Recent Comments
Prev 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Next
Comments 101 to 150:
-
Noel Yrrep at 07:56 AM on 19 January 2025Fact brief - Can CO2 be ignored because it’s just a trace gas?
Carbon dioxide is a trace gas, but to put that in perspective, the amount of CO2 in the atmosphere is not physically insubstantial.
Focusing on parts per million to the exclusion of actual amounts is employing a mathematical sleight of hand. A sleight of hand that suits those who don't accept the scientific orthodoxy behind the 'warming' role of CO2.
- Each part per million of CO2 in the atmosphere represents approximately 2.13 gigatonnes of carbon, or 7.82 gigatonnes of CO2.
- 422 parts per million represents 3,300 gigatonnes of CO2.
- 422 parts per million represents 3,300,000,000,000 tonnes of CO2.
Which is why "trace" amounts of CO2 supports the massive amount of plant life on land and in water.
-
Philippe Chantreau at 06:28 AM on 19 January 2025Is Nuclear Energy the Answer?
Michael,
I know I won't change your mind about nuclear as a large scale solution, but please give me the courtesy of keeping up the quality of argument expected from all participants here. I don't doubt that you have some notion of what kind of analysis is required before one can claim that some days are "representative" and I am sure that, if anyone on this forum was to make such claim on any subject, they would be asked to produce substantation.
There is an abundant literature about load following and and it disagrees with your contention that it is by nature uneconomical. Here is an example, that I have no doubt you could find just as easily as I could. It also contains information as to how the load following is achieved and it is not through shutting down the reaction. Virtually all reactors currently in use in France were designed for load following. Wikipedia has a page on load following, which is also very easy to find.
The reason EDf was partially privatized in 2004 was to satisfy a EU mandate. The following years saw underinvestment in infrastructure and maintenance. There was a number of problems with several reactors in 2021-22 but these have now been resolved and the parc is again exporting electricity all over Europe, so, once again, using the adverb "lately" is a little imprecise whren talking about the capacity decrease of 2021/22.
I won't dispute that nuclear electricity is more expensive than coal gas or oil, that is a fact. The carbon footprint, however, is much better. Perhaps there is a price to that, although that was not the reason why it was initially chosen as a solution.
I know very little about the US nculear reactor fleet and how it ise used, so I won't comment on that.
My contention is that France's programme has been largely successful and has produced yearly terawatts of carbon free electricity for decades. If not produced from that source, what would they have been using instead over the past 60 years? Oil, coal and gas, like England and Germany or even Denmark? Would that truly have been better?
-
Charlie_Brown at 05:30 AM on 19 January 2025At a glance - Is the CO2 effect saturated?
sychodefender @18 and @22
The band saturation effect does have a diminishing logarithmic curve as explained by Bob Loblaw and as shown in Fig. 6a of Zhong & Haigh 2023. But the rate of modern global warming has not diminished to being negligible. Your descriptors of the effect: "distinct leveling off" and "surely can only have a very slow impact on warming" are misleading toward an incorrect conclusion.
I described the main problem with Kubicki, et al., above @16.
I described the main problem with Wijngaarden & Happer in “Is the CO2 effect saturated?” @716, page 29, Oct 5, 2024.
Moderator Response:[BL] A direct link to Charlie Brown's comment #716 is here. There are discussions of work by Wijngaarden and Happer earlier in the comment thread.
Sychodefender: note that the main article on the "Is the CO2 effect saturated" rebuttal has many, many comments. Some are useful; some are extremely misguided attempts to argue in favour of the myth. In some cases, a head vice is needed.
Addendum: comments from Charlie Brown fall in the "useful" class.
-
sychodefender at 03:46 AM on 19 January 2025At a glance - Is the CO2 effect saturated?
Thanks Bob, I will try to take this all in, appreciate your very comprehensive replies.
-
Jeff Cope at 20:39 PM on 18 January 2025Producing and transporting wind turbine components releases more carbon dioxide than burning fossil fuels
With 90% of wind turbine emissions from its transport, electrifying & renewablizing it would obviously make a tremendous difference practically & even a little help advocating for them. (Most people acting in good faith already know they're vastly better; others won't be convinced by facts, no matter how remarkable.)
I'm curious about whether there are any electric transport vehicles for wind turbine blades & other components, & what emissions are or will be compared to what I assume are now diesel. I think I'd know if any service operation vessels were low emission but I wonder how close they are?
-
Bob Loblaw at 06:54 AM on 18 January 2025At a glance - Is the CO2 effect saturated?
Now for the question posed in #22, about the logarithmic effect of CO2.
There is more than one place in the CO2/climate system where we see logarithmic relationships.
The first place is in the fundamental aspect of the absorption of IR radiation by CO2 (or other greenhouse gases, such as methane). This is described mathematically by Beer's Law, which you can read about on Wikipedia, or by reading this post I made here at SkS about three years ago.
- In non-mathematical terms, the absorption of radiation by a given thickness of air occurs as a proportion of the radiation. If that thickness of air absorbs 10% of the radiation, it will absorb 100W/m2 out of 1000W/m2, but only 10W/m2 out of 100W/m2.
- So, if you start out with 1000W/m2, and absorb 100W/m2, you're left with 900 W/m2.
- But when that 900W/m2 passes through the next identical layer, only 90W/m2 is absorbed, and 810W/m2 is passed on.
- ...and in the next layer, 81W/m2 is absorbed, and 729 W/m2 is passed on.
- ...and so on.
- And this sequence is a logarithmic relationship.
- Although you never get to 0W/m2, after enough layers you do get to the point where it is essentially 0 for all practical purposes. At this point, you can say "with all those layers, absorption has reached a saturation point".
Those "skeptical" of the CO2 effect on climate focus on this "many layers already absorb all the IR" case and then argue "adding more CO2 will not absorb any more". They are wrong.
The catch is that this "saturation" idea only applies when you look at IR radiation that started in the beginning and passed through all those layers. There are two issues with this:
- Even if all the radiation is absorbed by many layers, adding CO2 will change how much was absorbed in layer 1, or layer 2, etc. Thus we are still changing where in the atmosphere the radiation is absorbed.
- This will alter the energy flows in the diagram I posted in the previous comment.
- This diagram shows how changing the absorption rate changes the amount absorbed in the earlier layers. (The diagram comes from this post, which I mentioned earlier.)
- The second issue related to what other commenters have said, that "skeptics" seem to ignore: that the atmosphere itself is emitting more IR radiation.
- Even though less and less of the original radiation entering layer 1 (closest to the surface, if we are thinking of our earth-atmosphere system) reaches the upper layers, constant emission of IR radiation locally (i.e., at that height) is replacing at least some of the IR radiation that was absorbed.
- The local emission depends on local temperature (the kinetic energy source)
- The local emission will be half upwards, and half-downwards, which means that it is now harder for that energy to reach a point where it can be lost to space (greenhouse effect discussion in previous comment).
There is another important place where a logarithmic relationship is seen. In Beer's Law, we talked about a layer containing something that absorbs IR radiation. What happens to the absorption ratio if we double the amount of CO2?
- We might think it doubles the amount of absorption, but this is only the case for low concentrations.
- At higher concentrations, the amount of absorption will not quite double
- ...and at yet higher concentrations, the absorption will not quite not quite double, etc.
- ...so we see a "law of diminishing returns".
- Eventually, at very high concentrations, there will be very little additional absorption. Again, we can call this "saturation".
