A Slow Burn on IPCC Coral Misrepresentation

One of my latest posts discussed how, after nearly a year's delay, the Intergovernmental Panel on Climate Change (IPCC) released material for its latest report. There's a lot of interesting things to be found in it. Today, I'd like to add to a previous post I've written, which you can find here. You don't need to read the post to understand what I'll discuss today, but I'd recommend it if you have the time.

As a quick summary of what I said before, the IPCC wrote about the supposed dangers faced by coral reefs if the world were to warm by 1.5°C or 2.0°C. That discussion took place in several parts of the report, with the different parts making inconsistent statements. On top of this, it was difficult to figure out where the numbers they used came from, and they seemed to misrepresent at least one source they cited. With the release of these materials, we can see how these inconsistencies happened and confirm the IPCC did in fact misrepresent sources to exaggerate the perceived threats of climate change.

Before I continue, I need to point out the "final" version of the report I'll quote here is not the same version the IPCC gave to the world during it's PR campaign announcing the publication of its report. That's because the IPCC told everybody its "Final Government Draft" (FGD) was the final report even though it was a draft report which could undergo any number of changes, even substantial ones. I don't think that will matter today, but it may cause confusion when comparing things quoted in this post to other things you've read. Also, this post will be fairly lengthy (it's a slow burn) so feel free to skip to the end to see the conclusion. With that said, here is text from Chapter Three of the final version of the latest IPCC report:

Even achieving emissions reduction targets consistent with the ambitious goal of 1.5°C of global warming under the Paris Agreement will result in the further loss of 70–90% of reef-building corals compared to today, with 99% of corals being lost under warming of 2°C or more above the pre-industrial period (Frieler et al., 2013; Hoegh-Guldberg, 2014b; Hoegh-Guldberg et al., 2014; Schleussner et al., 2016b; Hughes et al., 2017a).

The assumptions underpinning these assessments are considered to be highly conservative. In some cases, ‘optimistic’ assumptions in models include rapid thermal adaptation by corals of 0.2°C–1°C per decade (Donner et al., 2005) or 0.4°C per decade (Schleussner et al., 2016b), as well as very rapid recovery rates from impacts (e.g., five years in the case of Schleussner et al., 2016b). Adaptation to climate change at these high rates, has not been documented, and recovery from mass mortality tends to take much longer (>15 years; Baker et al., 2008).

There are some weird things in here, like referring to "the ambitious goal of 1.5°C of global warming under the Paris Agreement." The Paris Agreement did aspire to limiting warming to 1.5°C, but the primary goal was 2.0°C. This sentence muddles that. It's not important for today's post, but it is interesting as the IPCC went out of its way to make this sentence muddled. This sentence was given to the IPCC by a reviewer of the Second Order Draft (SOD, the draft prior to the FGD) who wrote:

On line 19, it is not correct to link "well below 2degC" with the "loss of 90% or reef-building corals because the 90% is predicted for the 1.5C scenario (in the reference already cited). Suggest instead saying "Even achieving emission reduction goals consistent with the most ambitious target in the Paris Agreement of 1.5C will result in ...." [Australia]

Of note, this reviewer referred to this goal as "the most ambitious target in the Paris Agreement." This makes it clear there was more than one target in the Paris Agreement, with the text referring to the most extreme case. The IPCC team responded, "Accepted and text changed," but while they accepted this sentence, they specifically chose to delete the word "most," muddling the sentence.

I don't know why the IPCC went out of its way to make a sentence less clear, and I'm not sure how important that is. However, this shows the value of having access to things like reviewer comments. Because of it, we know this wasn't just poor writing. It was an explicit change to text that had already been written. I don't know why it was made, but it shows us how careful we need to be in examining what the IPCC says.

With that in mind, let's look at the text for this section in previous versions, like from the FGD:

Even achieving emission reduction goals consistent with the ambitious goal of 1.5oC under the Paris Agreement will result in the further loss of 90% of reef-building corals compared to today, with 99% of corals being lost under warming of 2o 21 C or more above the pre-industrial period (Frieler et al., 2013; Hoegh-Guldberg, 2014b; Hoegh-Guldberg et al., 2014; Schleussner et al., 2016b; Hughes et al., 2017a).

