW and ~1K ECS.”
Err no. you are now trying to say .6C because of volcaoes? And solar? What happened to UHI
Further there is no great change in solar. Zip. zero. nada.
Ron
1. Where is my answer? How did you get .6C from your assumption of 1/3 UHI
next
‘As far as UHI, why can’t we use the wind and rain data that is not compromised by other confounding aspects? ”
1. Because we dont have good global and long time series for Wind and Rain.
2. That will lead you to poor spatial sampling and horrible temporal sampling
You will miss deserts, you will miss the poles where AGW says the warming will
be most proniunced. This will reduce the warming trend artificially.
3. I already showed you that in the windest city I could find the warming rate was no different
4. I already showed you your rain ideas are busted.
5. You continue to give me homework and questions to answer without answering a simple one.
“I’m not saying you need to do this. They probably spent millions on CRN and millions more to maintain it. Why not establish a committee made of green team and red team to find if there are signals in the data?”
1. Because Nobody who knows anything about the data would sign up to waste their time.
2. because when you reduce the data ( only windy and only rainy ) you end up with
MORE NOISE and the data s already noisy enough to make the detection of UHI
difficult
3. because no skeptic will ever join a team that might vindicate a record. the GWPF
started there own ‘red team’ no results.
Ron
1. Where is my answer?
next
“Sorry for any grief, Steven, nothing I am claiming says your work was faulty but sometimes you are claiming that UHI does not exist,”
1. UHI exists. What does this mean. This means as I have explained to you many times over at lucias that you can.
A) Pick a city ( A large one helps)
B) Pick a SEASON ( like summer)
c) Pick a time period ( like calm cloudless non rainy days )
and
D) Find a difference between the Temperature of a Urban Thermometer and a rural one.
2. The question is does this Difference manifest itself as a bias in the global record?
and the answer to that is NO.
Whats that mean? It means that if you build a record from urban stations you will
see a trend of X +- and if you build a record from Rural stations you will see a
trend of Y+- and X -Y will give you a spread of something like 0+- .15C
In other words the very real UHI is so small that it lost in the noise. We know its
real by studying individual stations. Huge cites like tokoyo 1 of 41,000 stations
may have a few days in summers where the UHI is 7C.. But when you monthly
average this shrinks. When you average tokoyo with other stations and downweight
it the signal shrinks even more.
“DTR correlation to UHI does not exist, ”
1.ron you keep changing my words and it doesnt help you.
2. IF you have UHI, then you may have a change in DTR
3. IF you have a change in DTR, that doesnt mean the cause was UHI
https://en.wikipedia.org/wiki/Affirming_the_consequent
That is WHY I am telling you that changes in DTR do NOT imply that UHI is present.
very simply.. If you were to look at the temperature at the arctic or antartica where there
is no city.. And if you find a change in DTR that does not mean the cause is UHI.
A simple change in cloudiness will change the DTR. GW will cause a change
The error you have made is affirming the consequent
“and sometimes you claim UHI is adjusted for. ”
1. In GISS there is an EXPLICIT bottoms up adjustment. They identify Urban sites
as urban sites and they adjust them using sites identified as rural. EXPLICITLY
The algorithm knows the site is urban or rural.
2. In BE we do it differently but it has a similar effect.
A) Sites are NOT coded as rural or urban
B) The algorithm just looks for odd balls. 1 station trends at 3C, all its neighbors trend at 1C
C) The stations are given quality weights based on how well they agree with their neighbors
D) AFTER THE FACT…. I can go look at sites ( like tokoyo) and say “Hey pretty cool we adjusted out the UHI” The algorithm that knew nothing about the fact that Tokoyo was 13 million
people, managed to detect that it was an odd ball in the neighborhood and downweighted it
E) What is REALLY COOL is that I can then go look at CRN stations… Was the algorithm
intelligent enough to leave the pristine stations alone? did it correct a reference station?
95% of the time the reference stations are untouched.
So, yes UHI is real, you can find a station that warms more than its neigbors. So can an algorithm. and it does a fair job of fixing it.
Ron
1. Answer my question please?
Next
‘Steve, harsh tone does not dissuade or persuade. I’m interested in sound logical supported arguments. Maybe we break down our questions and acknowledge answers. Also answering questions with openness and sincerity goes a long way to bring acceptance.”
1. this is too funny. You have been asked dozens of times to answer one simple question
2. you have not answered that question
3. I have only question for you, you ignore it. I have answered the vast majority of your questions. even when the answers were already in our writings.
“I hear you saying that UHI is real, not real, adjusted for, not adjusted for. I will read over your linked pages and try to understand how that can be.”
lets make this simple.
you have 1000 people.
You take there weight for 10 years
1 person doubles his weight
999 dont gain any weight
You take an average weight gain and come up with zero.
there is no weight gain in the general population.
you look at the one guy.
His weight gain is real
So weight gain is REAL
But, it doesnt effect the metrics of the general population.
Next.
We do not EXPLICITLY correct for UHI, but it gets adjusted.
Imagine you have an algorithm that compared all 1000 people.. pair by pair
the algorithm is just looking for odd balls.. it sees the fat guy. It is looking
for odd balls.. if it found someone who lost weight that would be an odd ball too.
some on a yoyo diet would be an odd ball. Its not EXPLCITLY looking
for fat guys..
