## Residual airborne fraction

Hope everyone had a good, and safe, festive season. Towards the end of last year I wrote a couple of posts about the ocean carbonare cycle. I then ended up in a debate about what it would take to stabilise concentrations (hint: stabilising emissions will not stabilise concentrations).

One issue that people still seem unsure about is why we expect some fraction of our emissions to remain in the atmosphere for thousands of years. I thought I would try to illustrate this using the basic ocean carbonate chemistry that I described here.

Basically, in equilibrium, the amount of dissolved inorganic carbon (DIC) in the ocean determines the partial pressure of CO2 and, hence, the atmospheric CO2 concentration via Henry’s Law. The top panel of the figure on the right shows this. A DIC of $2002 \mu$mol/kg produces an atmospheric CO2 concentration of 280ppm (pre-industrial). As the DIC increases, so does the equilibrium atmospheric CO2 concentration.

We also know that the ocean holds about 38000 GtC (giga-tonnes of inorganic carbon). Therefore, we can associate a change in DIC with a change in the total amount of carbon in the ocean (i.e., the increase in inorganic carbon in the ocean is approximately ${\rm DIC} \times 38000/2002 - 38000$). Similarly, you can get the increase in atmospheric CO2 using $(p{\rm CO2} - 280) \times 2.12$. The sum of the increase in inorganic carbon in the ocean and the increase in atmospheric CO2 would be our total emission; the fraction of that in the atmosphere would be the residual airborne fraction. This is shown in the bottom panel of the figure on the right. The residual airborne fraction increases from about 15% for emissions of 100s of GtC (we’ve already emitted 600 GtC) to almost 30% if we were to emit as much as 5000 GtC.

The above is all approximate, but I think the basic idea is about right (happy to be corrected if it’s not). It also looks similar to what is obtained in Archer (2005), from which I’ve taken the table below. The “yes” refers to the analysis including temperature feedbacks (as does mine), while the “no” refers to it not including CaCO3 and silicate weathering (mine doesn’t either). The 4th column is total emissions (in GtC) and the 3 final columns are after times of 1kyr, 10kyr, and 100kyr and the reason the values are constant is because weathering isn’t included.

Credit: Archer (2005)

Essentially, in the absence of weathering (which occurs on kyr timescales) the oceans cannot dissolve all our emissions, and the residual amount in the atmosphere increases from about 15% (for total emissions of around 1000GtC) to around 30% (for total emissions of around 5000 GtC).

If anyone would like to download the code I used to produce these figures, it is here. You may need to uncomment some of the lines to get both figures.

## 2016: A year in blogging

Since the year is almost over, I thought I would summarise some of what has gone on here.

January was a rather quiet month. I wrote about a poignant essay by Piers Sellers, who sadly passed away just before Christmas. I also discussed the whole attribution issue, which I may cover again.

February was mainly about the statistical forecast, that not even the author believed.

March saw me writing about some of my own research and pointing out that the Global Warming Policy Foundation (GWPF) doesn’t get that “no warming since 2016/2017” is meant to be a joke, not a suggestion.

April included a post in which I argued that the TCR-to-ECS ratio probably can’t be above 0.8 (which might have relevance for energy balance estimates) and, of course, had a lengthy thread about consensus on consensus (if I ever want a lengthy comment thread, I know to just write about the consensus).

May saw a couple of posts about consensus messaging, including one where I point out that I don’t really like it either (but that’s not necessarily an argument for not using it).

June included a post about a very nice paper that largely reconciles the various climate sensitivity estimates. I also wrote a post about how we don’t even agree on the basics which then had a comment thread that largely illustrated the point.

July’s highlights were probably pseudoscience at UCL (the meeting moved somewhere else, eventually) and the furore over Gergis et al..

August saw me write, again, about some of my own research. The post that caught everyone’s attention was probably the one about Reiner Grundmann’s article calling for more Social Science.

September was again fairly quiet with a post about arguing online and a post about carbon taxes (discussing a Joseph Heath post that I found very useful).

October saw my finally write a post with some details about ocean CO2 uptake (I learned a lot writing it), discussing the Royal Society renting space to the GWPF for a meeting, and a post about Matt Ridley’s lecture at the GWPF meeting.

November included a follow-up of my ocean CO2 uptake post and the start of a my posts about David Rose and his it woz El Nino Wot dunnit.

December included David Rose going further down the rabbit hole (for someone who appear to hate being called a science denier he seems to have no idea about how to not sound like one) a post commiserating with Roger Pielke Jr (okay, not really), and a discussion of why anyone should care about the (scientific) views of sociologists who don’t understand science.

