RealClimate logo

Unforced Variations: Oct 2013

Filed under: — group @ 1 October 2013

This month’s open thread. We’re going to guess that most of what people want to talk about is related to the IPCC WG1 AR5 report… Have at it!

286 Responses to “Unforced Variations: Oct 2013”

  1. 1

    I have just published a review what the IPCC WG1 report writes about my area of expertise (homogenization of station data). I personally expect uncertainties in the global mean temperature to be larger, but would say that the IPCC gives an honest summary of the scientific literature. That is what the IPCC is supposed to do.

    Such reviews are a good way to judge the quality of an information source. Please help me by finding or writing further reviews.

  2. 2
    P. Lubitz says:

    Lost in the media coverage of the IPCC report and its definitive confirmation of the evidence of global warming, and its almost certain cause as human activity, was any mention of the dramatic new data on temperature reconstructions for the past 10K years, and of the ocean warming, especially as measured down to 6000 meters as discussed in the previous two RealClimate posts. These offer stark confirmation that the moderation of atmospheric global warming in the last decade is ephemeral and mainly the result of increased mixing of the increasing heat input into the deeper ocean. These results should be promulgated to the media with all means and urgency available! Also to be remarked, again as gleaned from earlier RC posts, is the slight cooling effect of the present solar minimum, and the more significant effects of increasing emissions, mainly of sulfates, in growing economies, mainly but not only, China. I l

  3. 3
    prokaryotes says:

    Victor is the a WUWT fail compilation? Thanks.

  4. 4
    prokaryotes says:

    the = “there”

    Btw what kind of climate sensitivity is used in the RCP scenarios? On the bottom line, it should be more clear that we heading to a early Pliocene state (+20 m SLR). Re +2 C is a disaster target.

  5. 5
    Lauri says:

    The new report illustrates how crucial the ocean – atmosphere interaction is for the atmospheric temperature development. Are there any analyses about the uncertainty range that the heat exchange between these two creates? Is that included in the spread of temperature constructions in the model runs? If not, is there any idea about the possible rates of acceleration or deceleration due to changes in ocean uptake?

  6. 6
    Ray Ladbury says:

    Gavin et al., In all of the copious spare time you will undoubtedly have during the shutdown, might you consider an update on developments in climate sensitivity?

    In looking at new developments in the latest IPCC report, one notable change is the decision not to give a single best estimate. The denialosphere has gone ape caca over this. However, there is actually something interesting here–that is that the distribution of sensitivity estimates has bifurcated in the last 7-8 years. During this period, you have several estimates that find relatively low sensitivities (1.5-2.3) and several estimates with sensitivity in the 3-4.5 range and not much in between. What’s going on?

    It occurs to me that they could be measuring two different sensitivities–one with rapid equilibrium time yielding low sensitivity (e.g. Schwarz 2007) and another with added slower feedbacks yielding a much higher sensitivity. The paper:
    “On the diagnosis of climate sensitivity using observations of fluctuations” – Kirk-Davidoff (2009)

    seems to support this, as do the studies showing more significant warming in the deeper oceans.

  7. 7
  8. 8

    prokaryotes, as far as I know, a list with WUWT fails does not exist yet. It would be nice if someone would make a list. Or maybe someone can make a pre-selection and then everyone can vote for his favourite WUWT classics.

    My contributions to that list would be:

    A view of climate “on the ground” from a reporter who was there at the beginning

    On the plus side, there’s no reason for William M. Connolley to comment here anymore

    Careful post selection at WUWT

    I only have a few posts on misinformation by WUWT. The experts are: Wotts Up With That Blog, Hotwhopper, Wott’s Up With That?, What’s Up With That Watts?, VVatts Up With That. They regularly explain WUWT’s errors.

  9. 9
    Joseph O'Sullivan says:

    According to the IPCC homepage, the section of WGII that addresses ecosystems effects will be released in March. Does anyone know of any recent papers that deal with ecosystems and anthropogenic global warming (AGW)? A review paper would be ideal.

    On the mitigation front, there were nine legal challenges to the EPA’s current plan to reduce AGW pollution. All were rejected, so the small amounts of progress will continue.

  10. 10
    Nick O. says:

    #6 Ray

    Could you say a bit more about this, please, as it strikes me as being a particularly important point. For example, what difference is being used to distinguish between a “rapid” equilibrium time and a presumably longer time (but higher sensitivity).

