New IPCC report considered dull

Is it just me or is the new IPCC report a bit dull? I'm talking about The Ocean and Cryosphere in a Changing Climate. I'm looking at the Summary for Policymakers, formally approved at the Second Joint Session of Working Groups I and II of the IPCC and accepted by the 51th Session of the IPCC, Principality of Monaco1, 24th September 2019.

The particularly boring bit is:
The global mean sea level (GMSL) rise under RCP2.6 is projected to be 0.39 m (0.26–0.53 m, likely range) for the period 2081–2100, and 0.43 m (0.29–0.59 m, likely range) in 2100 with respect to 1986–2005. For RCP8.5, the corresponding GMSL rise is 0.71 m (0.51–0.92 m, likely range) for 2081–2100 and 0.84 m (0.61–1.10 m, likely range) in 2100. Mean sea level rise projections are higher by 0.1 m compared to AR5 under RCP8.5 in 2100, and the likely range extends beyond 1 m in 2100 due to a larger projected ice loss from the Antarctic Ice Sheet (medium confidence). The uncertainty at the end of the century is mainly determined by the ice sheets, especially in Antarctica {4.2.3; Figures SPM.1, SPM.5}
Didn't everyone agree the AR5 was a bit conservative and they'd do better next time? But these are hardly changed. I stopped at that point, so if there's something wildly exciting I missed in the second half, do let me know.

Update: reading with Carbonbrief

They're fairly enthusiastic about it, so I could read what they wrote: In-depth Q&A: The IPCC’s special report on the ocean and cryosphere. Well, I could skim it.

* these impacts are projected to have huge costs. In monetary terms, for example, “declines in ocean health and services are projected to cost the global economy $428bn per year by 2050”, the report says, “and $1.979tn per year by 2100”: meh yes, it's easy to get large numbers, but $2T is only 0.37% of global GDP.
Chapter two of the report describes how by the end of the century, glaciers are projected to lose around 18% of their mass compared to 2015 levels under a low-emissions scenario. This anticipated loss doubles to around a third under a high-emissions scenario: just for once I think they're underestimating the impact, by averaging. Those values might be correct globally but locally - e.g. in the Alps - losses will be much higher. Ah yes they continue: In non-polar regions with relatively little ice cover, such as Central Europe and North Asia, the projected outcomes are far more pronounced, according to the report, with on average more than 80% of their current glacier mass gone by 2100.

Or you can try their Explainer: How climate change is accelerating sea level rise (which of course does no such thing); wherein you can sense their frustration with the new report.


* Would you like to be told that IPCC report paints catastrophic picture of melting ice and rising sea levels – and reality may be even worse? Then read Mark Brandon at The Conversation.


1. Nice place, or so I'm told. Still, you can't expect the IPCC to meet in some grimy post-industrial northern city.



Anonymous said...

Is it me or is Stoat going 'Climate Depot'?

I mean to a Climate Depot reader science is dull...Who needs dull old science when sarcasm (he said sarcastically) is easier.

Crikey, I ever misread the host country as 'Principality of Morano'...

Everett F Sargent said...

I think the basic problem is with these ...

Perhaps not quite linear, perhaps quasi-linear to slightly quadratic (say over just the last 40 years or so). The other thing is noone appear willing to put significantly large p-values onto the upper limits of SLR.

You can play connect the dots to have say a two meter SLR by 2100, but to do that you will almost certainly have to see about an order of magnitude increase in the rate of SLR, almost certainly heavily back loaded into the last say 40 years or so.

IMHO they should really be looking at 2150-2300 SLR estimates (which they do but are also very much less certain of those longer time spans).

So yes, the SLR estimates are boring (I would have been really surprised if they weren't). If AR6 is just as boring then all we can say is that SLR science is moving at about the same rate as ice sheet losses, as in really slow.

Heck, we are still stuck with an ECS estimate of 1.5C-4.5C after 40 years.

David B. Benson said...

Bypassed Monaco on the train as I had been told that the place was rather rundown.

William M. Connolley said...

> science is dull

So is there new and interesting science in it? You don't mention any.

