RealClimate logo

What has science done for us?

Filed under: — rasmus @ 13 December 2016

Where would we be without science? Today, we live longer than ever before according to the Royal Geographical Society, thanks to pharmaceutical, medical, and health science. Vaccines saves many lives. Physics and electronics have given us satellites, telecommunications, and the Internet. You would not read this blog without them. Chemistry and biology have provided use with all sorts of products, food, and enabled the agricultural (“green”) revolution enhancing our crop yields. The science of evolution and natural selection explains the character of ecosystems, and modern meteorology saves lives and help us safeguard our properties.

So what is science? It’s more than just a body of knowledge. It’s a mindset and strategy to build an understanding of our world. This understanding is extremely valuable for our society, especially when it comes to establishing where we stand and what the likely outcomes will be from perceived future actions.

The scientific method is perfect for resolving uncertainties such as controversial claims about facts. It builds on the principles of transparency, testing, and independent replication. Every scientifically trained scholar should get similar results when the analysis is repeated for a finding that is universally true.

Scientific testing and replicating scientific facts are usually based on data analysis and require an understanding of statistical reasoning and what the data really represent. The data analysis is often the point where differences arise. Climate science is no different to other science, and I have myself contributed to the process of checking the findings in a number of controversial papers (Benestad et al., 2016).

There is always a story behind each conclusion that goes back to its roots. The difference between science on the one hand, and dogma and propaganda on the other, is that the latter is not traceable. In other words, you should be more confident about scientific results and sceptical when it comes to intransparent or undocumented claims.

The scientific community has a well-established system for taking care of scientific findings, mainly through publication of papers in the scientific literature. A scientific paper should provide sufficient information for others to replicate the work done and reproduce results. Scientific results are also presented and discussed at conferences, such as the present American Geophysical Union (AGU) fall meeting. The science presented in conferences, however, is not readily broadcasted to the wider society, partly because of difficult language and partly because of limited media presence.

I strongly believe we need a public voice of scientists and historians (see Defending Climate Science), but there is a concern for the future of Earth and space science. It is not just a potential problem for the science community. This is also a genuine worry that affects the wider society and its right to scientific facts and objective information. It is also an issue when it comes to education.

Science benefits everyone and is part of the fabric of our civilisation. It is therefore unwise to dismiss or twist for short-term benefits. The concept “science denial” has been discussed in the magazine called Physics World (September 2016), Nature, blogs, videos, as well as books, just to mention some examples. One of my favoutites is nevertheless the book with the title ‘Agnotology: the Making and Unmaking of Ignorance‘ by Proctor and Schiebinger

History of science can explain how absurd the notion is regarding global warming being a hoax from China. We only need to search for scientific publications from the past, as I did when I wrote a review about the greenhouse effect, based on a paper from 1931 by the American physicist Edward Olson Hulburt (Benestad, 2016)). There is an excellent historical account of modern climate science American Institute of Physics written by Spencer Weart.

It is also a disservice to our society to close down faculties, such as earth observations and climate science. We need both observations and updated analysis more than ever in the times of unprecedented global warming. They are essential inputs to fact-based decision-making concerning our global environment on which we all depend. Our society has progressed and become great much thanks to science, and it would be a sad story for everyone if we were to undo that.


  1. R.E. Benestad, D. Nuccitelli, S. Lewandowsky, K. Hayhoe, H.O. Hygen, R. van Dorland, and J. Cook, "Learning from mistakes in climate research", Theoretical and Applied Climatology, vol. 126, pp. 699-703, 2015.
  2. R.E. Benestad, "A mental picture of the greenhouse effect", Theoretical and Applied Climatology, 2016.

222 Responses to “What has science done for us?”

  1. 201
    Keith Woollard says:

    Ray @ 190

    There are only two things that can increase global temperature on a 100 year timescale? – WOW!

  2. 202
    Romain says:

    Zebra 196,

    Here is a very simple example:
    Take the actual warming of 0.9°C in 120 years. Add the symetric cooling after. Then flat temperatures.
    The rate of increase would be lost in a Marcott like reconstruction.
    And if you think such a scenario is not possible, could you please direct me to the relevant study/blog post/book?
    (we are running in circle, I know…)

  3. 203
    zebra says:

    Romain 202,

    When I gave you my example, you did not do as I asked and describe what the Marcott plot would look like, but you agreed that the temperature change would be “detected”.

    Now, you are saying “the rate” would be “lost”.