So our code just looks for odd balls. Sometimes its die to TOBS, sometimes instrument
changes,, sometimes UHI.. and it fixes oddballs. It does not diagnose WHY it is odd
it just fixes oddballs. So we dont adjust for UHI Explicity, but UHI gets corrected. there is no
code that says ‘UHI bias is .25.. correct it’ it looks for odd balls. If tokoyo runs 2C warmer
than its neighbors the weighting will correct it to minimize the error.. If reno runs 3C warmer
that too will get weighted.. a different amount to minimize error.
“You acknowledge to some degree the proposition is illogical. ”
1. No. it is not illogical.
2. it was confusing until I did the math. you have not.
“UHI peaks at 7C in Tokyo and studies reveal a trend in temp and an associated decreased diurnal temperature range that diverges from a rural control in Kim(2002) and Hamdi(2010) and others.”
1. As I explained over an over again.. You will be grossly mislead ( i was at first) if you
concentrate on studies that show huge peaks. THEY ARE DESIGNED TO FIND THE MAX
we are interested in the average.
2. You need to study LLN
3. Kim 2002 studies Max UHI. When you see the term max UHI, put the paper down
it will only confuse you. It confused me for years.
4. Hamdi (201)doesnt help your case. we know that certain cites suffer from UHI.. For that series
It warmed at rate slightly higher than the global average.. But it was an odd ball
so it got DOWNWEIGHTED in the average.. after downweighting.. It warms less than
the global average
http://berkeleyearth.lbl.gov/stations/152708
“But Karl creates a project to disprove this and that becomes the final word. And you say Anthony buys into it?”
1. Please go read the history of CRN
2. please go read the GAO reports
3. please go read the conference notes and presentations leading up to CRN
4. Dont ask me to do your homework
5. Anthony and CRN. Does he buy into it?
you forget that he was paid ( rightly so) to produce this
‘How did those CRN sites look?
REPLY: Unsurprisingly, they looked like CRN1 and 2’s. I was impressed with the diligence they have gone through to ensure siting is not an issue in the present or future. I am more convinced than ever that the CRN will be accurate. I threw about a half dozen what ifs and they had relevant and applicable answers for each issue. It is well thought out.”
“You yourself endeavored to do a study somewhat similar to what I proposed. I have a copy of your draft. I’m not sure why you wanted to do it, or if you ever did it. Seeing your draft is what got me started in January to realize wind and rain nullify UHI and present a mode of segregation to create a control. It was shortly after that I found DTR divergence from the rural control was noted by Kim(2002). Then Hamdi(2010) focusing in on divergence of DTR trend and quantified it in a longer interval as 2.5X to the rural control. I find it hard to accept that all the wind and rainfall records are lost. If we still have them and it turns out we have very little relative warming on windy and rainy days how could that be explained in an AGW context?”
do you not read.
1. In many cases there ARE NO WIND OR RAIN RECORDS TO BEGIN WITH..
nothing is LOST.
2. You cant calculate a LONG TERM TREND from a few days!
3. Selecting for windy and rainy places will give you a skewed sample spatially
4. Picking the rainiest cities and windiest cities you can already see your idea is busted
Here is a clue Ron.
I have way more time and answers than you have questions.
I will only ask you one question.
1. How did you derive a global temperature of .6C from 1850 to today, given your argument
that 1/3 of the land record was “plausibly” infected with UHI?
here is another clue.. You need to choose. You can challenge forcings, or temperature or OHC.
why do you need to choose?
because you cannot be an expert in all three.
Steven, I’m impressed by your patience…
However, my view on UHI is that it’s real and anthropogenic, caused by waste heat from energy use, albedo changes, etc.
AGW means Anthropogenic Global Warming, not exclusively Anthropogenic Greenhouse Warming. UHI is a part of the global warming and should not be “removed”..
However, urban met stations must have a proper area weighting, and only represent urban areas, not the surrounding countryside. As I understand, the BEST algorithm does this, it find the odd balls and shrink their area influence..
Homogenisation algorithms are conservative, they don’t like fast changes and extreme events, adjusting them down or simply removing the data.
UHI corrections like that of GISS downplay global warming to that of a non-urban world.
GHCN v3 dislike the rapidly warming Arctic, and permanently adjust down stations from Svalbard eastward through Alaska by 1-2 C, cooling the global trend since 2000 by about 20%. The recent Winter this wasn’t enough, a persistent hotspot over Svalbard, Franz Josef and Ostrov Vize, with anomalies of +10 C from November through April, was disliked by GHCN v3. It simply removed data from those three stations, thus cooling the world by 0.02 C. The Gistemp loti record from Feb 2016 is no longer 1.35, only 1.33 C
Sorry for this OT, but I dont think that the global temperature datasets exaggerate global warming, on the contrary they are conservative..
Ron,
You keep saying this, but I think you’re wrong. For an ECS of ~1K to be plausible, non-anthropogenic influences would have had to produce more than 0.5C of the observed warming and the planetary energy imbalance in the mid-1800s would have had to be substantial (> 0.5 W/m^2). Neither of these is likely, so even though I didn’t include some element of natural warming in my post is not going to change the basic conclusion.
And that would show up in sea level change and it doesn’t. Ron rather ignored this point earlier.