- ...but current atmospheric CO2 concentrations are a loooong way from reaching this "saturation" point. We're a little over 400ppm now, and we'd need to get to concentrations several times higher before "saturation" is reached.
- From our starting point at 300ppm, we'll see 2-5C rise when we double to 600ppm, and then another 2-5C rise if we double again to 1200ppm, so we can see the logarithmic relationship.
- ...but a climate with 1200ppm of CO2 will not be a pleasant place compared to what we have now. The logarithmic decreasing effect will not save us from a very different world.
Whenever you see a "saturation" or "logarithmic" argument, you need to try to understand which version someone is claiming. The key error in all of them is that they are isolating one small part of a complex system and ignoring other parts that are affected by increasing CO2. Only by including the complex relationships among all the parts of the system can you determine the warming effects of CO2.
-
Bob Loblaw at 06:08 AM on 18 January 2025At a glance - Is the CO2 effect saturated?
sychodefender:
Thanks for providing follow-up questions. It helps us determine just what it is you are trying to understand, and what information we can pass on to help. It is often difficult to know what a person already knows, and what level of explanation is needed.
First, for your question in #21 about where the energy comes from for re-emission of IR radiation.
- All objects (gas, liquid, solid) that have a temperature about 0 Kelvin (-273.15 Celsius) contain kinetic energy that will lead to emission of radiation. Colder object emits primarily at long wavelengths, while hot ones emit at shorter wavelengths. The sun emits primarily in the visible spectrum, while the earth-atmosphere system primary emits in the infrared range.
- Objects will not continually emit radiation unless they have another source of energy. If there was no energy input, the objects would eventually cool to 0K.
- The main source of energy input to the earth-atmosphere system is the sun.
- In a stable climate, the energy absorbed from the sun is exactly offset by the emission of IR radiation to space (averaged over the globe and over a suitable length of time).
- Most of the sun's energy is absorbed at the surface (land, ocean). The warm surface then sends energy back up by three main mechanisms:
- IR radiation
- Thermal transfer of energy from the surface to the air.
- Evaporating water at the surface, moving the water vapour up into the atmosphere, and then condensing the water vapour in the atmosphere. We call this "latent heat transfer" because it involved the latent heat of vaporization of water.
- So the gases (CO2 or others) that emit IR radiation at various points in the atmosphere get the required energy from the sun, after it gets moved around the earth-atmosphere system via the three mechanisms mentioned above.
A key aspect of this is that to understand how CO2 affects climate, a model has to look at all energy flows - not just radiation transfer. Adding CO2 alters the radiation part of the equation, but you can't just isolate the radiation terms. You need to watch that energy play out in the system as the thermal energy and latent heat terms respond.
- The really short version of the greenhouse effect is that the presence of the atmosphere makes it a lot harder for the solar energy absorbed at the surface to get emitted back to space as IR radiation. The surface ends up stabilizing at a much warmer temperature (about 33C warmer) than it would with no atmosphere at all.
- The really short version of adding more CO2 is that it makes it even harder, resulting in an even warmer surface.
Here is a diagram that shows those energy transfers (global averages) pictorially. Note that there are additional fluxes of energy within the atmosphere and back from the atmosphere to the surface. In a full climate model, you also need to consider how these vary globally, and over time (daily, seasonally, etc.)
I'll answer your other question in another comment.
-
One Planet Only Forever at 05:38 AM on 18 January 2025Nobody’s insurance rates are safe from climate change
Eric (skeptic),
Goats were used in 2024.
County of LA Fire Department - Goat Grazing post (linked) opens with the following:
On Thursday, August 15, 2024, the County of Los Angeles Fire Department (LACoFD) was joined by several media outlets in a demonstration showcasing the use of 150 goats and sheep for effective vegetation management near Helispot 69 Bravo in Topanga.
Topanga is the region immediately west of Pacific Palisades.
Also, you state "But I have no weather risk here in rural Virginia." Do you have a scientific reference for your confidence that your region will remain 'weather risk free'? I notice that the regions of North Carolina hit by Helene were less less than 200 miles from the less populated parts of Virginia. With the tracks of hurricanes and their remnants changing due to climate change, hopefully you are correct about your region remaining impact free in the future.
-
One Planet Only Forever at 05:09 AM on 18 January 2025Exploring the drivers of modern global warming
RickyO,
Saying that the 'rate of something has plateaued' means that the rate is constant. The rates had been increasing, meaning more change per year. Now the rates are no longer increasing. That is an improvement. But what is still required is the reduction of the rate to get down to 'zero' net annual impact - which would be when CO2 levels stop increasing.
And, since it is undeniable that 1.5 C warming will be exceeded, to be fair to future generations the current generations need to more rapidly develop to 'net-zero impact'. They need to also draw-down the over-shoot of CO2 and other ghgs levels (years of net-negative ghg impacts). Future generations will have enough trouble dealing with a 1.5 C increase.
-
sychodefender at 22:59 PM on 17 January 2025At a glance - Is the CO2 effect saturated?
Just to be clear, are you saying that co2 does not have a diminishing logarithmic effect ?
The moderator kindly provided a link to a page showing a graph from Zhong and Haigh 2013, which plots a distinct levelling off as co2 levels rise, a logarithmic response that surely can only have a very slow impact on warming.
Apologies if I am traveling old ground but I'm trying to explain this to my children and they are inquisitive!
-
sychodefender at 21:25 PM on 17 January 2025At a glance - Is the CO2 effect saturated?
Thanks for the rapid replies. I understand that co2 absorbs and constantly re-emits in all directions but I'm wondering how this additional kinetic energy comes into existence?
-
RickyO at 21:23 PM on 17 January 2025Exploring the drivers of modern global warming
Zeke,
I'm troubled by the CO2 curve on the graph 'Analysis of decadal warming rates as the sum of different contributing factors between 1905 and 2024' and the explanation "The rate of warming from CO2 has increased over time as emissions have increased, though it has plateaued over the past decade as CO2 emissions have plateaued.
Surely, in considering CO2 as a driver it's the measured amount in the atmosphere which is important, not the partially contrived CO2 'emissions' (as in eg. the carbon neutral burning of wood pellets).
I'm struggling to understand why the rate of warming from CO2 has plateaued in the last 10 yrs, while the conc of CO2 in the atmosphere continues to increase relentlessly (independant of 'emissions')
www.climate.gov/news-features/understanding-climate/climate-change-atmospheric-carbon-dioxide
(Figs 1&2)
-
MA Rodger at 20:41 PM on 17 January 2025At a glance - Is the CO2 effect saturated?
sychodefender @18,
It is incorrect to group the work of Lindzen with that of Kubickiet al, Van Wijngaarden & Happer and that of Schildknect. Lindzen does not dispute the calculated forcing resulting from increasing CO2. His argument is that the resulting feedbacks counteracts the effect of such forcings rather that amplifying them. Dispite may attempts, Lindzen has yet to provide a satisfactory basis for his claim.
The others you list are sinilar in that their basic thesis ignores the IR emissions from the atmospheric gases like CO2 which are generated through the kinetic molecular actions (and thus are dependent on temperature). Such IR comprises the vast majority of IR absotption/emmision within the atmosphere and is readily shown through measurement. Instead, the likes of Kubickiet al, Van Wijngaarden & Happer and Schildknect examine solely the IR directly resulting from surface-emitted IR. Because of such childish error, their work is deemed nonsensical and igmored or, as is the case with Kubicki et al (2024), retracted.
-
One Planet Only Forever at 13:04 PM on 17 January 2025Nobody’s insurance rates are safe from climate change
Erik (skeptic),
The link to your graph using FRED - StLouisFed works.