> https://climate.nasa.gov/vital-signs/sea-level/

OK, 3.3 mm/yr, but I knew that before.

Everett F Sargent said...

"OK, 3.3 mm/yr, but I knew that before."

I think that that IS my point.

That nothing much has changed is the last six years wrt SLR.

If anything, the ice sheets losses have gotten MORE linear. Remember Hansen et. al. (2016) tried exponential doubling rates (5, 10 and 20 year doubling rates), but the current ice sheet rates have more or less flat lined.

One needs a model that is at least as consistent as the observational data we have to date.

Regional rates? Sure. WAIS goee? Sure. Greenland gone? Sure.

But it is still all a matter of timing. :/

William M. Connolley said...

> That nothing much has changed is the last six years wrt SLR

Hmmm: I thought this was supposed to be focussing on new science. I read This Special Report on the Ocean and Cryosphere1 in a Changing Climate (SROCC) was prepared following an IPCC Panel decision in 2016 to prepare three Special Reports during the Sixth Assessment Cycle. By assessing new scientific literature, the SROCC4 responds to government and observer organization proposals as saying that. But now I re-read it, perhaps it isn't really. OTOH, Carbonbrief claims "The special report “assesses new knowledge” since the IPCC’s fifth assessment report (AR5) – published in 2013-14 – and its 1.5C report". Meh.

Everett F Sargent said...

"Still, you can't expect the IPCC to meet in some grimy post-industrial northern city."

Or even in a pre-industrial tin-shack shanty southern city. You know, like holding their meeting in the ghetto.

At least Cartman tried ...

Everett F Sargent said...

This sums up the human condition, in just under five minutes even ...

izen said...

I expect the Saudis will be delighted that their line-by-line input on the report has rendered it dull...

Phil said...

Economics and morality. Or why climate change isn't just an economic problem.



"Still, you can't expect the IPCC to meet in some grimy post-industrial northern city."

A Silesian coal-mining center like Katowice, say?

No, wait a minute , they just held COP-24 there.

William M. Connolley said...

Katowice is exotic to me :-)

> https://www.nytimes.com/2019/09/30/opinion/climate-change.html

Reasonably good at the non-controversial bits (limiting warming as a public good, problem of commons). But alas runs right off the rails as soon as he tries to solve the problem. I think what this shows is that the proposed formulation was just fluff, designed to lead to his solution.

> This analysis of how hard it is solve the problem of climate change makes clear that the United States needs to take seriously the search for a technological solution to the challenge it poses.

Well no. All the analysis to this point leaves the free-rider problem intact, including tech solutions.

> Consider the problem of street lighting in the 19th century. Suppose you are so much wealthier than everyone else that you have more to lose walking city streets at night than anyone else. You’ve got so much to lose that it’s worth it to you to pay for street lighting all by yourself

Or, Consider the problem of street lighting in the 19th century. Suppose you are so much wealthier than everyone else that you have bodyguards and people to carry torches for you; it's not worth it to you to pay for street lighting, even if others are prepared to contribute too.

> Countries and corporations convinced that their gains from mitigating climate change can outweigh the costs to them will provide the public good to everyone as a byproduct

Indeed. But Countries and corporations convinced that most of the losses from climate change willfall upon others (recall all that stuff about "GW will hit the poor hardest"?) will not have an incentive to act.

So his analogy is just an analogy. Does it apply? It needs far more work that he's put in to see.

Everett F Sargent said...

"Would you like to be told that IPCC report paints catastrophic picture of melting ice and rising sea levels – and reality may be even worse? Then read Mark Brandon at The Conversation."

The following part is about as Bonkers as Monkers ...

"Of course, sea level isn’t static. Intense rainfall and cyclones – themselves exacerbated by climate breakdown – can cause water to surge metres above the normal level. The IPCC’s report is very clear: these extreme storm surges we used to expect once per century will now be expected every year by mid-century. In addition to rapidly curbing emissions, we must invest millions to protect at-risk coastal and low-lying areas from flooding and loss of life."