  11. 11
    Ray Ladbury says:

    Nick O.,
    We’ve known for a long time that a rapid equilibrium time tended to imply a low sensitivity–that is the mistake Schwarz made. The thing is that it all depends on how long one must wait for the system to approach closely enough to equilibrium. If it’s rapid, then whatever temperature rise we’ve had in that period determines the sensitivity. For a longer equilibration time, we heat the air and shallow ocean, then the deep ocean, then the air/shallows some more and so on. Slower feedbacks come into play. This latter scenario seems to be what is going on wrt the Skepticalscience escalator. Anyway, this is the way I understand things.

    It also seems to be supported by the study in Kirk-Davidoff. And if this is the case, then depending on what data you use, you will get a sensitivity in either of the two modes.

  12. 12
    FP says:

    There have been a lot of fires. Australia, California, Denver, etc… I would be surprised if they didn’t effect the climate and weather. Perhaps causing some cooling? The soot certainly transfers heat into ice rather than bouncing it back up into the air. Is fire smoke different than volcano smoke?

  13. 13
  14. 14
    David B. Benson says:

    Source Of 13th Century Volcanic Calamity Discovered
    Can you find this in the GRIP ice core data?

  15. 15
    prokaryotes says:

    Thanks Victor for the overview, but please don’t hyperlink to WUWT, part of this is the reason why WUWT is listed on many climate related search queries at Google. Use a link shortener instead. The misjudgement and the debunked claims from WUWT should be used to derank WUWT at Google.

    What is missing is a website which explains current climate science – on a daily basis – accurately, written for the average Joe.

  16. 16
  17. 17
    Gordon Kenney says:

    AR5 is an amazing collaborative achievement. The report was released on Friday, (when most news wishing to be buried appears). During press coverage, the IPCC consensus was compared to the views of the usual suspects, without voice to those who believe that the situation may be a little more dire than RCP 8.5.

    A general lack of discussion in the media about the unknown implications of albedo change, future ice sheet dynamics, and other feedbacks is understandable given that these concepts are not well understood; the IPCC ignores chaotic factors in its assessment.

    In translation to a social context this scientific rigor unfortunately downplays the severity of our current situation, and promotes inaction by the public. ie “A RCP 2.6 is possible too and didn’t you say the temperature is not rising very quickly”? An average citizen does not understand the social transformation required to attain RCP 2.6, nor the consequences of a 4 degree C change by 2100. The vested interests surely wish to obscure the truth.

    My question is; how can the findings of climate science be effectively communicated more frequently than the IPCC in a language that can transcend propaganda?

    Sincere thanks to those who seek truth!

  18. 18
    Nick O. says:

    #11 Ray L.,
    Many thanks for this, it makes sense. I’ve been concerned for a long time that the sensitivity is higher than supposed, and likewise that the mixing of the extra retained heat within the Earth system will be very complex and take much longer than supposed. Anyway, I’ll try to have a look at the Kirk-Davidoff paper at some point, and get this clearer in my mind.

  19. 19
    Juerg says:

    @1 Victor, thanks very much for the nice review!
    Please note, IPCC did not a review but an assessment!

  20. 20
    prokaryotes says:

    “This provides evidence that some CMIP5 models have a higher sensitivity to greenhouse gases and a larger response to other anthropogenic forcings (dominated by the effects of aerosols) than the real world (medium confidence).” The ASK results and the initialised predictions both suggest that those CMIP5 models that warm most rapidly over the period (1986– 2005)–(2016–2035) may be inconsistent with the observations. (page 123) Link

    The IPCC report is not making it very clear that the lack of projected ground temperature records is related to the current IPO state, which distributes heat energy to the deep sea rather than atmosphere. Do these models of the past, account for that IPO related heat distribution?