    That is not an honest answer, that is a troll answer. I never suggested the Marcott plot would be identical to the true temperature plot; that is obvious, since I asked you to show the difference between true temperature and Marcott in the first place.

    Your 100 years up 1.0C and 100 years down would still be “detected” as an anomaly. In fact earlier you gave an example like this and said that it would show up as .5C– so, where is that .5C change on the Marcott plot?

    If this is what you think of as a “precedent”, then we don’t even need the other reasons why it didn’t happen.

  4. 204
    Ray Ladbury says:

    Keith Woolard: “There are only two things that can increase global temperature on a 100 year timescale? – WOW!”

    Those are not forcings/causes. All I am saying is that you either have to have a very large source of energy to warm the fricking planet or the planet is effectively decoupled into a segment with small thermal mass that warms rapidly and one that is effectively thermally inert. The evidence we have–not least the data on warming in the deep ocean–favors the former. However, if the latter is true, then we are in the soup as we know that anthropogenic CO2 is putting a whole helluva lot of energy into the system.

    What you and Romain are doing is playing “God of the Gaps”. Because you have no evidence to support your side of the argument, you retreat to where there is as yet no evidence at all.

  5. 205
    Thomas says:

    This whole discussion is about temperature change, not temperature.

    Why you believe I didn’t know that is a huge mystery. Thanks for not answering my question and replying with your new question. (sigh) It clarifies what the real problem is here.

  6. 206
    Bob Loblaw says:

    Keith @ 199:

    You say “I know you can transform with non-evenly spaced samples”, but at #185 you said “Yes, I know to do an FT or an FFT requires a constant sample rate”.

    Unfortunately, you can’t seem to decide on what you know.

    You are also now saying “The only reason I brought up Nyquist is it is a simple concept that explains the limitations of sub-sample interpolation.”, yet I pointed you to a web site that shows that Nyquist does not create the same limitation on non-uniformly-space samples – you can identify features that occur at sub-Nyquist intervals. So, your “simple concept” does not explain what you want it to explain.

    I’m afraid your credibility is shot.

  7. 207
    Ric Merritt says:

    I am not interested in the details of the discussion about whether warming in recent decades is “unprecedented” by some definition, or whether past brief spikes may not be evident in some reconstructions. So have at it if you are interested. Seems to me the larger questions run along different lines. Is there any good reason to believe that current warming is a brief spike? Is there any good reason to anticipate a decade (for non-specialists, it’s easiest to stick to calendar decades, 2010-2019 and the like) that will be no warmer than the last, barring very large volcanic eruptions, or more importantly, several decades in a row that fail to show an upward trend of 1-2 degrees per century, or even more? If you don’t believe the past century, and especially the steady decadal rise since the 1970’s, indicates a problem yet, how many more decades would it take to indicate a problem worth combating to you? If you do recognize the current steady rise, along with the evidence it will continue, and find that a problem worth combating, then we’re on the same side about the important things, and I don’t have a beef.

  8. 208
    Romain says:

    Thomas, 198,
    There was no need for your condescending trivialities. No need to bring Hitler into this neither.
    I can only hope you feel better now.
    As for the actual added value of your diatribe, I can’t see any.

    Zebra, 203,
    If you could stop the troll meme, that would be great.
    “Your 100 years up 1.0C and 100 years down would still be “detected” as an anomaly.”
    Not necessarily. Resolution problem. The rate of increase would be lost for sure, and the reconstruction could even be close to flat (like the original Marcott). Anything lasting only a century could be entirely lost.
    My 0.5 degrees increase upthread was just an example of data smoothing, don’t use it for anything else.

    Ray Ladbury 204,
    This is a fair summary, thanks.
    As for the “God of the gaps”, that is a stretch! I am not using gaps in knowledge to prove anything. My only problem is that people don’t even acknowledge there is a gap!

  9. 209
    Keith Woollard says:

    Bob, clutching at straws. There are many ways to transform a time signal to the frequency domain (or pseudo-frequency) I said FT (meaning the classical mathematical way and FFT. Sure there is DFT, and Laplace and a host of others. But again that is all irrelevant. Tamino’s error is he showing you can find periodicities, not that you can reconstruct an unknown signal from sparsely sampled datapoints. Take up my challenge – put the proxy portion of Marcott through Tamino’s hyper-clever transform and show me the solar cycle. That is something we know is there and so you should be able to see it. What I am saying is there is stuff in there we don’t know, can’t see and probably never will.