The assumptions underpinning these assessments are considered to be highly conservative. In some hypothetical cases, ‘optimistic’ assumptions in models include the rapid thermal adaptation by corals (0.2-1.0oC per decade and 0.4o 26 C per decade; (Donner et al., 2005; Schleussner et al., 2016b), respectively) as well as very rapid recovery rates from impacts (i.e., 5 years; Schleussner et al., 2016b). Adaptation to climate change at these high rates (if at all) has not been documented and rates of recovery from mass mortality tend to be much longer (> 15 years; Baker et al., 2008).

The text isn't as polished, and some of the numerical values are slightly different. This shows the final version of the report addresses some issues I had pointed out in my previous post on this topic. I don't know if the changes between these two versions was because of my post or if somebody else happened to notice the same problems independently of me. Either way, it's nice to know I correctly identified a number of discrepancies.

Now that we've seen the public versions of this text, let's look at the previously undisclosed versions. Here is the text from the Second Order Draft (SOD):

Even achieving emission reduction goals consistent with the Paris Agreement (“well below 2oC”) will result in the further loss of 90% of reef-building corals found on reefs today, with 99% of corals being removed under warming by 2oC or more above the pre industrial period (Hoegh-Guldberg et al., 2014; Schleussner et al., 2016b). In some of the latest analyses, the risk of losing coral reefs under 1.5o 22 C increases up until mid-century at which point 90% of coral reefs will have been eliminated. Under this scenario, the loss decreases to 70% by late century, as conditions stabilize. In a 2°C scenario, however, most coral reefs have been eliminated by mid-to-late century.

The assumptions underpinning these grim assessments are considered to be highly conservative. In some hypothetical cases, ‘optimistic’ assumptions adopted by modelers include rapid thermal adaptation by corals 1.0oC per decade and 0.4o 28 C per decade, Donner et al., 2005; Schleussner et al., 2015, respectfully) as well as very rapid recovery rates from impacts (i.e. 5 years, Schleussner et al., 2015). Adaptation to climate change at these high rates (if at all) has not been documented and rates of recovery from mass mortality tend to be much longer (> 15 years, Baker et al., 2008).

This text makes it clear I was right in my previous assessment. My previous post theorized the mentions of 90% were in reference to coral loss of 2050 while mentions of 70% were in reference to coral loss as of 2100 (after some corals had regrown). This text shows I was correct. That information was laid out clearly in the SOD, but in the FGD when the text was compressed, things got muddled.

As for what the text was in the First Order Draft (FOD), there's really not much to say. There was text in this location of the FOD, but it is so entirely different it's difficult to know what one would quote to compare. The reason for this is the authors of the chapter had their attention drawn to certain literature by a reviewer in the reviewer comments.

Section on 'abrupt changes in avoided risks' discusses coral bleaching and mortality at 1.5 and 2 degrees, with no reference
to Schleussner et al. 2016, which found that under 1.5 deg 70% of corals would be at risk of serious degradation. [Bill Hare,
Germany]

Accepted: work by Schleussner et al. 2016 now discussed in considerable detail.

After this exchange, the text was overhauled to provide a different set of conclusions. Given that, and given the results can be taken from Schleussner et al. 2016, it seems clear this study is the source of those results. The numerical values all seem to have been taken from a single study. None of the other studies seem to contradict those results, but I am not sure of the validity of citing numerical results from a single study as representing the "consensus" on a topic. It is particularly strange to rely on a single study's results if one is going to distort what they are. Remember the final version of this line:

Even achieving emissions reduction targets consistent with the ambitious goal of 1.5°C of global warming under the Paris Agreement will result in the further loss of 70–90% of reef-building corals compared to today, with 99% of corals being lost under warming of 2°C or more above the pre-industrial period