600 years at 0.41 oC per decade is 0.25 oC. A bit of that is due to orbital decay. In fact McGregor et al consider that land use changes provide as strong a contribution to the cooling as their modeled volcanic contribution. The “reboundable” component of temperature suppression from some (undefined) “equilibrium” value must be small based on all the paleotemperature data (not just McGregor et al).
No that’s incorrect. McGregor et al. present data with bins recentred at shifted 50 year intervals (Supp. Fig S5) and however the bins are centred there isn’t a Medieval Warm Period in their data, whether or not there was one in reality!.
Getting back to the subject of this thread, all of the global temperature series (NASA Giss, Hadcrut, NOAA, Best) shows that we’ve had over 1 oC of warming since the late 19th century (pre-Krakatoa), and we’ve pretty much established using the data you’ve linked to here that any putative “rebound” contribution is small.
In fact, we can get do a useful sanity check on the likelihood of an ECS below 2 oC:
For a climate sensitivity of 2 (oC per doubling of atmospheric [CO2]) we expect an equilibrium warming of:
ln(405/286)*2/ln(2) = 1.0 oC
If atmospheric [CO2] levels were to stay at current levels of 405 ppm.
As pretty much established on this thread, we’ve had at least 1 oC of warming since the late 19th century (pre-Krakatoa) even taking account any putative small negative radiative imbalance in the late 19th century for which there isn’t actually good evidence for. So we’ve had all the warming already expected for an ECS of 2 oC. And we’re far off actually doubling [CO2] levels. And we have a substantial current TOA radiative imbalance that we have yet to equilibrate with. And a significant part of our greenhouse induced warming has been offset by atmospheric aerosols. Offsetting this is the fact that some of the man-made warming arises from methane nitrous oxide and CFC emissions.
An ECS below 2 0C isn’t too likely!
Steve, thanks for the detailed reply. I am assuming Anthony’s paper is not published and you did not read it.
Steven Mosher’s calculation: “The OCEAN ALONE with ZERO UHI has warmed around .9C
the OCEAN is 70% of the total.
The land has warmed about 1.5C
Even IF 30% of that were UHI… you’d STILL have 1C of warming. Now, Assume that ALL OF THE 1.5C land warming is UHI. Calculate the global average..
done?
.63C”
The above includes the following assumptions:
1)The land and sea records are dead on accurate.
2) BEST says Land warmed 1.5 since 1850, AND is more accurate than the more commonly accepted HADCRUT, which you say got it wrong by -0.25C in the 1800s.
3) All of the warming is AGW from GHG
I question all three of these assumptions and replace them with the following:
1) There is high uncertainty in the first 100 years of the land record and first 156 years of the sea record.
2) The natural bias of the investigators is toward warming and there is no adversarial controls established to check self-serving reporting and results. The assumption must be weighted with a suspicion of warming bias. This has been demonstrated in evidence such as MBH(98, 99) and Climategate emails.
3) Judging from the OCEAN2K (2015) analysis there is fluctuation in the amount of natural forcing from volcanic and solar, or perhaps cloudiness, that cools the oceans on a millennial time-frame as well as centennial.
4) Two of the natural forcings, volcanic and solar, have evidence being temporary cooling agents for the middle 600 years of the last millennium to the extent of perhaps 0.7C.
5) If the last 166 years has experience forcing more typical of the 1-1250CE time interval then one could expect a rebound, a warming of perhaps 0.4C on a 166-year interval.
6) If UHI is large in cities, medium in suburbs and small in rural, we may not be correcting for it properly with coding that assumes suburban and rural = 0 UHI
I thus present the calculation presuming:
1.25 C Land warming since 1850 assuming HADRUT dead on. 0.9C Ocean warming per you less 0.2C uncertainty (call if Karl bias) = 0.7C ocean warming
Double the ocean warming and adding to land and divide by 3 equals global warming
((0.7 x 2) +1.25)/ 3 = 0.88C
Subtracting for 0.4C being natural warming rebound from LIA yields:
0.88 – 0.4 = 0.48C
Assuming Anders’ given radiative imbalance numbers, his equation thus becomes:
The equation then becomes Lamda = -(0.7-0.1-2.3)/0.48 = 3.54.0W/m2
EfCS = 3.7/Lamda = 3.7/3.54 = 1K
Therefore 1K is demonstrated plausible even without accounting for UHI bias.
Now for UHI bias.
to be continued….
Ron,
Except you have to do this
and you have provided no actual justification. There is no physical process in our climate called “rebound”. It is not a bouncy ball. So, yes, if half of our observed warming is natural, then the EfCS could be 1K, but there is little evidence that half could be natural and you still have not provided a physical process that could justify this.
Furthermore, you still have the problem of explaining how if non-Planck feedbacks are 0 the response to natural warming can produce a positive energy imbalance. There is an inconsistency that you have not addressed.
Ron Graf writes: “Steve, thanks for the detailed reply. I am assuming Anthony’s paper is not published and you did not read it.”
First, I rather think that was Steve’s point — no published paper has ever resulted from AW’s misbegotten 2012 fiasco. It was followed by the 2015 AGU poster, but that also has severe problems. You were provided links where the 2012 ‘paper’ and the AGUposter and the ideas behind them were discussed at length by one of the co-authors – Evan Jones. Apparently you didn’t even bother to follow them. And here you were all excited by AW’s ideas, yet apparently not excited enough to read more about them in detail. Imagine that.
oneillsinwisconsin says:
May 19, 2016 at 1:22 am
Ron
‘The above includes the following assumptions:
1)The land and sea records are dead on accurate.