I note, however, that the graph in the article is ‘relative to inflation’. What that means is not clear (the methodology is not described). But the following appears to confirm the reliability of the graph in the article based on the FRED graphs not being inflation adjusted (note that because of the significant increase in the Home insurance premiums later in 2024 the month chosen in the FRED graph makes a difference in how the values work out):
- The CPI inflation from 2008 to 2024 was 46% (many sources checked – some indicate 1.47). So $1.00 in 2008 is equivalent to $1.46 in 2024.
- The FRED Car insurance premium index went from 125.4 in June 2008 to 216.3 in June 2024. So the increase relative to 2008 was 216.3/125.4=1.725. Relative to inflation it increased by 1.725/1.46=1.18 (18%).
- The FRED Home insurance premium index went from 161.1 in June 2008 to 247.8 in June 2024. So the increase relative to 2008 was 247.8/161.1=1.538. Relative to inflation it increased by 1.538/1.46=1.054 (5.4%). For Sept 2024 the index was 259.0. So the increase relative to Sept 2008 (162.4) was 259.0/162.4=1.595. Relative to inflation the increase would be 1.595/1.46=1.092 (9.2%).
Those values appear to be in line with the end points of the lines on the graph in the article.
-
Bob Loblaw at 11:13 AM on 17 January 2025At a glance - Is the CO2 effect saturated?
sychodefender at 18:
What is your point? All the authors you mention have serious problems in their analysis, so you won't find their results given any credence in the IPCC summaries.
There is no such thing as "the IPCC models". The IPCC reports are a review of the scientific literature, so any models they discuss are models that are developed by individual scientists or groups. Those scientists cooperate on model comparison studies. RealClimate keeps a permanent page on CMIP model comparison results.
There is no need for climate science to come up with a "more recent finding or mechanism" to refute that list of authors, as those authors do not refute the previously-existing understanding that explains current warming (and predicts future warming). Since the old understanding isn't broken, it doesn't need fixing.
Also note that these "at a glance" posts include a link to full rebuttals of the myth, at the bottom of the post (Click for further details), above the list of other rebuttals.
-
sychodefender at 09:59 AM on 17 January 2025At a glance - Is the CO2 effect saturated?
The Kubicki paper seems much the same as those from Happer, Van Wijngaarden, Lindzen, Schildknect and others who all centre their conclusions around the ability of co2 above 300ppm to absorb infrared radiation only at a diminishing logarithmic rate, not comlete but near saturation. I assume the IPCC models include this information as it was mentioned on their website about 20 years ago so they are aware.
I'm guessing that there has been some more recent finding or mechanism by which co2 is declared culpable of creating warming, could someone bring me up to speed please?
Moderator Response:[PS] You can see responses to Kubicki by atmospheric physicists here.
-
Eric (skeptic) at 00:55 AM on 17 January 2025Nobody’s insurance rates are safe from climate change
Here's a different perspective on the chart from the article: https://virtualsnowclub.com/home-and-auto-premiums.png I created my chart at FRED (not sure if their link works): https://fred.stlouisfed.org/graph/?g=1CXFZ
The chart in the article appears to have artifacts from the starting point and the scaling, but also somewhat different recent data.
The grave damage and loss of life in Los Angeles county was a spectacular failure by local officials. In general if fire exclusion is the de facto California policy then they need goats as mentioned above or mechanical clearing. And during a well-predicted and warned fire weather event they need to stage fire fighters to put out fires when they are reported instead of 45 minutes later: https://www.washingtonpost.com/weather/2025/01/12/palisades-fire-origin-new-years-eve-fire/ The Post article also contains an interesting theory about the origin: hot spot from a fire a week earlier. Again, a complete failure by local officials. Caruso, who is far more competent and proposed increasing fire fighting budgets in his campaign, lost the election, but saved his own retail property with private fire fighters.
Anyway about insurance. I had an obscure company for home and GEICO for car. Got a mailing from an Allstate agent and signed up with them. GEICO called after that and tried to pull me back with a new offer, but not as good. Bottom line my homeowners insurance came down a bit from $1600 to $1500 a year. My car insurance (I was overinsured) went from $1300 a year to $600 a year. Part of the decrease is multiple policy discounts.
Allstate is a bait and switch company. The mailing estimated $800 for homeowners. On the day I signed up over the phone for the new $1500 estimate I got another similar low-ball estimate from another Allstate agent in the mail.
Bottom line is insurance is a scam. Yes, they are trying to cover gigantic nationwide liabilities from real disasters. But I have no weather risk here in rural Virginia. Also have not made a homeowners policy claim since 1991 when I started getting homeowners insurance.
-
One Planet Only Forever at 05:18 AM on 16 January 2025Nobody’s insurance rates are safe from climate change
There is an important point regarding the costs of natural disasters. As the article states “Somebody has to pay for the costs to repair, rebuild, and replace damaged homes and vehicles, ...”.
The damage repair costs produce no improvement (they are only attempts to recover and repair. Improvement costs more) but they can count as GDP. So increased natural disaster impacts can be counted as part of increased GDP. And some people only measure progress by simplistic GDP evaluations rather than the more complicated evaluation that would remove any ‘repair or recover value’ and identify the magnitude of damage as a measure of ‘improvement or progress’ (less damage = more improvement or progress).
Note that the 2018 US wildfire season costing 30 Billion (as stated in the earlier YCC article “The role of climate change in the catastrophic 2025 Los Angeles fires” linked to at the end of the 2nd para.) was 0.15% of the 2018 US GDP (20.7 Trillion). That means that wildfires in the US in 2018 caused at least 0.15% decrease of GDP improvement (the decrease was larger since many costs/harms are excluded from the evaluated 'damage done' -> How much is a life worth? How much do legacy health impacts cost?).
-
One Planet Only Forever at 03:33 AM on 16 January 2025Is Nuclear Energy the Answer?
An additional reference for turning off nuclear reactors specific to Canada's CANDU system is Government of Canada: Nuclear Power Plant Safety Systems. It includes the following points:
When the reactor is operating, the chain reaction (or power level) is controlled by moving adjuster rods and varying the water level in vertical cylinders.
Sensitive detectors constantly monitor different aspects, like temperature, pressure and the reactor power level.
When necessary, CANDU reactors can safely and automatically shut down within seconds.
...
Following shutdown, the amount of energy produced by the reactor decreases rapidly.
The nuclear fuel will, however, continue to produce some heat and must be cooled.
That heat, called decay heat, represents a small fraction of the heat produced during normal operation.
-
One Planet Only Forever at 03:24 AM on 16 January 2025Is Nuclear Energy the Answer?
Regarding turning nuclear power reactors on and off.
The World Nuclear Asociation: Nuclear Essentials / How does a nuclear reactor work? includes the following quote:
In order to ensure the nuclear reaction takes place at the right speed, reactors have systems that accelerate, slow or shut down the nuclear reaction, and the heat it produces. This is normally done with control rods, which typically are made out of neutron-absorbing materials such as silver and boron.
Also note that some heat is still produced after the control rods 'stop the reaction'.
-
michael sweet at 01:36 AM on 16 January 2025Is Nuclear Energy the Answer?
Phillippe Chantreau:
The post in the other thread was made in August 2023. The days are from August 2023.
The days were not cherry picked. That affects their statistical value. Random days are more statistcally significant than cherry picked days. I stated how I picked the days.