Any reader who sees the range "70-90%" here would assume this is a range of potential values due to uncertainty. That is wrong. As seen in the text of the SOD, Schleussner et al. estimated a loss rate of 90% by 2050 and a loss rate of 70% by 2100. That is not an uncertainty range. It's two different results for two different questions. The IPCC combining these like it did is highly inappropriate as there are actual uncertainty ranges given for Schleussner et al's results:

Two models gave a lower range of a 14% loss rate by 2100 under 1.5°C of warming, with one giving a lower range of 48% and the other 60% for 2050. A third model gave an expected value of 9% by 2050 and 1% by 2100, with the lower ranges being 2% and 0% respectively. Presenting these values as a range, rather than explaining they are answers to different questions, hides the significant uncertainties in these results.

Additionally, the IPCC chose to ignore one set of results from this study, with no stated explanation. The "Adaptation" scenario is described in the study:

In addition to the constant scenario, an extremely optimistic scenario of strong thermal adaptation of the corals is assessed (Adaptation). Under this scenario, the critical DHM threshold constantly increases from 2 ?C in the year 2000 up to 6 ?C in 2100. The assumption of a thermal adaptation of 0.4? per decade appears very ambitious

One could perhaps argue a highly optimistic scenario such as this is unrealistic enough the IPCC would be justified in not considering it. However, the IPCC did not ignore this scenario; the IPCC relied on it. This is where the text from the second paragraph I quoted all those times above comes into play. Here it is again, from the final version of the report:

The assumptions underpinning these assessments are considered to be highly conservative. In some cases, ‘optimistic’ assumptions in models include rapid thermal adaptation by corals of 0.2°C–1°C per decade (Donner et al., 2005) or 0.4°C per decade (Schleussner et al., 2016b), as well as very rapid recovery rates from impacts (e.g., five years in the case of Schleussner et al., 2016b).

This is almost unbelievable. We see one of the assumptions of the Adaptation scenario from the Schleussner et al. study listed amongst the assumptions the IPCC says "are considered to be highly conservative." These assumptions only apply to the Adaptation scenario, a scenario whose results the IPCC intentionally excluded from their report.

This means the IPCC took the most optimistic results of this study and threw them out, pretending the model which created them didn't exist. It then took the assumptions of that model to say those assumptions are so extreme the other models, which didn't use them, are likely underestimating the dangers of global warming.

I don't have it in me to discuss just how utterly reprehensible this is. The IPCC hid results which downplay the dangers of global warming, then it used the problems with those results to say other results which didn't have those problems underestimate the dangers of global warming.

There's a lot more wrong with this section, but I'm going to stop here.

5 comments

  1. Hi,
    Thanks for highlighting this.

    I'm not really surprised by your findings. Any group connected to the UN is by definition a political beast. Mixing truth and half truth and putting a spin on things is to be expected.

    I think what suprises me is how they feel they can do this without being challenged in the forum of public debate. There is currently a lot of hot air about "fake news". What they mean is that people are challenging the accepted narratives. The one currently doing the rounds is that we have a climate emergency. It is driven by emotions and not evidence. So the IPCC for a whole host of reasons panders to the current zeitgeist. Perhaps by egging the evidence a little. They have form in that, surely. That is the political animal for you.

    Of course, "scientists" are supposed to be above all that. Hah! As if! They are as easily influenced as a pussy cat and a laser pointer.

    Do keep up the good work. I always enjoy reading what you write. I know it can be dispiriting being the "contrary" voice in a crowd of lemmings, but we do need the contras to try and redirect emotions in a more productive way.

    (I write this as XR is trying to fly drones across the runways at Heathrow.)

    Peter

  2. What intrigues me the most about all this is I can't see how it'd be intentional. If this was malicious, I would expect a certain level of competence. Given how shoddy I'm realizing this work (the part of the report) is, I have to assume the cause is some mix of bias, laziness, ignorance, stupidity or other factors which don't involve nefarious motives. Going through the drafts of the report makes it clear the commentary on corals in this report was written by people with no significant knowledge or expertise on the topic. This isn't an organized conspiracy of people out to push some malevolent plan. It's a conspiracy of stupidity and intellectual laziness.