    Another approach to making projections involves weighting models according to some measure of their quality (see Chapter 9). A specific approach of this type, known as ASK (Allen et al., 2000; Stott and Kettleborough, 2002) is based on the use of results from detection and attribution studies (Chapter 10), in which the fit between observations and model simulations of the past is used to scale projections of the future. ASK requires specific simulations to be carried out with individual forcings (e.g., anthropogenic greenhouse gas forcing alone), and only some of the centres participating in CMIP5 have carried out the necessary integrations. Biases in ASK derived projections may arise from errors in the specified forcings, or in the simulated patterns of response, and/or from non-linearities in the responses to forcings

  21. 21
    Urs Neu says:

    There are a number of possible reasons that are discussed, and probably it is (as usual) a combination of some (or all) of them. A short summary with some comments, (the ‘slow-down’ being a reduction of the global temperature trend from about 0.17°C per decade 1976-2000 to about 0.05°C per decade 1998-2012, i.e. by about 0.12°C per decade):

    1. ENSO: The negative trend of ENSO over the last 15 years can explain – depending on the analysis – from only a small contribution (Fyfe et al. 2013) over about two thirds (Foster and Rahmstorf 2011) to almost all (Kosaka and Xie 2013).
    2. Volcanic Aerosols: The increase in volcanic aerosols over the last 15 years can explain from a very small contribution (Foster and Rahmstorf 2011) to about a third of the slow-down (Solomon et al. 2011)
    3. Solar activity: The negative trend of solar irradiation over the last 15 years (due to the current low maximum of solar cycle 24, showing about half the activity of the maximum in cycle 23) might explain from only a small contribution (based on the direct forcing value) up to more than half of the slow-down (based on several regression analysis)
    4. Heat uptake by the ocean: Some studies explain the slow-down by heat uptake in the deep(er) ocean. This explanation involves important uncertainties and relatively short time series in deep ocean measurements. It is important to note that this explanation might overlap quite strongly with the influence of ENSO, since the ENSO effect mainly works through heat exchange with the ocean.
    5. Overestimation of climate sensitivity by (some) climate models: there is some evidence that at least some climate models might overestimate climate sensitivity. However, this can hardly explain the slow-down (i.e. a strong change in the trend), since climate sensitivity of the models has not changed in the year 2000. The argument is the other way round: the slow-down might point to a somewhat slower climate sensitivity, because it lowers the long-term trend over 50 years or so.
    In summary, external (solar volcanic) and internal variability (ENSO) might well explain the slow-down, since the latter is by far in the combined range of explanations 1-3.

    An interesting point when trying to investigate these factors is the fact that the slowdown is more or less restricted to the Northern hemisphere winter (see e.g. Therefore one should discuss these factors in the light of a possible mechanism leading to an influence mainly in winter time on the Northern hemisphere. Let’s look at these influences:
    – ENSO: Kosaka and Xie have a corresponding explanation, arguing that the meridional heat exchange out of the tropics is much larger in winter, and therefore the ENSO tropical warming/cooling affects the higher latitudes much stronger in winter. Something to consider might be that El Niño/La Niña events normally are most pronounced in winter/spring, but regression analysis shows that the corresponding effect on global temperature has a time lag of a few months and therefore might be most pronounced in summer/autumn…
    – Volcanic aerosols / solar activity: Both factors influence solar irradiation, which has its strongest effects in Northern hemisphere summer, therefore this is inconsistent with a main effect in winter.
    – Heat uptake by the deep ocean: at first sight I can’t see a reason, why this should have a distinct effect in winter. While mixing and heat uptake in the upper layers have seasonal characteristics, the mixing down to deep layers might less depend on seasons. Any thoughts by oceanographers?
    – Climate sensitivity: I don’t know an effect explaining a distinct winter effect.

    Another effect not in the list above has been discussed by colleagues during the last days:

    6. Energy put into the melting of ice: The strong increase of ice sheet melting over the last 10 years has absorbed quite a lot of energy. This seems to be not a negligible amount, at least when compared to the energy taken up by the atmosphere. Does somebody know a recent calculation of the energy budget including recent ice sheet melt increase? However, it is not clear either, why ice (sheet) melt, which absorbs energy mainly in summer, should lead to the slow-down in winter… On the other hand, this effect is the only one, that is not due to variability and therefore will not disappear after some years. Thus it should be considered carefully…

    Any thoughts?

  22. 22
    Ray Ladbury says:

    BTW, the two modes are centered at roughly 2.16 and 3.64 degrees per doubling–even the lower mode is not cause for celebration.

  23. 23

    prokaryotes, you criticise the wrong person. ;-)

    I have just written a long blog post about linking to WUWT and Co: NoFollow: Do not give WUWT & Co. unintentional link love.