2) BEST says Land warmed 1.5 since 1850, AND is more accurate than the more commonly accepted HADCRUT, which you say got it wrong by -0.25C in the 1800s.
3) All of the warming is AGW from GHG
A. You have not explained your .6C calculation
B) You dont need dead on accuracy. YOU need to SHOW not merely question. you need show
the estimate is biased in one direction or the other
C) Hadcrut? commonly accepted? All the appeal to the crowd. We showed ( not speculated)
that there method was biased in the early years. Just as Cowtan and way showed their bias
in recent years
D) This has Nothing to do with the cause of the warming
YOU HAVE TO SHOW how you get to .6C warming with 1/3 UHI
and yes
as predicted
You cant do that without attacking SST
This is silly isn’t it? Can’t we just agree that Ron would like to think that ECS is very low, and he thinks we haven’t warmed as much as the evidence indicates? This sort of argumentation by assertion, special pleading and loaded assumption is particularly uninteresting. If Ron could point us to some published or archived data that supported his argumentation he presumably would have done so by now. Otherwise we get the point already.
Chris,
It would still be interesting to see if Ron can recognise that an ECS ~ 1K means non-Planck feedbacks are 0, or even slightly negative. Hence any natural warming should produce a negative planetary energy imbalance which is inconsistent with the positive planetary energy imbalance we currently have.
Here’s the only way I can see an ECS of ~ 1K being plausible. Let’s start with what we currently “know”
1. Change in external forcing ~ 2.3W/m^2.
2. Overall warming ~ 1K
3. Current planetary energy imbalance ~ 0.7W/m^2.
This tells us that if ECS ~ 1K, that if we started in energy balance, warming of 0.62K would return us to energy balance. Since an ECS of ~ 1K implies that non-Planck feedbacks are close to 0, any warming beyond 0.62K would produce a negative planetary energy imbalance. Given a Planck response of 3.2W/m^2/K, an extra 0.38K of warming would produce a negative planetary energy imbalance of 1.2W/m^2.
Hence if the ECS is 1K and the planetary energy imbalance today is 0.7W/m^2, then that would imply an initial planetary energy imbalance of 1.9W/m^2. That seems rather implausible.
Next possibility is that our assessment of the overall surface warming is horribly wrong. Let’s imagine now that we started with a planetary energy imbalance of ~0.1W/m^2 and now have a planetary energy imbalance of 0.7W/m^2 after a change in external forcing of 2.3W/m^2 and some as yet unknown surface warming. Since the Planck response is 3.2W/m^2/K, what we need is an amount of warming that produces a radiative response of 2.3 – (0.7 – 0.1) = 1.7W/m^2. That would be 1.7/3.2 = 0.53K.
So, the two possibilities that I can see is that
1. The initial planetary energy imbalance was 1.9W/m^2, almost 3 times bigger than it is now.
2. We’ve actually only warmed by 0.53K, almost half what almost all estimates suggest.
3. I’ve made a silly mistake.
I wouldn’t rule out 3, but – if not – 1 and 2 seem highly unlikely.
I guess there is another possibility.
Estimates for the change in external forcing are very wrong. For ECS to be ~ 1K, for us to have warmed by 1K, and for us to still have a planetary energy imbalance of 0.7W/m^2 would imply
4. The change in external forcing is actually around 3.9W/m^2, almost twice the current best estimate.
The only method of producing iron core evidence that DTR is affected by AGW or natural long-term trends not associated with UHI – MS would be an Argo study, which we could do now with the 11 years of data. Or, we could do the rainy/windy vs clear/calm analysis I proposed comparing DTR with meteorological data of 50+-year continuous record stations. But in the absence of such elimination of DTR trend affect from non-UHI we can leave the assumption as still plausible that All of the DTR trend is UHI related, thus:
According to Vose, Easterling et al (2005) a 0.7C/ century decrease trend in diurnal temperature range DTR occurred from 1950-2004. Half of the DTR is reflected in Tavg, which is the metric used to calculated anomalies for month and year.
0.7C/2 = 0.35C
Vose and Easterling are buddies with Karl and Hansen and have no anti-AGW personal bias, quite the opposite.
If 0.35C were subtracted from the Hadley land record HADCRUT then we have 1.25-0.35 = 0.9C
We all agree that land is more responsive to atmospheric change due to greatly less heat capacity. Thus a warming of 0.9C on land would imply a 0.65C warming of the ocean, which we know has a much less certain record pre 2005, thus we would give the land record more weight of confidence.
((0.65 x 2) + 0.9)/3 = 0.73C
Now lets assume that only 0.13C of that gain is from “return to normal” volcanic and solar activity.
0.73C – 0.13C = 0.6C
My comment one week ago was: “And I would place the beyond a doubt consensus at > 0.6C.”
Steve, if you are referring, with your incessant “show me” taunting, to a comment I made at Lucia’s, I reasoned there UHI could be as high as 0.35C of the land record as I showed above from Vose. Then I had a momentary lapse of sanity and reasoned that 0.35 is about 1/3 of 1C, and that 1C was the claim of 160 years of global warming.
[Shirt ripping. -W]
The points I made are valid regarding natural centennial level variability and uncertainties with our climate records.