One should NEVER combine a long term non-stationary SLR with an essentially stationary short term event. It makes absolutely no sense whatsoever from the standard frequency of occurrence standpoint.

An example or three. In 2100AD SLR is two meters, so a storm surge of 10 meters is reduced to 8 meters in keeping with the same total temporal water level.

So what exactly did they do wrong? They took a frequency of occurrence graph and jumped up the whole curve by two meters. Unless they explicitly state otherwise (include appropriate SLR water levels into the the FoC graph and note the long term SLR component explicitly, meaning a 10-year DLR is only appropriate for 10-years in the future). You don't ever map directly from the time domain to the frequency domain, when presented a very long term temporal component.

The SLR is not a stationary event or ever assumed to be a stationary event, it is a long term event that has no frequency of occurrence. On the other hand, pure storm surge is almost always assumed to be a stationary event contingent on the joint probability assumption of iid.

Another example, if SLR in one year is 3.3 mm, then the new storm surge (of 3.3 mm) occurs twice (or once) a day at high tide. Probability of occurrence? To infinity and Beyond!

A third example, eventually SLR is two meters but your tide range is only one meter, part of you is now underwater 100% of the time, must now use negative storm surges.

Finally, my fear is that coastal engineers become very rich as the rate of SLR is currently small enough to spend trillions of dollars in coastal protection structures.

In other words, we will eventually be spending trillions, not millions, in protecting expensive shorelines for up to two meters of expected SLR.

Everett F Sargent said...

Boy, I sure wish you had not mentioned that boring SLR! As I've had a chance to look at their ORIGINAL unpublished research ...

" Relative sea level and extreme sea level events based on tide gauge records"

"To quantify the average return period of extreme sea level events, a peak-over-threshold method is applied following Arns et al. (2013) and Wahl et al. (2017). Tide gauge records are detrended by subtracting a running mean of one year. Peaks above the 99th percentile of hourly water levels are extracted and declustered by applying a minimum time between peaks of 72 hours. This threshold of 99% was recommended by Wahl et al. (2017) for global applications. Using a maximum likelihood estimator, a Generalized Pareto Distribution (GPD) is fitted to these peaks, allowing for an extrapolation to return periods beyond the available period of observations."

Which our people have been doing for over two decades (mostly via MOU's with FEMA and internal research work unit funding) now. Like the method used in this IPCC report is so last century, at least in some parts of the world.

In certain places where SLR is falling due to GIA (e. g. the western side of Hudson Bay), you can get extreme events that have a strong temporal component, in other words a ranking of extreme events is essentially flat lined due to their GIA nature. A ranking of extreme events from low to high loses the GIA temporal nature completely. In those places you have to get a negative amplification factor (see their Figure 4.11 (labeled "Example") as in the allowance is negative and the arrow points towards the RHS not the LHS).

The method of 'so called' amplification is thus almost totally bankrupt, it is used to scare people by artificially generating arbitrary large numbers of 100 or 1000 or even 10,000 (or in very unique situations infinity). Storm surge is storm surge and SLR is SLR, one is permanent the other is not, please don't mix them up willy silly, like ever.

Still their 'so called' analysis IMHO should be considered ORIGINAL unpublished work (not originally published in the peer reviewed literature).

OK, next up is their revision to SLR as exhibited in Table 4.4. You might ask where the 10 cm came from? Well it came from inclusion of Antarctica (which was excluded from AR5 due to excessive uncertainties).

Total AR5 - Antarctica AR5*; 2081–2100 RCP 8.5 = 0.60 (0.43–0.78) meters
Antarctica 2081–2100 (this report) RCP 8.5 = 0.10 (0.02–0.23) meters

Total GMSL 2081–2100 (this report) RCP 8.5 = 0.71 (0.51–0.92) meters

Doing the mate we get 0.71 - (0.10 + 0.60) = 0.01 meters.

A net change after six years of new publications of, wait fo it, a whopping 10 millimeters!

William M. Connolley said...

Yes, I'm a bit surprised it is as little as 10 cm more, after all the complaints after AR5 and work since.

David B. Benson said...

Brexit drives one crazy:

Just pull the covers over your head...