You have provided no information to support your claim that that reactors are not turned off on weekends. Wikipedia said "some reactors close at weekends" which I read as stopping the chain reaction. Since you do not know what the capacity factor of nuclear reactors is, a basic number that everyone should know, why should I believe your unsupported word the reactors do not stop the chain reaction instead of the references I have read and linked? Please link your references that say the chain reaction is not stopped when the reactors are closed.
The 92% capacity factor for nuclear reactors is a number for the US reactor fleet that is published every year. David-acct has cherry picked the highest year for the reactor fleet with the highest numbers. The USA fleet is usually around 90%. If the reactors run 100% of the time full out they have a capacity factor of 100%. Lower numbers indicate that reactors were off line for some reason. Lower numbers are not economic.
The world fleet has about an 80% capacity factor and France about 70% before 2020. France has been much lower lately due to unplanned maintenance. Since these factors are released every year Daid-acct and I do not need to reference them, they are common knowledge.
EDF, the French nuclear power company, used to be partially privately owned. In 2023 the French government bought out the private investors because the company was bankrupt due to enormous maintenance expenses and the extraordinarily overbudget new build nuclear plants. Despite their 2023 profits they would not survive without very large government support. I am surprised a French citizen is not aware of the buy out. Your comments on government ownership of nuclear plants are simply uninformed.
I always argue that nuclear is too expensive, takes too long to build and there is not enough uranium. These factors are clear in the French fleet.
-
Bob Loblaw at 02:08 AM on 15 January 2025Skeptical Science New Research for Week #1 2025
Yes, Michael and nigel (comments 49, 50) - those other sources support Wallace et al, not refute it.
David-acct, in comment 45, uses my quote from the paper (in comment 44) as an indication that the authors admit the study was "not robust". Let's look again at the key part of the phrase I quoted:
...it is possible that the deaths that our study data did not include may disproportionately occur among individuals registered with a particular political party...
It may, or it may not. To assume that it does is a classic case of confirmation bias. To answer the may/may not question, you'd actually have to obtain the extra data on deaths that Wallace et al were not able to obtain from the sources they used. And then you would have to analyze that data to see if the association between excess deaths and party affiliation is different from the one that Wallace et al found in the data they looked at. And if it is different, is it different enough to affect the overall results.
Note that in the same quote I provided in 44, Wallace et al indicate:
Although overall excess death patterns in our data are similar to those in other reliable sources, such as the Centers for Disease Control and Prevention National Center for Health Statistics data...
So Wallace et al have looked at other sources to see if there are indications that their data might have issues (and they have not found evidence that this is the case).
David-acct would prefer that we believe - as he does - that the "missing" data is a problem. Even though David-acct does not have access to that missing data, and has provided absolutely no evidence or analysis to indicate that those missing data would paint a different picture (i.e., they would have a different statistical distribution from the Wallace et al data that covers 83.5% of deaths in the US).
Of course, we have no idea (i.e., no evidence to support an argument) as to why David-acct is hugely skeptical of these results. His "analysis" presented here is extremely weak. Hypothetical reasons for such resistance could include any or all of the following, which are all varieties of confirmation bias:
- An extreme reluctance to accept the possibility that individual Republicans have been making worse choices than Democrats about their health care.
- An extreme reluctance to accept that people who are making worse choices about their health care can be labelled as Republicans.
- There is a subtle, but important, difference between this possibility and the first one I listed.
- An extreme reluctance to accept that Covid was (and still is) a serious illness that can cause death (i.e., isn't just "another mild flu").
- An extreme reluctance to accept that vaccination is a safe and effective tool in reducing the serious effects of infectious diseases such as Covid.
- A serious reluctance to accept that government-led action in response to community issues (such as health care) can sometimes lead to a better life for its citizens.
- ...or many more...
Note that all of the above fall into the class of "I don't like the results or implications, so there must be something wrong with the study". Conclusions first, analysis later. (Or no analysis, just fling mud at the study.)
And even if we were able to determine what motivates David-acct, it would be an error to extend that to a community level. To do that, we'd need to collect and analyze data...
...and to get back to the OP and the paper listed there, this is what Ecker et al said in their abstract:
Public Significance Statement
This article refutes claims that misinformation is an insignificant issue. Through a critical literature review, we demonstrate that misinformation represents a nontrivial part of the information environment and can causally and adversely influence people’s beliefs, decisions, and behaviors. We clarify through our discussion why misinformation continues to be a significant problem that should not be ignored by communicators and policymakers.To use a classic cartoon from the Christian Science Monitor:
-
Philippe Chantreau at 06:47 AM on 14 January 2025Is Nuclear Energy the Answer?
No worries.
Like I said my focus was not on future solutions that would be applicable worldwide., and I did not claim that nuclear would be that.
I went to the other thread to find the data you and David argued about. I am not sure which year it was from, as the the August 2024 data on these weeks does not seem to correspond (the nuclear share does not dip much below 40GW).
I still have some remarks. In your post responding to David-acct, you said "The days are not cherry picked, they are the only days I looked at." They may not be cherry-picked as in intentionally chosen to show something, but it must be acknownledged that they can not be asserted to be representative of the bigger picture. Furthermore, as was already discussed, the plants are not shut down. This language is somewhat misleading, as is the argument that these 2 days show they can not satisfy demand when needed.
I do not see in either David-acct or your post a proof for or against the 92% claim. It would demand much more data and analysis. What are we exactly considering? An average across all plants? A percentage of time would be a strange way to look at it anyway. Once a reaction is started, it is on pretty much 100% of the time until major maintenance or emergency. There are better metrics.
The reason that France continues to own these reactors perhaps has something to do with the fact that they represent such large infrastrucure investments, are of public utility, of highest concern to public safety and national security. They are indeed an enormous investment. I am not sure it would make more sense to privatize them. Would we consider privatizing all roads? Freeways in France were built with public money, then their exploitation was handed over to private entities, and it is highly questionable whether that was in fact a good move. Some knowledgeable people argue that it was not. As for EDF, they had a net profit of 10 billion euros in 2023.
-
nigelj at 04:56 AM on 14 January 2025Skeptical Science New Research for Week #1 2025
I agree with Michael Sweet @49. Some additional data. The red states had considerably lower vaccination rates than the blue states and the red states had a considerably higher covid mortaility rate after vaccinations were introduced. Please refer:
The most likely explanation is republican voters failed to vaccinate and thus died at a higher rate than democrat voters. The republican voters form majorities in those states and its known they were sceptical of vaccines from polling. I know there are other possible explanations but this is the simplest explanation and Occams Razor tells us its therefore the most likely correct explanation.This is all consistent with the findings of the Wallace study. Therefore I see no reason to be hugely sceptical of the studies findings.
-
michael sweet at 03:47 AM on 14 January 2025Skeptical Science New Research for Week #1 2025
David-acct:
The Our World in Data report I linked at 24 showed that the death rate among the unvaccinated was ten or more times higher than the death rate for the vaccinated. We know that many more Democrats than Republicans were vaccinated. Where is the surprise thar more Republicans died? That is the result we all expect.
As I previously posted about the Wallace et al 2023 paper, Florida had a successful program to vaccinate everyone over 65 when the vaccine was first released. Most deaths were over 65's. That means that Wallace greatly underestimated the Republican death rate. Wallace states the difference they measured was primarily from Ohio.
The death rates from Our World in Data are an independent confirmation of the Wallace paper. Independent confirmation means that Wallace is not misinformation but your claims are misinformation.