BTW, Steven, where is the best place to get historical meteorological records if a layperson was going to set up a study of them for DTR?
> The points I made are valid […]
The possibility point only, RonG – establishing possibility is more or less trivial, and a trivial point is trivially valid.
The plausibility point has yet to be backed up, after many, many, many, many requests.
***
> where is the best place […]
You don’t get to ask anything until you answer the questions asked, RonG.
Ron Graf writes: “…we can leave the assumption as still plausible that All of the DTR trend is UHI related”
Except that paper you cite Maximum and minimum temperature trends for the globe: An update through 2004, Vose et al, GEOPHYSICAL RESEARCH LETTERS, VOL. 32, L23822, doi:10.1029/2005GL024379, 2005, says:
So are you *also* positing that this UHI effect suddenly stopped for 25 years (1979-2004)? Hmm, I don’t suppose you happen to have a physical theory to tell us how the UHI/DTR effect decides when and when not to act?
BTW, Ron, what do you suspect the DTR is on our neighbor Venus? Compare and contrast with the DTR on our other neighbor, Mars. Does the differences tell us anything?
> what do you suspect the DTR is on our neighbor Venus?
No need to burden RonG with rhetorical questions. He has enough to answer already.
Ron:
“The only method of producing iron core evidence that DTR is affected by AGW or natural long-term trends not associated with UHI – MS would be an Argo study, which we could do now with the 11 years of data. ”
1. Err No. You have merely asserted that that there is no other way. To prove there is no other
way, you’d have to canvas all known and untried methods.
2. An argo study tells you nothing about the effect GW has on DTR and nothing about what
AGW does to DTR.
3. there are at least three ways DTR can be impacted: UHI ( raising tmin), changing cloud cover
(effecting Tmin and Tmax) and AGW. The second, changing cloud cover, is an interesting
to look at. In fact one of the studies of the impact of GCR argues thusly:
DTR changed, that implies a change in cloud cover, that implies a GCR effect
See the references here https://arxiv.org/ftp/arxiv/papers/1304/1304.7879.pdf
yes Ron, read the papers in the bibliography. Here is the point: DTR can changes for
reasons other than UHI. Whether this supports the GCR hypothesis I am skeptical.
“Or, we could do the rainy/windy vs clear/calm analysis I proposed comparing DTR with meteorological data of 50+-year continuous record stations. But in the absence of such elimination of DTR trend affect from non-UHI we can leave the assumption as still plausible that All of the DTR trend is UHI related, thus:”
1. You cannot even begin to explain what you mean by rainy day/ windy day analysis
2. Looking at the rainiest windiest places, your theory is already busted.
3. You claim “plausibility” but its more plausiable that all the DTR is GW related.
You see how weak “arguing from,plausibility is?”
“According to Vose, Easterling et al (2005) a 0.7C/ century decrease trend in diurnal temperature range DTR occurred from 1950-2004. Half of the DTR is reflected in Tavg, which is the metric used to calculated anomalies for month and year.”
1: ” Changes
in cloud cover, precipitation, soil moisture, and
atmospheric circulation likely accounted for much of the
trend differential during the period (e.g., Dai et al., 1999;
Przybylak, 2000; Braganza et al., 2004). Changes in
land use also impacted the DTR in some areas (e.g.,
Balling et al., 1998; Bonan, 1999; Small et al., 2001).
Unfortunately, data constraints have historically limited
global-scale analyses of DTR trends and their causes;
in particular, the most recent global assessment
(Easterling et al., 1997) only covered about half of the
land surface and ended in 1993. Consequently, in this
study we use new data acquisitions to expand spatial
coverage and update global trends in maximum
temperature, minimum temperature, and the DTR
through the period 1950-2004. For consistency with the
IPCC Fourth Assessment Report, trends during the
satellite era (1979-2004) are also discussed. ”
2. “Both maximum and minimum temperature
increases from 1979-2004 whereas the DTR is basically
trendless. The maximum and minimum temperature
trends are nearly identical (0.287 versus 0.295°C dec-1),”
3. Note that the actual science attributes the changes in DTR to other causes, so its
plausible you are wrong.
4. Your theory would require UHI to STOP between 1979 – 2004 a period of
RAPID urbanization.
#################################################
next you assert a change in DTR of .7C. the trend in DTR is -.066 per decade
1.”and in DTR is -0.066°C dec-1.
2. You extrapolate that backwards. Big assumption
3. .066 per decade would be 1.056 over 16 decade
4. That .53C
Your calculation is wrong
0.7C/2 = 0.35C
“If 0.35C were subtracted from the Hadley land record HADCRUT then we have 1.25-0.35 = 0.9C”
1. You made a math mistake
2. 1.25 -53 = .72
Now all your estimates come from a fundamental error in extrapolating a linear regresssion
if you want to see how DTR changes from 1850 to today
http://berkeleyearth.lbl.gov/regions/global-land. A linear regression shows it is “trendless”
If you study it in detail you will find periods where the DTR Widens.. OMG negative UHI?
and where it narrows. and then widens again.
It actually contradicts your hypothesis.