  10. 210
    Keith Woollard says:

    Thomas@205 :-
    “If I have this right, then your opinion is that the ‘evidence/info’ based on ‘proxies’ for Temp going back 10-20K years is insufficient to logically or reasonably conclude today’s temps are ‘unprecedented’ in this time frame?”

    that is why

  11. 211
    Thomas says:

    @210 Keith Woollard “that is why”

    Well at least I accurately understand what you mean.

    Now your problem is to turn your “opinion” into a valid peer-reviewed paper/s and convince this field of science why your opinion is the more correct one by using evidence and rational analysis of that evidence in a compelling argument.

    Good luck with that.

  12. 212
    Thomas says:

    208 Romain, well you’re welcome to your opinions. I simply say no one likes being shown up for their faulty arguments and having what they actually said repeated back to them, so it’s clearer how little logic was contained in it in the first place.

    Please try and find the point of difference between rhetoric/sophistry and genuine dialectical discussions. Maybe you’re right but just saying is is hardly convincing. (I get that complaint all the time myself) The tough issue is having everyone on the same page and who already agree with the basic knowledge/evidence in an objective sense.

    On social media emotions rule (unfortunately) and social graces were abandoned years ago in favour of ego protection and time saving. :-)

    Good luck with your opinions and theories.

  13. 213

    Woolard, you’re assuming the 11-year Solar cycle (or the 22-year) has a measurable effect on Earth’s mean global annual surface temperature. It probably doesn’t. Want the math?

  14. 214
    zebra says:

    Romain 208,

    You have now officially failed the test.

    “resolution problem”…..”data smoothing”….”close to flat”

    Can you provide any numbers or methodology to demonstrate your conclusions? What’s the difference between a “resolution problem” and “data smoothing” as you understand it? (Complete sentences please.)

    It is obvious that you don’t know what you are talking about here. Keith Woollard knows a little more jargon but is playing the same troll game.

    God of the gaps is not exactly the correct term– what you two are doing is equivocation and moving the goalposts. If you wanted to have a serious discussion, you would specify a lower bound to the physical phenomenon being measured. Are we also required to provide “a study” with appropriate resolution to prove that… the global mean surface temperature did not go up 1C in 1 second, and then down 1C in 1 sec, sometime in the past? Where does it end?

    So, is it possible for the current energy increase in the global climate system, which has been caused by anthropogenic CO2, to return to pre-industrial in 100 years? Absolutely no. (Maybe in 500 if humans radically change their behavior and intervene.) Would that also be true for some hypothetical case in the past 10K years, where something other than CO2 caused the energy increase which caused the temperature increase? Absolutely yes. The latter is true both because of physical constraints and because it would have showed up as an anomaly in Marcott.

  15. 215
    Bob Loblaw says:

    Keith Woolard @ 209:

    Please remember that I first entered this discussion when I commented on your use of an argument I characterized as “but Nyquist…” . You have now (it seems to me) backed away from making that claim, after making a couple of more inconsistent claims that I have pointed out, and you are in the process of trying to shift the goalposts.

    Now, you want to see solar cycles in Marcott – I presume you mean 11- and 22-year cycles, rather than the 10,000-100,000-year cycles associate with Milankovitch theory that clearly could be detected?. Even now, when we have detailed data on short-term solar cycles and temperatures, we can tell that such solar cycles do not have any significant effect on multi-decadal climate variation. Thus, physics tells us that the cycle you have selected would not be detectable even with higher temporal resolution. In addition to shifting goalposts, you are working your way through the common techniques of climate myths by setting Impossible Standards. Arguing that I probably won’t see a flea in my living room is not going to convince me that I can’t see an elephant either.

    You clearly have not thoroughly read Tamino’s posts to understand what they do show, and are trying to challenge me to show something else. I am not interesting in what you think I should be doing. As I said: your credibility is shot. Continue to wave your hands as you see fit, but don’t expect anyone to think you’re providing a sound argument.

  16. 216
    Ray Ladbury says:

    Keith and Romain,
    See, that is the wonderful thing about science–it depends not just on empirical evidence but also on theoretical understanding. Evidently, you are willing to concede that the temperature reconstructions do not provide evidence of such rapid increases in the past. If you are going to contend that such periods are there, but merely not visible, then it is incumbent upon you to propose a mechanism that not only causes such a rapid, sustained rise over a short time period, but than disappears, leaving no trace. What is more, it should be a mechanism that recurs with some regularity, because it could otherwise be dismissed as a rare mechanism that likely has little relevance to current conditions.