-
Bob Loblaw at 12:10 PM on 13 January 2025Skeptical Science New Research for Week #1 2025
David-=acct @ 45:
...and at least Wallace et al know the difference between Florida and Texas, between death rates and excess death rates, and they are consistent in how they use their party affiliation analysis, etc.
I certainly trust their analysis a lot more than I would trust yours.
-
Bob Loblaw at 12:07 PM on 13 January 2025Skeptical Science New Research for Week #1 2025
David-acct @ 45:
The authors do not state that their study in "not robust". As all good scientists do, they have considered possible uncertainties, but there is lots in the paper to support the argument that they are making.
You really need to work on your reading skills.
-
Bob Loblaw at 12:05 PM on 13 January 2025Skeptical Science New Research for Week #1 2025
OPOF @ 43:
I agree that David-acct seems hell bent for leather to discredit Wallace et al, and is blocking out anything that might run counter to that preferred outcome.
As for Ecker et al (the paper listed in the OP), what does it have to say about Wallace et al (2022 - the earlier version of Walalce et al)? Here is the only reference they make to Wallace et al:
....and the partisan gap in COVID-19 vaccination rates
between Republicans and Democrats—which is now
associated with a widening gap in mortality rates (Wallace
et al., 2022)...Nothing more than mentioning an "association". They don't even try to call it a cause-effect relationship. Not proof. Not "shows". Just an "association".
For David-acct, even an "association", a slight implication, is too much for him to bear. The mental gymnastics he is going through to try to find something that will stick to the wall is absolutely astounding.
In comment 27, I gave a list of speculations as to why we see no evidence that David-acct has actually read the Wallace et al paper. I am going to add one more. Granted, this is a highly unlikely explanation, but it's not impossible, so here it goes:
- David-acct is paid to troll web sites and try to inject uncertainty and doubt into climate science, Covid, etc.
- He has a list of responses he has been directed to add to comments sections whenever certain key words show up.
- Whenever misinformation is mentioned, he is supposed to cry "censorship!", and claim "both sides..." arguments.
- Whenever Covid is mentioned, he is supposed to claim the studies are wrong.
- He probably has been given a "cheat sheet" for talking points if Wallace et al is mentioned.
- Since he can't read or understand any of the science in any of these sources, he just keeps returning to the same tired talking points ad nauseum. When his canned answers run out, he just rewords them and repeats...
Now, as I said, this "paid troll" explanation is extremely unlikely, but it is possible and fits the pattern. On the other hand, perhaps he is just arguing in his spare time.
Of course, as he takes all these hits that destroy his arguments, he keeps claiming "tis but a scratch". Tell you what, David-acct: why don't we "call it a draw"???
-
David-acct at 11:51 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
Reply to Bob at 44 -
In summary the authors admit their study is not robust - which is exactly what I stated.
Yet you continued to defend a non robust study!
-
Bob Loblaw at 11:35 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
David-acct @ 42:
OMG. You still haven't read their study, have you?
From their Limitations section:
Second, our mortality data, although detailed and recent, only included approximately 83.5% of deaths in the US and did not include cause of death. Although overall excess death patterns in our data are similar to those in other reliable sources, such as the Centers for Disease Control and Prevention National Center for Health Statistics data, it is possible that the deaths that our study data did not include may disproportionately occur among individuals registered with a particular political party, potentially biasing our results.
They already know that their source of data is not complete, and that this may bias their results.
Have you figured out that your "can't get the death by party affiliation" argument total destroys your eTable 1 "analysis" yet?
Keep waving your hands. Maybe your opponents will catch a cold.
-
One Planet Only Forever at 08:14 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
Bob Loblaw,
David-acct certainly appears to exhibit the behaviours of an unwitting victim of Morton's Demon.
However, I am not inclined to reconsider my update @31.
I am inclined to believe that they are likely aware of, and understand, that they are raising unreasonable doubts ... perhaps hoping to appeal to Others who are unwittingly possessed by Morton's Demon.
I had originally been overly-considerate in my comment @7 by suggesting that they were perhaps unwittingly presenting misunderstandings.
The opening statement of their comment @2, "I agree that misinformation is a problem. However, calls to stop misinformation are essentially calls for censorship.", followed by the unreasonable claim that the Wallace et al 2022 study was misinformation, appears to be a deliberate attempt to unjustifiably discredit or dismiss the Ecker et al study.
I would be inclined to change my mind if David-acct openly declared their full agreement with the Ecker et al study except for having some extremely minor, and possibly incorrect, doubts about Wallace et al.
-
David-acct at 07:18 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
Bob - You have put considerable effort into your responses . As expected you failed to cross check your own work before you criticized my response.
In your responses at #30 - The study stated 10,325,730 US Deaths age 25 or older.
However, the CDC shows 11,567,394 total deaths less 171,444 under 25 leaving 11,395,950 in the US. The study has an error approximately 10% of basic data.
With an error that large with such an easily verifiable number, are you going to continue to defend the reliability of the data and the robustness of the study?
Two serious math errors - How does the study survive scrutiny?
From the #30- It turns out that the first paragraph of that section contains the following sentence, explaining where they got the mortality data:
o We obtained detailed US weekly mortality data from January 1, 2018, to December 31, 2021, from Datavant, an organization that augments the Social Security Administration Death Master File with information from newspapers, funeral homes, and other sources to construct an individual-level database containing 10 325 730 deaths in the US to individuals aged 25 or older during this period..Bob - If the study cant get total deaths correct - How do you expect to get death by party affiliation correct. That belief is simply absurd, not matter how much its sugarcoated.
total under 25
2018 2,839,205 39,434 2,799,771
2019 2,854,838 38,944 2,815,894
2020 2,403,351 44,968 2,358,383
2021 3,470,000 48,098 3,421,902
11,567,394 171,444 11,395,950
10325730
1,070,220
0.09391 10.3645941 % errror My appologies on the formatting of the data from the cdc - It did not copy well from excel. Note - that review of source/raw data is quite valuable. -
Bob Loblaw at 06:55 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
OPOF @ 39:
That CBC story about the cancellation of The Penthouse's X account is priceless.
It sounds to me like X would benefit from being bought by an individual who takes free speech seriously, and wants to establish policies where everyone is free to express themselves, regardless of popularity. We don't want censorship!
[What's that? Oh, right. I forgot about that. Well, ignore that request.]
In addition to the link on motivated reasoning, it's worth reposting the link to Morton's Demon.
-
nigelj at 06:16 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
Bob Loblow @37 &38, very convincing points. You have put a lot of work into that.
-
One Planet Only Forever at 06:06 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
It is undeniably important to increase awareness and improve understanding of the harm of misinformation.
In my comment @15 I make the point that misleading claims need to have limited influence and any harm done needs to be effectively corrected, undone, and/or neutralized.
That is common sense. For an organization (or community or nation or global humanity) to be sustainable and develop lasting improvements it is important that actions and interactions are governed by everyone learning to be less harmful and more helpful to Others.
The following pair of articles are relevant to the issue of harm done by misinformation:
1. CBC News item “'Zombie facts' live on after black plastic and other studies get corrected or retracted” has the sub-heading “Corrections take time and rarely get the attention that the original finding did”.
It opens with the following quote:Headlines warning people to throw out their black plastic kitchen utensils live on, as do social media posts warning of "secret toxins" in your kitchen.
Less prominent? A correction to the peer-reviewed study those headlines were based on.
It includes the following quote:
Though regrettable, errors happen, including in studies that have been peer-reviewed. They can range from a typo or miscalculation that gets a correction, to mistakes so large the paper is retracted, to rare but full-blown fraud. The promise of the scientific process is that by exposing work to the scrutiny of others, any problems will be corrected over time.