    In short: you can link to WUWT in comment below blogs. The software typically automatically adds a tag NoFollow, which tells Google not to count this link as a recommendation. URL shorterners do not always have this effect. In Forums you sometimes have to add the NoFollow tag yourself. In a blog post you always have to add it yourself. There are apps to help you see which links as NoFollow and which are DoFollow.

    A blog with daily climate science would need funding. WUWT can report daily because they do not have any quality standards, but a pro-science blog would have to have standards. ThinkProgress Climate reports daily about climatic issues, both science and politics.

    Juerg, I am just a scientist and also not a native speaker. What is the difference between a review and an assessment and a synthesis?

  24. 24
    Dan H. says:


    You may want to add heat loss due to open Arctic waters. The decrease in sea ice area (or extent) was only about 7% between 1979 and 2001, but a whopping 30% since. At the end of summer, the warm open waters mixing with rapidly cooling Arctic air creates the potential for large evaporation, and subsequent cooling. Both GISS and CRU show temperature decreases starting around 2002, immediately following the higher (relative) sea ice minimum of 2001. Prior to then, changes in heat loss in the Arctic would have been relatively minor. See the following:

  25. 25
    prokaryotes says:

    Victor, that is interesting and thanks for pointing latest linkage features out. CP is a good blog but the scope is somewhat to specific around Washington politics and Energy. There are many different concepts but most can be considered to be publications from academic sources written for this audience. What i have in mind is a bit like HuffPost or The Guardian, but both sources only cover climate within a broader news coverage.

  26. 26
    Ray Ladbury says:

    Preaching to the choir. I haven’t gone to Tony “Micro” Watts blog in years except by accident.

  27. 27

    “ENSO: The negative trend of ENSO over the last 15 years can explain – depending on the analysis – from only a small contribution (Fyfe et al. 2013) over about two thirds (Foster and Rahmstorf 2011) to almost all (Kosaka and Xie 2013).”

    I am beginning to think the latter, and so are several other analysts.
    The Southern Oscillation Index (SOI) which you can get a detailed record from NCAR, maps to the vast majority of the fluctuations in the GISS global temperature record. Since the SOI has a reversion to the mean property and the mean is zero, when used as a correction to GISS it essentially completely wipes out the pause-like variations leaving the warming trend behind.

    A tool such as Eureqa is fun to play around with and will do all the compensation automagically.

    I have been pushing this view at Curry’s blog, and some denier said that GISS is already doing this SOI correction in its data set and that it was a fudge. I responded that if they are, they can certainly do a lot better!

  28. 28
    Yvan Dutil says:

    #24 I am wondering if the world climate might simply had shifted from a state to another around that date. Tamino graphics from Kosaka and Xie 2013 seem to indicate the present slowdown in winter is anomalous.

    Do we have pass trough a magic door without noticing it?

  29. 29
    Nick O. says:

    #22 Ray L.
    Quite so, Ray, neither figure is at all reassuring.

  30. 30
    MARodger says:

    Urs Neu @21.

    It perhaps could be pertinent that the decline in sea ice and the ‘slow down’ both appear to be challenging climate models.

    We can be a little more precise about the timing of the ‘effect’ during the winter. If you look at the GISS NH data by the month, the period that is responsible for most of the ‘slow-down’ is actually December to February.
    HadCRUT4 gives similar results although March has been showing more decline in the last couple of years than it does within GISS.

    With all the fluctuating values, the year of the start of this phenomenon is more ambiguous from NH anomalies for individual months of the year. Feb has been declining since 2000 but most months are turning nearer 2006/7. Geographic data may show further clues regarding the start year. For instance, this graph shows the CET 5-year rolling ave anomaly & the SST for the surrounding seas turning down in 2005.

  31. 31

    # 25, @ prokaryotes
    An alternative that may be doable and would be interesting would be a weekly podcast on climate. An example could be Econtalk. Typically, they discuss a recent article or book. From the article you learn economics, from the discussion scientific thinking. Especially in the podcasts where the libertarian host is confronted with reality, the discussions can be quite enlightening and interesting. I could imagine that RealClimate could do something similar for climate.

    #27, @WebHubTelescope
    The Climate Abyss has made a beautiful plot showing the relationship between global mean surface temperature and SOI. By given El Nino and La Nina years a different color and symbol you can see the influence of SOI, without having to “fudge” the data.