PSST ….. here is one paper we worked on ( haha I was waiting for you to cite Vose)
http://onlinelibrary.wiley.com/doi/10.1002/2015JD024584/abstract
and the follow up
http://onlinelibrary.wiley.com/doi/10.1002/2015JD024583/abstract
[Extraneous copy-paste. -W]
1. your land warming, as I showed, was calculated wrong EVEN IF you extrapolate the DTR trend
2. you can’t extrapolate the DTR trend […] because the history before Vose 1950 shows the opposite.
3. You pulled .65 out of [thin air -W].
4. You randomly assign more confidence to records after the fact. […]
Wrong. […]
land (your […] number) .9 *.3 = .33
SST .65 * .7 = .45
Total = .78 NOT .73
Further, [u]sing your approach land would have been .72 NOT .9 (.7C per decade)
So you will have to do more pulling numbers out of [nothingness -W].
1. HUH? you specifically claimed that the reduction from 1C to .65C was BECAUSE OF UHI
2. Now that your math is busted you make up .13 for solar?? there is no solar change
3. At some point you will admit you subtracted 1/3 from 1C and forgot that land was only 30% of the total.
1. No they are not.
1. See the papers above which totally destroy your position.
2. Thank you for playing.
[Etc. Try to use the spacebar instead of tabs, and please use paragraphs. As Bender would say, “needles in the eyes.” -W]
Of course, Ron could have learned about the dangers of extrapolating Vose 2005 IFF
he had read AR5
Click to access WG1AR5_Chapter02_FINAL.pdf
Note: Ron, Peter Thorne is the author of that chapter. brilliant guy. so when he asked us to
participate in his first paper ( cited above) of course we said yes.
“In AR4 diurnal temperature range (DTR) was found, globally, to have
narrowed since 1950, with minimum daily temperatures increasing
faster than maximum daily temperatures. However, significant multi-decadal
variability was highlighted including a recent period from
1997 to 2004 of no change, as both maximum and minimum temperatures
rose at similar rates. The Technical Summary of AR4 highlighted
changes in DTR and their causes as a key uncertainty. Since AR4,
uncertainties in DTR and its physical interpretation have become even
more apparent.
No dedicated global analysis of DTR has been undertaken subsequent
to Vose et al. (2005a), although global behaviour has been discussed
in two broader ranging analyses. Rohde et al. (2012) and Wild et al.
(2007) note an apparent reversal since the mid-1980s; with DTR subsequently
increasing. This decline and subsequent increase in DTR over
global land surfaces is qualitatively consistent with the dimming and
subsequent brightening noted in Section 2.3.3.1. Donat et al. (2013c)
using HadEX2 (Section 2.6) find significant decreasing DTR trends
in more than half of the land areas assessed but less than 10% of
land with significant increases since 1951. Available trend estimates
(–0.04 ± 0.01°C per decade over 1950–2011 (Rohde et al., 2013b)
and –0.066°C per decade over 1950–2004 (Vose et al., 2005a)) are
much smaller than global mean LSAT average temperature trends
over 1951–2012 (Table 2.4). It therefore logically follows that globally
averaged maximum and minimum temperatures over land have both
increased by in excess of 0.1°C per decade since 1950.
Regionally, Makowski et al. (2008) found that DTR behaviour in Europe
over 1950 to 2005 changed from a decrease to an increase in the
1970s in Western Europe and in the 1980s in Eastern Europe. Sen Roy
and Balling (2005) found significant increases in both maximum and
minimum temperatures for India, but little change in DTR over 1931–
2002. Christy et al. (2009) reported that for East Africa there has been
no pause in the narrowing of DTR in recent decades. Zhou and Ren
(2011) reported a significant decrease in DTR over mainland China of
–0.15°C per decade during 1961–2008.
Various investigators (e.g., Christy et al. (2009), Pielke and Matsui
(2005), Zhou and Ren (2011)) have raised doubts about the physical
interpretation of minimum temperature trends, hypothesizing that
microclimate and local atmospheric composition impacts are more
apparent because the dynamical mixing at night is much reduced.
Parker (2006) investigated this issue arguing that if data were affected
in this way, then a trend difference would be expected between calm
and windy nights. However, he found no such minimum temperature
differences on a global average basis. Using more complex boundary
layer modelling techniques, Steeneveld et al. (2011) and McNider et al.
(2012) showed much lower sensitivity to windspeed variations than
posited by Pielke and Matsui but both concluded that boundary layer
understanding was key to understanding the minimum temperature
changes. Data analysis and long-term side-by-side instrumentation
field studies show that real non-climatic data artefacts certainly affect
maximum and minimum differently in the raw records for both recent
(Fall et al., 2011; Williams et al., 2012) and older (Bohm et al., 2010;
Brunet et al., 2011) records. Hence there could be issues over interpretation
of apparent DTR trends and variability in many regions (Christy
et al., 2006, 2009; Fall et al., 2011; Zhou and Ren, 2011; Williams et
al., 2012), particularly when accompanied by regional-scale land-use/
land-cover (LULC) changes (Christy et al., 2006).
In summary, confidence is medium in reported decreases in observed
global DTR, noted as a key uncertainty in AR4. Several recent analyses
of the raw data on which many previous analyses were based point to
the potential for biases that differently affect maximum and minimum
average temperatures. However, apparent changes in DTR are much
smaller than reported changes in average temperatures and therefore
it is virtually certain that maximum and minimum temperatures have
increased since 1950.
########################################
Even a layman can read this and understand that building a case for 1C ECS
on the uncertain and demonstrably WRONG approach of extrapolating from
one study… is.. well.. implausible.