Guest posts:

I had a number of Guest posts (thank you to those who write them). Richard Erskine discussed if we can end the antagonistic climate debate. Steven Mosher had a post about the surface and satellite discrepancy. Lawrence Hamilton, had a post about post-factual perceptions of weather and one on a Tipping Point. Willard also contributed a number of posts. I think that is all, but if I have forgotten someone, let me know.

2017:

2017 looks to be a rather interesting year, possibly for all the wrong reasons. I’m not quite sure what I’ll do with the blog, but then I’ve never been entirely sure, so nothing particularly new there. I think I’m starting to better understand the various issues with how to communicate this topic, but I don’t think that there’s much that I can do that’s different to what I’m currently doing. There’s also only so many times one can explain the scientific evidence, but sometimes it is worth repeating. Will just have to wait and see. I am potentially open to suggestions, but with the proviso that I’m not really trying to take this too seriously, and that there is only so much time I can commit to this.

Update:

There is also now a year in Stoats.

| Tagged , , , , | 27 Comments

## Merry Christmas

Just a quick post to wish everyone a Merry Christmas (or whatever season’s greeting seems most appropriate). Given what’s happening in the Arctic, the cartoon on the right seems apt. I really do hope that everyone has a very pleasant festive season. I plan to take some time off to just relax, spend some time with the family, and to get ready for what is likely to be a busy year. Maybe I’ll try to write some kind of summary of the year post, but we’ll see.

Posted in Climate change, Science | | 20 Comments

## Tipping Point

So the tipping point was social. That thought — a wrong one I hope — came to mind in the aftermath of a US election that set back prospects for reducing greenhouse gas emissions before major ecosystem or physical tipping points are reached. Such a setback has always seemed possible, but its sudden arrival took me by surprise.

Except that in this instance, public opinion has not reversed. Figure 1 combines data from 4 US surveys and 26 quarterly New Hampshire surveys that asked the same climate question from 2010 to August 2016, with highly replicable results. Over this period, the proportion who think that climate change is happening now, caused mainly by human activities, drifted upwards by about 10 points: from the low 50s to well over 60 percent. The pace has been glacial, still this seemed grounds for mild optimism. But how does this upward drift in public acceptance square with the drastic shift to rejection in Washington?

Figure 1

Posted in Climate change, Politics, Research | 36 Comments

## Why should anyone care?

Steve Fuller, who is a sociologist at the University of Warwick, recently wrote an article in the Guardian called Science has always been a bit post-truth. I thought it was confused. Others were somewhat blunter

I’ve written before about the Science and Technology Studies (STS), and have often been unimpressed by what I’ve encountered (not all, though, to be clear). So, I thought maybe this article was just trying to be provocative; to stimulate some kind of discussion from which we might learn something. I tweeted the author to check if that might be the motivation. The answer appears to be not really, but we had a discussion anyway and, sadly, it rather re-inforced my negatives views about STS (also, Steve Fuller seems to prefer retweeting, and then commenting, rather than directly responding, which didn’t help either).

Part of the issue seems to be that some sociologists are jealous (bitter?) that science appears to be taken more seriously, or has more credibility, than they regard as reasonable.

Well, science/research is simply the process of gaining understanding of whatever is being studied. It provides information that we can use to inform whatever decisions may, or may not, need to be made. I’m not entirely sure in what way science is elite, but if it is, maybe that’s because it’s providing information that actually improves our understanding of systems that are of interest. If some socioligists would like their research to be taken as seriously, maybe they should make stronger arguments, rather than trying to normalise science as part of the larger social fabric. Of course, understanding something doesn’t define what we should do – given that information – but that doesn’t somehow imply anything about our understanding.

It also seems that some regard the response from scientists as indicative of a reluctance to be studied.

Well, I have to admit that it’s still not clear to me how we benefit from one group of researchers studying another, but I don’t think that’s really the issue. The real issue seems to be that STS researchers often seem to infer much more from their research than is reasonable. It may well be interesting to study scientists as a group, but it can tell you very little (if anything) about science itself. If you want to understand the systems being studied, you study those systems, you don’t study those who are doing the research.

However, some sociologists seem to think otherwise. I asked about using sociology to conclude things about our actual scientific understanding, rather than just about scientists, and I got this response.

This is really the tweet that motivated this post. It’s clear that scientists have biases that can influence how they interpret their research. However, part of scientific training is to minimise the influence of bias. Also, the scientific method involves reproducibility and replication; we shouldn’t trust individual studies, or individual researchers. We start to trust our understanding when there is a consistent picture across many different studies and many different researchers/research groups.