  32. 32
    sidd says:

    ” What is the difference between a review and an assessment and a synthesis?”

    Heehee, lemme try

    X says this and that and such and so, agreeing with Y who says this and that and such and so but disagreeing with Z who says this and that and such and so.

    X and Y are wrong. Z is correct.

    X,Y and Z are all correct.

    (This is to be contrasted with some professors and others whom I have known, who seem to live by:If the article supports my work, it is redundant. If it disagrees, it is wrong. If it does neither it is irrelevant. In any case, therefore, it should not be published.)


  33. 33
    SimplyConcerned says:

    About 5 years ago, and again on this blog in 2010, there was a great deal of criticism of a paper published by Keenlyside et al in Nature (back in 2008). The abstract for that paper stated that:

    “Our results suggest that global surface temperature may not increase over the next decade, as natural climate variations in the North Atlantic and tropical Pacific temporarily offset the projected anthropogenic warming.”

    At the time, the paper was heavily criticised by this blog and others both on scientific grounds and, as far as I can tell, because the arguments advanced might be misused by climate scientists. At the time, when I read the paper, it seemed clear to me that the predicted offset of warming would be TEMPORARY, so I could not understand some aspects of the criticism. Frankly, to an outsider it looked like a little bit of jealousy and professional rivalry between modelers about a high profile paper.

    I wonder whether now, given the pause in warming which is being attributed to atmosphere-ocean interaction and subsurface heat storage changes, whether it would be worth a reinvestigation of that 3-5 year old debate.

    Personally, as a non-modeler, I suspect that a little less grandstanding by this blog and others would have been appropriate, given the way things have worked out to date.

    [Response: Sorry – but I disagree. The Keenlyside et al predictions were flawed from the get go, and were quite wrong for their first prediction period and will very likely be wrong for their second prediction period as well (ending in 2015). There is little or nothing to be learned from those simulations about recent trends. – gavin]

  34. 34

    “The Climate Abyss has made a beautiful plot showing the relationship between global mean surface temperature and SOI. “

    That is a nice qualitative view.

    One can further reduce the variability of the GMST by subtracting out the AMO Index. This is more questionable though because the AMO Index is detrended from SST so the variability reduction may be baked in.

    This is how smooth a reconstructed curve looks like when the SOI,volcanic, and AMO components are removed from the GISS temperature record

    The smooth curve was the trend calculated by Eureqa.

  35. 35
    Nick says:

    #3,#8. There is a list of WUWT fails…it’s called WUWT.

  36. 36
    owl905 says:

    @21 who wrote: “the slowdown is more or less restricted to the Northern hemisphere winter” – that highlight may be the boat anchor, but the observations were clear: the Jet Stream went haywire with enormous north-south gyrations and breakdowns. Record-setting cold spells snapped south into Siberia and flooded west & south (snow in Libya). Iirc, 2010’s record high tail-ended with the opposite – a northern hemisphere ‘summer in winter’ when the Jet Stream buckled.
    The mission is really to determine if the La Nina’s and the Jet Stream anomalies are linked, or collateral co-incidences.
    NOAA monthly climate recaps would be helpful, but they’ve been shut down by the pro-pollutionist TeaPartiers. It’s worth checking, if America ever gets back its democracy.

  37. 37
    JesusR says:

    I feel a bit disapointed, the AR5 makes me feel as if I couldn’t read. I’ve been spending the last 5 years on the net denying there was any hiatus in temperatures, defending that the latest estimates of future sea level rise were more than 1 meter, that climate sensitivity estimates were the same as in AR4 and that the model ensemble wasn’t running too warm. Now the AR5 acknowledges that there is an hiatus in temperatures and that a possible source of error are models; projected sea level rise is lower than 1 meter and semi-empirical models are disregarded; the lower bound of likely sensitivity is 0,5 ºC less and there’s no “most likely value” any more (however, in the combination section (Box 12.2 Figure 1 bottom) you can see that combined results tend to lie below 3ºC, with one result lying even below 2 ºC and no result above 3 ºC; it also seems that some climate models of the ensemble are inconsistent with observations (running too warm) (Chap. 11, p 23, about the ASK apporoach).

    I understand that the basic message (the politically relevant) remains the same, but then I don’t see the point of arguing with the deniers about the nuances (how fast things will hapen).