The trouble is, it does take time — and the resulting fixes rarely get the public attention of the original errors, say journal editors.
Timothy Caulfield, author of The Certainty Illusion: What You Don't Know and Why It Matters, and a professor at the faculty of law and school of public health at the University of Alberta, studies the twisting of facts and information.
"It was interesting, exciting, it was scary and it got over-promoted," Caulfield said of the black plastic study. "The correction happens and the problem is, there's almost always less uptake of the correction and the original story lives on, right? It becomes a zombie fact that just won't die."
Zombie misinformation also lives on. There are many zombie misunderstandings regarding climate science and vaccine science.
Related to the ‘compartmentalization’ point made by Bob Loblaw @38 is the concept of ‘motivated reasoning (Wikipedia description linked here)’. People will understandably resist and argue against learning when the increased awareness and improved understanding challenges ‘beliefs that they have developed a passionate loyal faith in as an important part of their developed perceived self-identity’. They will understandably present unreasonable doubts about understandings that they ’feel the need to resist learning about’. They will also understandably claim that promoters of learning they feel the need to resist are ‘resistant to learning – they have to be if they ‘doubt the unjustified doubts that are raised’.
That clearly includes people who have to find ways to claim that increased awareness and improved understanding of climate science, vaccine science ‘is misinformation’, or any other important matter, especially when that improved understanding exposes a clear difference between political groups identifiable by their relative interest in learning to be less harmful and more helpful to Others (or themselves).
The article also includes the following advice:
Both Oransky and Caulfield pointed to the importance of media literacy, including critical thinking skills, to counter the spread of misinformation.
Their suggestions include:
• Remember science is complicated with few 'yes' or 'no' answers.
• An immediate recommendation, like to start or stop doing X based on a single study, is rarely evidence-based.
• Keep in mind how scientists are under pressure to produce research quickly that's immediately relevant, which drives science hype.
• Since no study is perfect, the most trustworthy findings are supported by multiple studies that stand up to scrutiny over time."The more evidence that a news article or a TikTok video or a government pronouncement includes, the more I trust it, especially if it includes some nuance and some evidence of 'here's what we don't know,'" Oransky said.
2. CBC News “Vancouver strip club marquee cited as hate speech on X” with the sub-header “Known for its cheeky signage, The Penthouse's latest quip took gentle aim at president-elect Donald Trump” opens with the following:
The X account of Vancouver's Penthouse strip club has been suspended, and not for what you'd think.
The social media platform formerly known as Twitter took action after a photo of the club's latest marquee reading, "Forever neighbours, never neighbors" went viral.
The wording references president-elect Donald Trump's recent trolling of Canada by calling it America's 51st state, and uses the juxtaposition of the Canadian spelling of "neighbour" against the U.S. "neighbor" for political satire.
This is a clear indication of how harmfully misleading some leaders are willing to be and how easy it can be to make social media platforms, or even legacy media organizations, harmful misleading weapons. Yet many people will still passionately defend and excuse undeniably harmful misinformation exploiters and promoters.
-
Bob Loblaw at 02:43 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
Oh, I missed this the first time through.
David-acct says in comment 32:
The implausibility of any state maintaining a reliable data base of deaths by party affiliation.
Yet his favoured eTable 1, with its "devastating" evidence that discredits Wallace et al (2023) contains data tables of the following:
- Counts of Democratic voter deaths, by age.
- Counts of Republican voter deaths, by age.
Somehow, "deaths by party affiliation" are completely unreliable when Wallace et al use them throughout their analysis, but majickly become completely reliable when David-acct wants to use eTable 1 to support his own conclusions.
It is these kinds of logical conflicts that permeate so much of the contrarian argument space. Again, there is a psychological term for this defence mechanism: compartmentalization. From the Wikipedia page:
Compartmentalization allows these conflicting ideas to co-exist by inhibiting direct or explicit acknowledgement and interaction between separate compartmentalized self-states.
Of course, Wallace et al have described how they analyzed the data the obtained to link mortality to party affiliation. I even quoted part of the paper in comment 30. Hint: they did not obtain that data directly from any "state database".
Once again, David-acct could get a better idea of what the authors did - if he'd read the paper. (Or, just fully read all the comments here.)
-
Bob Loblaw at 02:06 AM on 13 January 2025Skeptical Science New Research for Week #1 2025
To further expand on why David-acct's "arguments" in comment 32 are trivial:
- Dismissing the Texas/Florida mistake as a "typo" is deflection. The mistake shows that you are sloppy and are not paying attention to detail.
- You originally stated in comment 2:
- "Per capita death rates for Democrats exceeded per capita death rates for republics [Republicans?] in all categories in both states except for the 85+ age group in Ohio and Texas."
- When I pointed out your error (comment 9) about the Florida age 25-64 age class, I said "David-acct ... may want to try argue that the difference is not significant".
- You have fulfilled that prediction, and now are trying to revise your argument, but that does not change the fact that your original statement was wrong.
- Once again, the mistake shows that you are sloppy and are not paying attention to detail.
- You continue to use a table (eTable 1) that shows total death rates, rather than excess death rates.
- Value for total death rates are not in conflict with any conclusions regarding excess death rates.
- The values in eTable 1 are not broken down by time, so it is impossible for those values to conflict with any analysis that does look at how death rates (excess or otherwise) change over time. (Hint: such analysis is included in the paper, in the portions you continue to fail to look at or discuss.)
- You have yet to explain how any "well known weaknesses in computing excess death rates" apply to the specific methodology used in Wallace et al (2023).
- You have not looked at how Wallace et al examined the sensitivity of their result to changes in assumptions for calculations of excess death rate.
- Even if there are "weaknesses in computing excess death rates", it makes no sense whatsoever to then assume that total death rates are a better way of looking at the issue of Covid-related deaths.
- Your argument about baseline periods ignores the information in the paper about various methods they used to examine the sensitivity of their results to changes in the baseline estimate methodology.
- Once again, you are just making a broad, sweeping generalization - without looking at the specifics in the actual study/paper.
- Your "implausibility" argument is simply an argument from incredulity. From the RationalWiki definition:
- The argument from incredulity is a logical fallacy that occurs when someone decides that something did not happen or does not exist because they cannot personally understand the workings.
There is nothing in your criticisms that would demonstrate that there is any significant problem in the Wallace et al study.
- That does not mean that there might be valid criticisms - it just means that you have failed to show any.
- Another hint: the paper itself has a section titled "Limitations".
- If you had bothered to actually read the paper, you might have been able to use that discussion as a starting point for your criticism.
You continue to grasp at straws. You continue to reject Wallace et al for no good reason. You continue to refuse to look at the paper in full.
There is a psychological term for someone that uses a defence mechanism involving a refusal to accept the truth of a phenomenon or prospect.
- The conclusion from the Wallace et al paper says "Our study found evidence of higher excess mortality for Republican voters compared with Democratic voters in Florida and Ohio after, but not before, COVID-19 vaccines were available to all adults in the US."
- There is something in your world view that simply cannot accept that this conclusion might be right.
- The authors of Wallace et al really have provided evidence, but you need to read the paper to see it.
-
Bob Loblaw at 11:33 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
David-acct @ 32:
You are just repeating your empty assertions. There is no "there" in your criticisms. Repeating them does not make them true. You are the one making arguments out of trivialities. And you can't even get the trivialities right without making elementary mistakes.