Now it is possible that biases might influence our understanding sufficiently that we’ll end up going down the wrong track for quite some time. Studying this may well help to avoid this in future. However, it cannot – as far as I can tell – tell us anything about whether or not something we are studying now might be being significantly influenced by the bias of those doing the research. We can’t determine the strength of our understanding of some topic by studying the researchers, but this seems to be what some in STS imply, even if they don’t say so outright. Also, going down the wrong track can be a perfectly reasonable outcome of scientific endeavours; we learn both from our mistakes and from our successes.

Furthermore, why should anyone care if some STS researchers have problems with both philosphers and scientists? Why should their views carry any weight? What research underpins their “problems” with philosophers and scientists? Also, if someone has a problem with aspects of science, or philosophy, they can become a scientist, or a philospher. Sniping from the sidelines is unlikely to be constructive.

Okay, this has all got rather long and I’m not sure I’ve expressed myself as clearly as I would have liked. The real problem I have with STS is the apparent over-reach; inferring much more from their studies than, in my view, is warranted. We really can’t say much about our understanding of some physical system, by only studying those who are doing the research. However, much of what I see from STS implies something close to this and, in my view, spreads more doubt about our understanding than is reasonable. As Michael Tobis would say, they’re swimming outside their lane and, mostly, confusing the public understanding of science. If someone wants to comment on our understanding of some system, they need some understanding of that system and to know something about how people study such systems, not just understand something about the people doing the research.

Update: I should admit that Michael Tobis has commented to point out that I have misunderstood his “swim in your own lane” post. The “swim in your own lane” was actually a reference to a social scientist arguing that physical scientists should not have opinions about public outreach, which – I agree with MT – is wrong. Maybe I’m going to get it confused again, but what MT is suggesting is that there are some social scientists who think that they should be at the interface between science and society. Science communication, however, requires an understanding of the science being communicated and so requires some domain knowledge, which many social scientists do not have.

## No, stabilising emissions will not stabilise concentrations!

I’ve written before about stabilising temperatures; stabilising temperatures requires getting net anthropogenic emissions pretty close to zero. See, for example, this Realclimate post, or Solomon et al. (2008). Stabilising atmospheric concentrations, however, would not require getting emissions to zero, but would still require substantial emission redutions (see this Steve Easterbrook post).

credit : xkcd

Clive Best, who is a physicist, has written a blog post claiming that stabilising concentrations would simply require stabilising emissions (i.e., constant emissions will mean constant atmospheric CO2 concentrations). His basic model is explained here and here. As far as I can tell, his argument appears to simply be that if emissions are increasing, the airborne fraction will be about 50% of our total emissions, but if emissions become constant, the sinks will retain balance – taking up as much as we’re emitting – and that this will happen rather fast (a few years). This, according to Clive, is something that Climate Scientists don’t yet understand. What I suspect Clive does not understand, is why people sometimes draw cartoons like the one on the right.

It should be fairly clear that what is being suggested is wrong; we’re dealing with a coupled system, so if you add new material to one of the reservoirs, it will rise in all reservoirs. However, there is a more formal way to show this. I recently worked through the ocean carbonate chemistry. It turns out that there is a factor called the Revelle factor, which is simply the ratio of the fractional change in atmospheric CO2, to the fractional change in total Dissolved Inorganic Carbon (DIC) in the oceans:

$R = \dfrac{\Delta pCO_2/pCO_2}{\Delta DIC/DIC}.$

The Revelle factor is about 10, which means that the fractional change in atmospheric CO2 will be about 10 times bigger than the fractional change in $DIC$. What this tells you straight away is that you can’t change the amount of CO2 in the oceans without also change the amount in the atmosphere; stabilising emissions will not stabilise concentrations.

Now, maybe if the fractional change in $DIC$ is small enough, the fractional change in $pCO_2$ might also be small enough to essentially stabilise concentrations. However, we know the quantities in the various reservoirs, and we’ve already emitted enough CO2 to change the $DIC$ by 1 – 2%, and – hence – the atmospheric CO2 concentration by 10 – 20%. If we stabilise emissions, we could easily change the $DIC$ by a further 1 – 2%. In fact, we have sufficient fossil fuels to change it by more than 10% and, therefore, enough to change the atmospheric concentration by more than 100% (i.e., to, at least, double atmospheric CO2).

There is, however, something I’m slightly glossing over, so will try to clarify a little more. The above is based on an equilibrium calculation. In other words, it is the changes once the system has retained a quasi-steady equilibrium. Our emissions are continually pushing the system out of equilibrium and so the fractional change in atmospheric CO2 is actually greater than what the Revelle factor would suggest. Given what we’ve already emitted, we would expect about 20% of our emissions to remain in the atmosphere, but it’s currently more like 45%. This is because the timescale for ocean invastion is > 100 years, and so the system hasn’t yet had time to return to equilibrium.