It’s much easier if you just answer straight forward questions..
or admit you goofed.
That said.. you can still make good arguments for ECS being less than 3C
you cant make good cases for it being lower than 1.5C or so.
If you COULD make a good argument peabody Coal would have called you
as a witness. They didnt. They called Lindzen. he failed too.
Steven, I submit that we all make goof’s as I did in underextrapolating Vose’s trend for 10 decades rather than 16.5. That does not make my assertion wrong that 1C EfCS is plausible, and Anders’ analysis is remiss not to consider the possibilities of bias in the temperature record as well as non-anthropenic warming.
I was citing Vose’s trend from its citing in Parker (2010). Now that you and Oneills have pointed out the inconsistency in that study, where the latter half of the 55-yr interval showed zero trend, I have studied the paper and pondered possible explanations.
1) It is likely most that after 1979 stations were being mostly protected from UHI and to the degree they were not was being cancelled by the introduction of data from stations that were free from UHI and the retirement of stations that were not.
2) Vose used Menne and another statistical evaluation looking for anomalous behavior to be cleaned from the data. If one out of six neighbors experienced the introduction of micro-site issue its data might be dropped from the evaluation.
3) Most of the 0.066C/decade trend occurred in the first decade. Thus, the trend from 1950-1960 was ~0.3C/dec, 3C/century, 5C from 1850-2015. Considering we only had 1.5C land warming at most, we know that did not happen but it demonstrates how sensitive the record could be to this systematic contamination.
Vose was not looking for UHI. Apparently, Easterling just wanted to update his earlier work. They make little attempt for attribution which seems astounding since DTR was known in Kim(2002) and other studies by 2004 to be related to UHI. The closest Vose comes to attribution is mentioned land use along with cloudiness as possible factors.
I urge all to look at the scatter charts of TMax TMin trends in Hamdi (2011). There is clearly and effect.
BTW, I read the Bishop Hill blog and noticed Anders had commented on the same logic as the above post and many bloggers immediately responded that he forgot to include natural warming from the recovery from LIA. Then today I saw Clive Best had a guest expert, Ken Gregory, post on 1C ECS, and it includes details on natural warming and UHI contamination.
Steven, you are the kung fu master. I just grasshopper.
Ron,
There is no such thing as a recovery from the LIA. Our climate is not some kind of bouncy ball that we squeeze and it then recovers when you stop squeezing. It responds to changes in forcing. If you force it one way, it warms. If you force it the other, it cools. If it cools, it won’t immediately warm, unless there is some change that causes it to warm.
I’ve done most of that here. However, the onus is really on you to show that these bias are not just plausible, but actually possible. It’s, of course, good to consider if our estimates might be wrong, but it’s not good to assume that they probably are unless you have actual evidence to support that; evidence that is more than simply UHI hand-waves.
Anders, I agree the GMST is not just a bouncy ball; the trend is chaotic as much as reciprocal. What we mean by “rebound” we are talking about recovering to an imaginary baseline of assumed similar forcing. There is real evidence by non-skeptics that there is centennial and millennial variability and that 1250-1850 was well below the statistical baseline. There is also real evidence of increased volcanic activity from sediment studies. Just after Galileo discovered sunspots they became absent from observation for almost a century. The inflection of solar inactivity is known as the Maunder Minimum. So, the seas were cool, making 1850 GMST a historically low baseline.
If we get some adversarial rigor into the science we will make everyone more productive. Competition = progress. I noticed the Christy has the idea of throwing Tavg out and going with Tmax only to easily decontaminate from UHI. What would be wrong with that?
“If we get some adversarial rigor into the science we will make everyone more productive.”
I think this has finally slipped into the realm of conspiracy. More hand waving, more side stepping, more speculating more ifs, no physics.
Ron Graf writes
Regarding recovery: Many – including our host – have explained ad nauseum that there’s no ‘recovery’ from the LIA. The forcings change and the GMST changes as a result. Since we are in an interglacial heading south, there’s absolutely no reason to expect temperatures to suddenly start increasing. This is why more than 100% of the change is often attributable to anthro.
oneills: ” Since we are in an interglacial heading south, there’s absolutely no reason to expect temperatures to suddenly start increasing.”
When you gave a best guess of 1.5C millennial natural variability range you meant forced, right? Well forcings must change naturally. As Eli would say, everybunny knows, ocean currents (polar transit of heat), solar, and volcanic are all chaotic. Orbital is the only gradual player, big on the multi-millennial scale, but small in any single millennium.
Ron you had 3 numbered points. You gave no evidence for the first two and you had the wrong sign on the trend for the 3rd. What does any of that have to do with 1.5C variability over the last 1ky or 2ky? Absolutely nothing. Look squirrel!
Besides, anthro forcing isn’t ‘natural’ and we’ve seen over 1C increase in the past 150 years due to that alone. I never said the variability was due to Milankovitch forcing – I just stated what I thought the reasonable range of Max-Min were over those periods. You are dependably wrong.
> the onus is really on you to show that these bias are not just plausible, but actually possible.
Or the other way around.
Ron,
And we have estimates for the changes in non-anthropogenic forcings. They’re small.
If you’re referring to what was presented on Clive Best’s blog, then it is simply curve fitting. What you don’t have is some mechanism that can provide internally-driven warming on these timescales. There are very good arguments that internally-driven warming operating on these timescales is unlikely.