I see you have not yet actually responded to anything in the Wallace et al paper other than eTable 1. When you show evidence that you have looked at anything else in the paper, it might be worth listening to you.
-
One Planet Only Forever at 08:36 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
There is a follow-up report from NPR for the item I linked in my comment @15.
NPR – Special Series - Untangling Disinformation: “Meta built a global fact-checking operation. Will it survive?” by Huo Jingnan, Shannon Bond, includes the following quote that is related to the evidence that David-acct repeatedly claimed that efforts to raise awareness about, and limit the harm done by, misinformation are ‘censorship’.
In a video announcing the change, Zuckerberg said fact checking contributed to "censorship" on Meta's platforms and that fact checkers were too "politically biased." Fact checkers point out it is the company, not them, that decides how to police posts on Facebook and Instagram.
"I'm just a simple European but…the United States seems to be the only country in the world where adding information is seen as censorship," said Maarten Schenk, Lead Stories chief operating officer and co-founder.
"Far from censoring, fact-checkers add context," said Laura Zommer, co-founder and CEO of Factchequeado, a nonprofit, Spanish-language fact-checking site that is not part of Meta's program. "We never advocate for removing content. We want citizens to have better information so they can make their own decisions," she added.
Note: The other items presented in the NPR – Special Series - Untangling Disinformation are very informative.
-
michael sweet at 08:35 AM on 12 January 2025Is Nuclear Energy the Answer?
Phileppe Chantreau:
I am sorry I misspelled your name in my last post.
We have already had this discussion in this thread starting at post 322. We disagree on nuclear power. If nuclear power was economic the French government would not have to own 100% of nuclear plants in France.
In the discussion we are following up on, David-acct claimed that no French nuclear plants shut down on weekends. He. demanded I support my claim and linked raw data that he claimed proved the reactors did not shut down on on weekends. When I examined the data I found that it proved that reactors shut down on weekends. I asked David-acct to explain the data but he has refused.
It does not matter why the reactors shut down. It is not economic for nuclear plants to load follow. David-acct claims nuclear plants work 92% of the time. I proved that his claim was false and in France many reactors shut down on the weekend.
Nuclear power is too expensive, takes too long to build and there is not enough uraniun.
-
michael sweet at 08:05 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
David-acct at33:
I note that you have refused to answer my analysis of your "raw data". My analysis showed without any doubt that your claim that French reactors do not shut down on weekends was completely false. Pileppe Changes looked at a different time period. Even in Phileppe's data French reactors shut down on the weekends. Bob Loblaw has shown that your recent " raw data" is not even raw data. I saw another thread where you cited the incorrect chart and the citation did not support your argument. You consistently post misinformation here.
I do not like you trying to hide your false arguments in "raw data" which on analysis does not say what you claim it does.
-
David-acct at 07:48 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
M Sweet at 26
The Wikipedia link was a 2012 post , 12 years out of data. Please cross check your work.
Phillipe Chantreau provided a good response which is based on the data from France's real time grid monitor / Eco2mix which refutes your claim of france "shutting down" their nuclear reactors on the weekends. Further, A broad and general understanding of electric generation from various source , including from nuclear reactors would show why that assertion is inane, its simply not cost or energy effective.
Finally I dont understand your aversion to Raw data or analyizing raw data - your statement - "Please do not cite raw data any more. "
That is the anti-thesis of science. -
David-acct at 07:42 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
Bob - you written extensively criticizing trivial items which do not resolve the deficiencies of the study that I have outlined. None of your comments address the substantive issues which devastates the alleged robustness of the study.
A )inserting Texas instead of Florida was a typo.
B) The 25-64 age group had the lowest per capita death rates and delta between the per capita deaths of republicans vs democrats is statistically insignificant. That does not change the substantive issues with the study.
Your response did not address the following substantive issues:
- There remains 5 of the eight age groups (2 & 3 in the two states ) that conflict with the conclusion. You have not provided any explanation for the results that conflict with the analytical data that survives basis statistical analysis.
- Well Known weaknesses in computing excess deaths with any degree of robustness even when using a statistically valid base period.
- Well known weakness in computing excess deaths using short term base periods, in this studies case, using an absurdly short 4 year base period.
- The implausibility of any state maintaining a reliable data base of deaths by party affiliation.
There is absolutely nothing the study or in your detailed response that would demonstrate the study is even remotely robust.
-
One Planet Only Forever at 07:42 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
Bob Loblaw’s excellent extensive presentation of aspects of the scientific method leads me to add that the scientific method constantly updates understanding as new information becomes available (Bayesian updating).
Based on the new evidence in this comment string I update my comment @7 as follows:
David_acct,
Thank you for (likely knowingly) providing examples of attempts “...to successfully produce misunderstandings through the presentation of misinformation...” that I referred to in my comment @1.
The real thank you goes to Bob Loblaw ... (the rest of my comment, and my other comments, is/are unchanged at this time)
-
Bob Loblaw at 05:21 AM on 12 January 2025Skeptical Science New Research for Week #1 2025
One more foray into the overall world of research. The term "raw data" now appears 42 times in this comment thread. It was first introduced in comment #2. In comment 27, I stated that the use of this term represents an attempt to give certain data an air of authority - "...it must be better because it's raw!" We see this ploy frequently from climate contrarians (a point made by nigelj in comment 6).
So, in the context of the Wallace et al (2023) paper, the infamous eTable 1 was presented as being "raw data". In fact, the participant that introduced the term here has (I think) used "raw data" every time he has referred to that table. Is that table indeed the "raw data" used by Wallace et al?
In a word, no. We can find out more about Wallace et al's data sources by reading the paper. (What a novel idea!) It turns out that they have a section of the paper titled "Methodology".
- It turns out that the first paragraph of that section contains the following sentence, explaining where they got the mortality data:
- We obtained detailed US weekly mortality data from January 1, 2018, to December 31, 2021, from Datavant, an organization that augments the Social Security Administration Death Master File with information from newspapers, funeral homes, and other sources to construct an individual-level database containing 10 325 730 deaths in the US to individuals aged 25 or older during this period.
- The second paragraph starts with the following:
- We linked the mortality data at the individual level to 2017 Florida and Ohio voter registration files;
- Later in the paragraph, we see:
- For each record, the linked data included week of death, age of deceased, county of residence, and 2017 political party affiliation.
- ...and they also explain how they assigned a party affiliation to each individual.
Wallace et al also have a section titled "Statistical Analysis".
- The first sentence in that section is:
- We aggregated weekly death counts from January 1, 2018, to December 31, 2021, at the county-by-party-by-age level.
I'm willing to bet that they also aggregated the data over the entire time period, and by state and age level, in order to get the totals in eTable 1.
- The values in eTable 1 are nowhere close to "raw data".
- Even the data sources used as "raw" data by Wallace et al (i.e., the data they accepted from other sources, rather than collecting themselves) are not "raw" data:
- The mortality data comes from "newspapers, funeral homes, and other sources".
- Even those are not "raw" data.
- Party affiliation is based on voter registration files.
- Even those sources are derived from lower-level sources.
There are many more clues to what Wallace et al actually did buried in the text of the paper - if you take the time to read it.
So, the next time someone tries to pull the "raw data" argument on you, you can be pretty sure that they are trying to pull the wool over your eyes.
Don't drink the "raw data" Kool-Aid!
-
Philippe Chantreau at 08:16 AM on 11 January 2025Skeptical Science New Research for Week #1 2025
Interesting Bob, thanks for doing the leg work.
-
Philippe Chantreau at 03:40 AM on 11 January 2025Is Nuclear Energy the Answer?