Therefore, the Revelle factor is – in some sense – a lower limit; if we continue to emit CO2 – even at a constant rate – we’d expect at least 20% of what we emit to remain in the atmosphere and, hence, atmospheric concentations will continue to rise. So, unless Clive can find some problem with basic carbonate chemistry, his claim that stabilising emissions will stabilise concentrations is simply wrong.

## Rose down the rabbit hole

To follow-up on his previous article, where he claimed that stunning new data indicates El Nino drove record highs in global temperatures suggesting rise may not be down to man-made emissions, David Rose has a new article in which he claims that Now SECOND set of data shows world temperatures have cooled… and spikes were caused by El Nino – NOT by man. His previous article was very heavily criticised, both in the media, and by scientists. His new article is now claiming that he was right all along, and that his critics are green propagandists.

What I find quite remarkable is that David Rose is essentially presenting his own scientific analysis, despite numerous scientists pointing out that he is wrong, and despite having no actual expertise. However, David Rose is being a little cleverer in this new article, as he has slightly shifted what he’s claiming; he’s now talking about spikes, rather than records. Let’s remind ourselves of what he said in his first article

Some scientists, including Dr Gavin Schmidt, head of Nasa’s climate division, have claimed that the recent highs were mainly the result of long-term global warming.

Others have argued that the records were caused by El Nino, a complex natural phenomenon that takes place every few years, and has nothing to do with greenhouse gas emissions by humans.

The new fall in temperatures suggests they were right.

What’s being suggested by some scientists is that the recent record warm years were not caused by the El Niño; they would have been records even without the El Niño contribution. The reason being that the contribution from El Niño is typically around 0.2oC, or less, and hence even correcting for this leads to these still having been record years.

I thought I would try, here, to illustrate the basic point. The figure on the right shows the HadCRUT4 monthly data relative to the 1961-1990 mean (blue), a 12 month running average (red), and the linear trend from 1970 – October 2016. I’m ignoring the uncertainty in the trend (it would be about ± 0.03oC) as it isn’t really relevant for the basic point. This will also be a little simple, so if you want more detail, read Tamino’s posts.

The table below shows the mid-year trend values, and global annual averages for a recent years that have either been record years, or close to record years. Let’s start with 1998. The mid-year trend value was 0.336; the global annual average for that year (which was associated with a strong El Niño) was 0.537. This is consistent with the El Niño enhancing global temperatures by maybe two-tenths of a oC. Of course, there may be other factors that also influenced this, but it was unlikely to be much more than this. Now consider 2014. The mid-year trend value and the global annual average are quite similar; there is little El Niño effect, and yet it still beats 1998. What about 2015? The global annual average is almost two-tenths of a oC higher than 2014, and yet the El Niño had only just started. There was almost certainly some El Niño influence, but probably not as much as two-tenths of a oC (most estimates suggest a bit less than 0.1oC). Hence, 2015 would probably have beaten 2014, even without some help from the El Niño.

Year Mid-year trend value Annual global average
1998 0.336 0.537
2005 0.458 0.545
2010 0.545 0.558
2014 0.615 0.576
2015 0.632 0.761
2016 0.649 (0.817)

We don’t yet know about 2016 (it’s not yet over) but the average to October is 0.817. It may be that it will only be warmer than 2015 because of a larger El Niño contribution, but that doesn’t change that the recent records were not simply because of the El Niño. I’ve also seen analyses suggesting that 2016 would still beat 2015, even if corrected for El Niño.

I should, however, make some things clear. This is a very simple analysis. The linear trend is not a perfect representation of the forced response, and variations from that trend do not necessarily correctly represent the variability about the forced response (if anyone thinks I’ve blundered spectacularly here, feel free to let me know). However, what this does indicate is that this variability is typically (for annual averages, at least) at the level of two-tenths of a oC, or lower. Hence, this variability (mostly El Niño in the last year or so) cannot explain the recent record warm years, as suggested by David Rose.

A couple of others things for David Rose to ponder (if he bothers to read this, that is).

• David Rose seems to suggest that he would prefer a politer debate about climate change. If so, maybe he should avoid using pejorative labels (like green propagandists) to describe those who disagree with what he presents. He doesn’t have to, but if he does he should probably avoid complaining when he gets labelled in a way he doesn’t like.
• David Rose seemed really put out about being regarded as a science denier. If this was genuine – rather than feigned – maybe he should avoid RTing, and associating with, those who very obviously are. Two examples would be Steven Goddard and James Delingpole.