Willard,
Yes, or maybe “not only possible, but actually probable”.
Anders, Oneills, you keep telling me what you don’t think causes natural variability.
I asked Oneills: “:When you gave a best guess of 1.5C millennial natural variability range you meant forced, right?” Oneills answered:, “What does any of that have to do with 1.5C variability over the last 1ky or 2ky? Absolutely nothing. Look squirrel!” So you are asking what does volcanic, solar, ocean current driven thermal gradients and M-cycle have to do with natural variability? I will add a couple more: asteroid strikes and cosmic ray variability effect on cloud formation. There could well be others. The point is that although the proxy studies often contradict on certain points, there is a good consensus of being natural millennial variability. Oneills put that range at 1.5C. I would say that is a good average estimate.
Anders, there is little evidence I am aware of that natural changes must be gradual. Volcanoes and asteroids are abrupt. Solar or ocean current effects are not constrained to gradual. Didn’t you see the movie “The Day After Tomorrow?” 🙂
Ron,
Here is the key point. If ECS is ~ 1K, then non-Planck feedbacks are essentially zero. If there is internally-driven warming of magnitude dT, then we would produce a negative planetary energy balance of 3.2W/m^2/K x dT. If cooling, then the reverse would apply.
Now the land/atmosphere has a mass of around 10^{19}kg and a specific heat capacity of 1000 J/kg/K, hence a total heat capacity of 10^{22}J/K.
Now, lets put some numbers in.
3.2 J/s/m^2/K x 4 x 3.14159 x 6370000^2 = 1.6 x 10^{15} J/s/K
Now imagined we have internally driven warming of 1K. We’d have increased the total energy in the land/atmosphere by 10^{22}J and be losing 1.6 x 10^{15} J/s. We’d lose all the extra energy in 6.25 x 10^{7} s. That’s two years. Can you at least acknowledge that you understand this?
That’s the key point. If you want to argue for an ECS of ~ 1K, then essentially you’re arguing that our climate is very insensitive to all changes.
Anything but CO2
I would argue that climate is more sensitive to abrupt changes in the contrary direction of the ambient glacial state. For example. the AMOC exciting its circulation from terminating in Labrador to the Norwegian Sea could have caused a very abrupt melting of ice sheets and glaciers to end the ice age. On the other side the asteroid that created Meteor Crater 50,000 years ago may have nipped a budding M-cycle interglacial. Sensitivity may lie more in the boundary area that causes the northern hemisphere to remain snow covered into its spring or not.
The paleo record does establish that CO2 lagged all the major climate changes in the last 500ka. When Hansen gave his famous warning in 1980 he thought the evidence pointed to the opposite, as does Al Gore to the present, but that’s another story. Do you dispute that CO2 concentration change lagged paleo-climate change?
” could have caused a very abrupt melting of ice sheets ”
“may have nipped a budding M-cycle interglacial”
It may also have been precipitated by the final defeat of the Jotuns by the Aesir.
Or maybe not. Without evidence who can say.
Physically, CO2 must lag natural climate change at least some. The greatest temperature-dependent store of CO2 is in the ocean, and warmer oceans hold less CO2. So as the climate warms, CO2 is released, and the increasing CO2 causes the climate to warm more. There’s a feedback loop between the two, but it’s kicked off by external forcings like orbital changes that cause the atmosphere and oceans to warm first.
That’s the natural cycle. Temperature changes first, and then CO2 is part of a feedback that creates more warming, which releases more CO2, etc. The timescale of the CO2 feedback is determined by how quickly the oceans warm and outgas.
Of course, that natural cycle can be broken if you have some sort of phenomena that produces CO2 first. Like, say, people dig up and burn sequestered carbon. The part of the cycle in which CO2 produces warming still works just fine.
[But RC & Al Gore. -W]
My view on paleo-CO2 lag of temperature is the striking lack of any apparent effect to inhibit drastic fall or climbs in temperature. If the orbital cycles are weak and gradual one would expect they would have a tough time overpowering CO2. If the cycles are strong and gradual one would expect smoother correlations with temperature to them. Have any thoughts on this question?
Yes – you are overplaying your hand. See eg. Leads and Lags at the End of the Last Ice Age Edward J. Brook (2013; Science) for a discussion of Parrenin et al. (2013) and Pedro et al. (2013).
Not sure if comments are allowed on older posts like this. But I think I have a comment that might be relevant here.
Bates recently wrote a shoddy report for the misinformation organization GWPF:
“Deficiencies in the IPCC’s Special Report on 1.5 Degrees”
https://www.thegwpf.org/content/uploads/2018/12/Ray-Bates2018.pdf?utm_source=CCNet+Newsletter&utm_campaign=355a04e07c-EMAIL_CAMPAIGN_2018_12_20_10_02_COPY_01&utm_medium=email&utm_term=0_fe4b2f45ef-355a04e07c-20158893
Judith Curry is approvingly citing the report, and I’ve made some criticisms in response:
So it looks like Bates is continuing in his campaign of distorting the science on anthropogenic climate change, while Curry and GWPF promote his distortions.
Atomsk,
Thanks, I hadn’t realised that he’d written a report for the GWPF. They’re really struggling now.
Pingback: The exoplanets around GJ 9827 | …and Then There's Physics