I also noted an inconsistency between posts #365 and 367 above:
In # 365, the following quote from Wikipedia is cited: "France's nuclear reactors comprise 90 per cent of EDFs capacity and so they are used in load-following mode and some reactors close at weekends because there is no market for the electricity."
In # 367, this comes back as: "the plants shut down during the highest demand periods during the summer and on weekends. Fossil fuels make up what nuclear fails to generate." That would be the very opposite of what the previous quote says: not following the load.
The reason why some plants are tuned down and/or taken off line is indeed the one in #367, i.e. they are used in load following mode. They are not "shut down" during the highest demand period, they are ramped up, that much is clearly visible in the data. As I have pointed before, the eCO2mix data shows that peak demand is, in the vast majority of cases, associated with peak nuclear production.
Furthermore, fossil fuel use in the generation mix in France is very limited. Oil and coal are almost negligible. Gas is marginal and, to my knowledge, used because of its very fast reaction time. Wind is very well developed: on the morning of January 1st, gas was at 1.8GW, when total wind production was close to 19GW (more than 10 times higher). Compare that to Texas, a leader in the US for wind electricity, where production peaks around 10GW on windy summer days.
Looking at the end of December and the beginning of January so far, I see that the share of gas is lower than wind, and the total amount exported is greater than the share of gas. Looking at longer periods, it is apparent that the overall share of wind power over time is larger than gas. France certainly can't be accused of being a bad actor in limiting the carbon emissions of electricity production in Europe. The European grid is highly interconnected and synchronous (except for the UK), even extending into North Africa; there are a lot of international factors involved in France's total production and level of export. I am skeptical of the claim that the exploitation of their nuclear plants is uneconomical.
In any case, achieving a carbon intensity per kWh that is a factor of 10 lower than neighboring countries, while retaining affordable rates is not a bad result. Despite high amounts of fuels used for road transportation, France has per capita CO2 emissions lower than Denmark, Germany, Finland or Italy. Sweden does better but, like Norway or Quebec, they are in a very privileged position for hydro generation; still, about 30% of their production is from nuclear.
That being said, the long term future needs serious planning. France's existing nuclear power plants can not last for ever. Much has been said of the outages of 2021/22. Some of it was valid, and some of it was spin. One could say that they showed a safety system that works, a grid resilient enough to withstand the outages, and problems that could be detected and solved.
Nevertheless, it must be acknowledged that the 30-50 year horizon demands solutions. I don't know that renewables can be increased to the 50+GW they would have to produce. That is way above my pay grade.
-
Philippe Chantreau at 12:31 PM on 10 January 2025Is Nuclear Energy the Answer?
I'll add that, looking at the eCO2mix tool for week 3 and 4 of July 2024, I see peaks in Nuclear share during the summer at the peak temperature times, which does not suggest that the plants were taken offline or ramped down at these times. I am not sure what data David-acct provided.
-
Philippe Chantreau at 04:10 AM on 10 January 2025Is Nuclear Energy the Answer?
Michael,
At the scale of the country and in the timeline where it happened, nuclear power in France DID significantly reduce reliance on fossil fuel, that is how they achieved such a low carbon emission per kWh. It was the very reason for the Messmer plan and it did succeed to a large extent. As to whether that can be scaled up far beyond the country, that is another question.
There may be a semantic issue between you and David-acct also: once the reaction is started, a nuclear reactor is not "shut down" except in an emergency or for major maintenance. The reaction is always going, albeit with up and down modulations. The part that is "shut down" is the electricity fed into the grid. It would be perhaps more accurate to say they are taken off line or partially off line, but they certainly do not stop the reaction.
-
Bob Loblaw at 02:09 AM on 10 January 2025Skeptical Science New Research for Week #1 2025
Part of this exchange with David-acct has been under the implication of "due diligence". I think it is worth discussing how I have gone about checking things, as it illustrates how science should be done.
(The topic of this thread is "New Research...". How research is done becomes relevant, even if this comment seems like a bit of a tangent.)
To start, in comment 2, David-acct draws attention to Wallace et al (2022) [although he actually meant Wallace et al (2023)]. He referred to data in eTable 1.
- I wanted to look at that data, as David-acct just made a claim that it "devastates the study's conclusion". A very strong claim, that needs supporting evidence. David-acct did not provide any description of what was in eTable 1, or how we should interpret it.
- Wallace et al (2022/23) is not a paper listed in the OP. Where does that reference come from? It is contained in a portion of the paper by Ecker et al (which is listed in the OP), quoted by OPOF in comment 1.
- I noticed that David-acct had erroneously claimed that OPOF made the statement. David-acct did not realize (or care?) that OPOF was quoting someone else.
Tangent: when citing scientific references, you are expected to cite the source where you obtained the information.
- If you are citing Columbus (1492) as a source of information about the discovery of America, you have to have actually read Columbus (1492).
- If you are reading Higgentoot (1776), which looked at the early history of the area that became the United States of America, and Higgentoot says "according to Columbus (1492)...", you are not allowed to repeat what Higgentoot said Columbus said, and cite it in your paper as "Columbus (1492)". You need to cite it as "Columbus (1492), as described in Higgentoot (1776)".
- OPOF correctly identify the quote he placed in comment 1 as coming from Ecker et al.
- David-acct failed to properly reference the quote when he referred to it in comment 2.
So, back to my pursuit of eTable 1.
- From OPOF's original comment, I knew that Wallace et al (2022) had been cited by Ecker et al.
- From Ecker et al, I could get a title.
- I used Google Scholar to look for the paper. I found a copy on-line.
- It did not contain anything called "eTable 1".
- I went back to Ecker et al. Their reference list provided a link to the paper they had used.
- It was the same link I had found using Google Scholar.
- Still no eTable 1.
- I challenged David-acct on this in comment 3.
- In comment 4, David-acct provided a link to the later version of the paper, Wallace (2023).
- From that main page, I was able to eventually find eTable 1.
- I was also able to see that much more information was available, and that David-acct was misrepresenting what was in that table - and being very selective in what information he was using from that paper.
Now, blog posts and comments are not scientific papers, but attention to detail is still important. David-acct was making strong claims about the reliability (or lack thereof) of scientific work.
- He failed to notice key differences between the Wallace et al paper versions.
- He did not provide a link to the paper he was talking about, until challenged.
- This led me on a goose chase.
- The goose chase was constructive, in a fashion, as it helped me realize how sloppy David-acct is with his work.
- He has been highly selective in the portions of the Wallace et al work that he references.
- Exposing this has required that I challenge him to provide links to his sources.
- [I still strongly suspect that David-acct is relying on a secondary source that tells him Wallace et al is unreliable, and David-acct has not actually read or understood the paper. But, as I stated in my previous comment, we'd need more evidence to isolate that explanation from several other possible explanations for him ignoring the bulk of the paper.]
So, this little commentaryhelps illustrate why people commenting here are often asked for references to support their claims.
Tangent 2:
I'm old-school. When I read a scientific paper (or write one), my expectation is that the paper will provide an explanation of methodology, the data used (Observations), the result of any calculations or modelling done with the data (Results), interpretation of the data (Interpretations), and conclusions (Conclusions).
- The ability to make a separation between Observations, Results, Interpretations, and Conclusions is essential at several levels:
- When reading previous papers that provide you with background.
- When doing the work.
- When preparing to write up the work.
- When presenting work to an audience.
Conclusions without clear indications of supporting observations, results, and intepretations are nothing more than opinions.