    The worst part is there are so many problems I don't know how to discuss them all. If I explain in thorough detail just how much the IPCC screwed up on this one issue, how many other issues will I miss out on? I've focused on this one topic because it's the first topic I came across, but how many other examples are there? I mean, look at this paragraph from the Summary for Policy Makers:

    B.4.2. Global warming of 1.5°C is projected to shift the ranges of many marine species to higher latitudes as well as increase the amount of damage to many ecosystems. It is also expected to drive the loss of coastal resources and reduce the productivity of fisheries and aquaculture (especially at low latitudes). The risks of climate-induced impacts are projected to be higher at 2°C than those at global warming of 1.5°C (high confidence). Coral reefs, for example, are projected to decline by a further 70–90% at 1.5°C (high confidence) with larger losses (>99%) at 2ºC (very high confidence). The risk of irreversible loss of many marine and coastal ecosystems increases with global warming, especially at 2°C or more (high confidence). {3.4.4, Box 3.4}

    For the moment, let's leave aside any other issues with this paragraph, like the numerical results listed being inappropriate. Look at those confidence values. This paragraph says "high confidence" and "very high confidence" on results I discussed in some detail in this post. I quoted the report's discussion of these results in multiple drafts. Did you see any mention of how confident we are in those results? I didn't. I checked every other reference to corals in the chapter too. I couldn't find any statement saying we have "high confidence" or "very high confidence" in these results. As far as I can tell, the IPCC report doesn't make those claims anywhere except for in the SPM.

    I may have missed something, but if not, the IPCC just randomly assigned confidence values for results discussed in the SPM that don't match anything in the text of the report. How does that happen? And how do we have "high confidence," much less "very high confidence," in results which come from a single study? Even if a single study could be sufficient for such claims, it's trivially easy to see the IPCC can't even describe that study's results accurately. If it can't do that, how could it possibly be so confident in its results?

  3. I misspoke in my last comment. I said those confidence values are only present in the SPM, but that's not correct. They're also present in the Executive Summary of Chapter 3, the chapter all this material is from. It doesn't change my point as both it and the SPM say the same thing, without any apparent basis for the stated confidence, but it's important to be precise about these things.

  4. Hi Brandon,
    The writing you quote (I haven't read the report directly) is, I would agree, less than precise. It is not clear, precise and accurate, which is the hallmark of good academic writing. (Forget the nuance and qualification that is also required.) In my discipline, you'd not get away with such shoddy writing, but I guess the ultimate editor(s) aren't prepared to wade through the whole thing sending it back to the writers to correct such poorly worded statements.

    I agree its probably incompetence by the IPCC and the academics who are working with them.

    But IMO, it's a worry since "policy" is being made on the back of such statements and I read somewhere that the IPCC is hard at it working on the next assessment. May be it will be better; but I'm not holding my breath.

    Keep up the good work of holding their feet to the fire.

    Peter

  5. I've looked into some other sections now, and I'm unimpressed by the quality of a bunch. I haven't found too many examples where I'd say the report is outright dishonest like I would say of this example,* but there is definitely a low quality of work involved. Quite frankly, I'd call much of this chapter (Chapter 3) shoddy. The only hopeful thing I can say of it is, this isn't one of the main reports the IPCC publishes. This is a more narrowly focused report, and maybe that means they put less effort into it. The Sixth Assessment Report is currently being worked on, and I can only hope they do a better job with it than they did with this one.

    *The report excludes "optimistic" results which make the threat of global warming (to tropical corals) seem less severe, then it used the assumptions for the optimistic results they didn't show to label the results they did show as unduly optimistic. Whether or not any individual intentionally made that misrepresentation (I'd assume not), the final result is dishonest. If nothing else, it's dishonest that the report pretends to be based upon some rigorous writing process when in reality the "best case" explanation is whoever wrote that was lazy and/or incompetent when working on that section. (Of course, the lazy approach may have been because of being overworked and busy, not just a lazy personality.)

Leave a Reply

Your email address will not be published. Required fields are marked *