    This is a rather tall order, but it is how science works. Try it some time.

  17. 217

    I can’t assume Mr. Woolard will respond to me, so I’ll just quickly do the math here for those who are interested.

    The flux density of sunlight absorbed by the climate system is

    F = (S / 4) (1 – A)

    where S is the Solar constant and A the Earth’s bolometric Russell-Bond spherical albedo. A good recent value for S (Kopp and Lean 2011) is 1361.5 Watts per square meter, and A is about 0.3 (recent determinations range between 0.28 and 0.33, unfortunately, with very little precision). F is therefore about 238.26 W m^-2.

    Kopp and Lean give the recent Solar minimum as S = 1360.8 W m^-2, implying an amplitude of 0.7 W m^-2 and a Solar maximum of 1362.2 W m^-2. Calculating S for minimum and maximum, we find Fmin = 238.14 W m^-2 and Fmax = 238.39 W m^-2.

    Earth’s radiative equilibrium temperature is

    Te = (F / ε σ) T^4

    where ε is the emissivity (always taken as 1.0 at top-of-atmosphere) and σ is the Stefan-Boltzmann constant (5.670373 x 10^-8 W m^-2 K^-4 in the SI). This gives Te(min) = 254.57 K, Te(max) = 254.64 K.

    But of course surface temperatures are higher by about 13% thanks to the greenhouse effect, as mitigated by convection and sensible/latent heat loss to the atmosphere. Let’s say we’re looking for an effect on surface temperatures between 287.66 and 287.74 K.

    The curve is, of course, not perfectly regular by any means, either in mean value, amplitude, or wavelength. Mr. Woolard, then, wants us to find a curve in surface temperature of about 0.04 K amplitude spread over a period of 11-22 years, in a curve with a variance probably several times greater than that. If we can’t do that, he says, neither can we detect a 1.0-1.5 K rise which took 100-200 years.

    You do the math.

  18. 218
    Thomas says:

    214 Bob Loblaw Arguing that I probably won’t see a flea in my living room is not going to convince me that I can’t see an elephant either.

    Excellent retort, not heard that one before. :-)

    Do you think Keith and Romain would agree that one doesn’t need to “observe” a flea visually in their living room to be convinced at least one is there after being bitten by it?

  19. 219
  20. 220
    Keith Woollard says:

    Bob @ 214.
    Yes I have read Tamino’s posts. They are just not very relevant. It is bizarre that there are three posts on the subject and none use real data. His/her/it’s argument is that we can spot a periodic function, therefore we can see all frequencies. Rubbish. Frequency decomposition is not really useful for the climate record unless you are looking for periodic influences.

    Do you honestly believe that the global temperature for the last 10,000 years is as smooth as the proxy reconstructions? And that what we see in the last 120 years is only high frequency because of man??

  21. 221
    Bob Loblaw says:

    Keith @220:

    Now you’re just making me laugh. You mentioned nyquist, first at #172. You then doubled down in #179, saying that “you need to sample that waveform twice“.

    In #184, I pointed you to a source (Tamino) that showed that nyquist is not a limiting factor in all cases – it provides evidence that you can (in at least some cases) detect sub-nyquist cycles when doing a Fourier transform on randomly-spaced data. You then tripled-down in # 185 when you said “I know to do an FT or an FFT requires a constant sample rate, and therefore there is an implied strict Nyquist frequency.“.

    In #194, I pointed out that FT does not require a constant sample rate, to which you replied in #199 that “I know you can transform with non-evenly spaced samples.

    …and now in #220, you just dismiss the Tamino posts as “not very relevant” and create a strawman of what Tamino posts and try to dismiss it.

    It is easy to see that the one thing that is surely “not very relevant” is your posts.

  22. 222
    Keith Woollard says:

    Ok Bob Et Al, let me paraphrase the last 70 or so comments.
    There was a comment that the recent warming was unprecedented in the last (let’s say) 10,000 years. The evidence was reconstructions in the peer reviewed literature. I, and Romain, suggested that the proxy record could not be used to say this due to resolution and smearing. The author of the study agreed due to those same two reasons, as Romain noticed.

    But we are, instead, to trust a blogger who wants to remain anonymous, who only uses model data to show a different problem.

    Wow, just wow.

Leave a Reply

Comment policy. (Note we have re-activated the ReCaptcha plug-in to reduce spam. Please let us know if there are any problems).

Switch to our mobile site