  38. 38
  39. 39
    Nick O. says:

    #31 Victor (and others)

    Have looked at the Climate Abyss site – – and it is a very striking presentation of the data, well worth looking at. Not much sign from that of any pause in the rate of warming.

    Just out today is the state of the oceans report, which is rather bleak reading:

    Not much in there about a pause in global warming either. I do just wonder how well the current GCM’s can account for increased acidification of the oceans and the likely reduction in CO2 drawdown that might result.

  40. 40
    Theo Hopkins says:

    I have just been trying to link from a Wiki article to the (USA) NAOO.

    The home page announces


    Is this the real reason for the temperature pause?

  41. 41
    prokaryotes says:

    Here is the executive summary from the State of the Ocean report 2013 by IPSO.

  42. 42
    prokaryotes says:

    Water in stratosphere plays key role in Earth’s climate

    “A better understanding of the stratospheric water vapor feedback could help explain some of the spread among predictions of future climate change from different models,” referring to the projections made by the recently released 5th Assessment report of the Intergovernmental Panel on Climate Change (IPCC) last week.

    Since Drew Shindell (NASA) wrote on this over a decade ago it would be great to have some input from him on the latest water vapor methane connection in the stratosphere.

    ps. for some reason many NASA webpages are unreachable. Such as “Reaction of Ozone and Climate to Increasing Stratospheric Water Vapor”‎

  43. 43
    prokaryotes says:

    Due to the Federal government shutdown, and most associated web sites are unavailable.

    Was trying to reach Unprecedented Rate and Scale of Ocean Acidification Found in the Arctic

    Wow, the US gov shutdown, it affects education and science.

  44. 44
    prokaryotes says:

    Ocean acidification due to carbon emissions is at highest for 300m years
    Overfishing and pollution are part of the problem, scientists say, warning that mass extinction of species may be inevitable

  45. 45
    Hank Roberts says:

    > I’ve been spending the last 5 years on the net
    > denying there was any hiatus in temperatures

    This is a teachable moment, an opportunity to explain that the warming signal is small compared to natural variability, but that the warming signal is a trend, that can be detected with enough data points.

    How many data points? More than 15 for annual global temperature data (at one point per calendar year), usually, although it’s possible to cherrypick to find, or avoid, results. That’s right out of Statistics 101.

    And you can explain that natural variability averages out to zero trend — over periods longer than fifteen years. How many? More Statistics 101.

    And then you can explain how long we’ve known about CO2 and the climate trend.

    Then point out how they can be misled if they believe the stuff constantly being shoveled out by Morano — and, to be clear, you can point out that there have been for many years sincere, worried and concerned people who have been exaggerating what we can know.

    Nuance is hard.

  46. 46
  47. 47
    Ed Barbar says:

    I’ve been wondering what temperatures would look like today in the absence of CO2, and assuming the models are correct.

    Figure 9.5 of AR-4 indicates temperatures without CO2 forcing would be the same as the 1900s, in the year 2000.

    figure 1.4 from the second order draft of AR5 indicates AR4 models would increase temperatures up to .4 degrees between 2000 and 2013 using the eyeball (and paper) method (note the zero baseline of the two graphs is different), yet temperatures have not increased, so presumably some natural variation sucked out extra heat.

    This means temperatures could be up to .4 degrees colder than in 1900. This historical temperature graph puts the 1900s about .2 degrees “C” above the historical low. Doing the simple math from the IPCC papers, this means the temperatures would be up to .2 degrees lower than in recorded history.

    It seems to me this is a reasonable back of the napkin calculation.

    A) Does anyone know whether AR5 has a similar graph to AR4?
    B) Anything obviously wrong in the above calculations?

  48. 48
    flxible says:

    “Wow, the US gov shutdown, it affects education and science.”

    Exactly the effect some wanted. More importantly, even cancer treatments for some children are affected. But take heart, at least they’re not currently monitoring your internet activity. ;)

  49. 49
    Robert says:

    @43 Prokaryote

    Slight correction on terminology: “US Govt Shutdown” ==> “GOP Shutdown of US Govt”. : )

  50. 50
    Anonymous Coward says:

    note to mods:
    I’m not up-to-date on your policies but Chris Dudley has posted links about nuclear power (not obvious unless you check them out) over at the September Unforced Variations in the last few days which were followed by discussion of nuclear power as people read the links.
    And now he’s starting to post the same links here as well…