The Thwaites Glacier in West Antarctic. Recent research suggests that part of the huge West Antarctic ice sheet is starting a slow collapse in an unstoppable way. Alarmed scientists say that means even more sea level rise than they figured. (AP Photo/NASA)

On Nov. 2, the United Nations' Intergovernmental Panel on Climate Change will release its "Synthesis Report," the final stage in a yearlong document dump that, collectively, presents the current expert consensus about climate change and its consequences. This synthesis report (which has already been leaked and reported on -- like it always is) pulls together the conclusions of three prior reports of the IPCC's 5th Assessment Report, and will "provide the roadmap by which policymakers will hopefully find their way to a global agreement to finally reverse course on climate change," according to the IPCC chairman Rajendra Pachauri.

There's just one problem. According to a number of scientific critics, the scientific consensus represented by the IPCC is a very conservative consensus. IPCC's reports, they say, often underestimate the severity of global warming, in a way that may actually confuse policymakers (or worse). The IPCC, one scientific group charged last year, has a tendency to "err on the side of least drama." And now, in a new study just out in the Bulletin of the American Meteorological Society, another group of researchers echoes that point. In scientific parlance, they charge that the IPCC is focused on avoiding what are called "type 1" errors -- claiming something is happening when it really is not (a "false positive") -- rather than on avoiding "type 2" errors -- not claiming something is happening when it really is (a "false negative").

The consequence is that we do not always hear directly from the IPCC about how bad things could be.

"Our motivation was really experiencing the IPCC process, and seeing the various ways in which the process, and sort of this seeking consensus, can lead to downplaying the full ranges of future scenarios," comments Bill Anderegg, a Princeton researcher and lead author of the new paper. Anderegg contributed his expertise on ecosystems and climate change in North America in Working Group II of the latest IPCC report.

To show why these researchers think the IPCC is conservative -- and emphatically not alarmist -- you need only consider what the leaked Synthesis Report (which, of course, is still subject to revision) says about the subject of sea level rise. Next to rising temperatures, rising seas are perhaps the most obvious outcome of global warming (because hot air melts ice and expands ocean water). They are also one of the most severe -- and an incredibly big deal if you live in Florida, or North Carolina, or Bangladesh, or the Maldives, or anywhere else with a beach or coast. Knowing just how much sea level could rise, and how fast, is thus vital to help cities and countries plan for how to adapt to a changing world.

By the year 2100, the leaked draft report claims, sea level rise "will likely be in the ranges of 0.26 to 0.55 m for RCP2.6 and of 0.45 to 0.82 m for RCP8.5 (medium confidence)," which is quite similar to what earlier documents from this round of the IPCC's work have said. To translate: For two different scenarios for future greenhouse gas emissions -- one a low end scenario, one a high end one -- there is a 66 percent probability that sea level rise will fall into these two corresponding ranges. And the high end of the range, in the high end emissions scenario, is .82 meters of sea level rise, or 2.69 feet.

Alas, it turns out that these numbers are misleading in several ways -- and may very well be too low. First, .82 meters is not actually the amount of sea level rise that is expected at the year 2100. If you sift carefully enough through the IPCC’s various reports, you will learn that it is rather the mean increase expected between the years 2081-2100, or during the last two decades of this century, when compared with the mean sea level between 1986 and 2005. The actual high end number for 2100 is .98 meters, or 3.22 feet – an amount that “would threaten the survival of coastal cities and entire island nations,” writes climate expert Stefan Rahmstorf of Potsdam University.

But it gets more complicated still -- that's not really the high end number either! Note above that IPCC only gives the range for sea level rise that it considers "likely." What that means, according to Princeton's Anderegg, is that "these ranges are only the middle 2/3 of the probability distribution." In other words, he says, "there is a 17 percent chance it could be lower than that, and a 17 percent chance it could be higher than that." You'd have to be pretty attuned to figure that out, though.

And just when you think you're finally figuring out how bad sea level rise could be by 2100, yet another problem pops up. There are many ways of determining an acceptable range for expected sea level rise, and the IPCC relies on one of them -- so-called "process-based models," which draw on physical equations that govern our understanding of the thermal expansion of the ocean, the melting of ice sheets, and other related factors. But that's not the only way of estimating future sea level rise.

Another way involves getting a group of the top experts together and trying to determine what they think. One such "expert elicitation" came out in late 2013, surveying 90 experts who all publish a lot of research on the topic of sea level rise. And when you do it this way, these experts give a "median likely range" of .7 to 1.2 meters of sea level rise by 2100 in a "high warming" scenario. So here, the high end would be 3.93 feet. (But of course, by the very definition of the process just described, many experts would say even higher than that.)

So by now, you should be getting a sense of how conservative the IPCC is. Indeed, here's a figure from the new paper in the Bulletin of the American Meteorological Society, showing the different ranges for sea level rise in 2100 that have been issued by the IPCC since its inception -- including its extremely controversial lowballing of sea level rise in the Fourth Assessment Report in 2007 -- and comparing those with the aforementioned expert elicitation, to the far right:


Figure comparing the IPCC's five estimates for the range of sea level rise, and one recent expert assessment (EE). IPCC assessments are the First Assessment Report in 1991 (FAR), the Second Assessment in 1996 (SAR), the Third Assessment in 2001 (TAR), the Fourth Assessment in 2007 (AR4), and the Fifth Assessment in 2013-2014 (AR5). Credit: William R.L. Anderegg et al, "Awareness of both Type 1 and 2 errors in climate science and assessment," Bulletin of the American Meteorological Society, 2014. ©American Meteorological Society. Used with permission.

And unfortunately, we still probably haven't discussed what the true worst case scenario is for sea level rise.

There's yet another problem with the IPCC process -- it only considers scientific papers that were published before a particular cutoff date, which in this case, was March 15, 2013. But in May 2014, long after that cutoff date, a blockbuster study came out suggesting that global warming has already irrevocably destabilized the massive West Antarctic Ice Sheet, which contains some 10 feet worth of sea level rise. That is not to say that all of that ice will fall into the ocean immediately and raise sea level, but rather to say that its disintegration, over time, is inevitable. How fast will it happen? That's the big unknown -- but obviously, it is unwise to underestimate an ice sheet, when the consequences around the world would be so devastating.

The lead author of that research, the University of California-Irvine's Eric Rignot, stressed in an interview that there is no scientific consensus yet about the validity of his alarming results. But adds that in his own opinion, the IPCC's estimate for sea level rise is "very conservative."

"We’ve been looking at these glaciers for 20 years, and what I see is defying all these models," adds Rignot.

So in summary, by 2100, sea level rise could be plenty worse than the IPCC suggests -- and realizing this might lead policymakers around the world to view global warming very differently. So then why are its scientific assessments like this? There are surely many reasons, but the authors of the new Bulletin of the American Meteorological Society paper suggest one of them is how much the IPCC has been blasted -- especially over past errors, such as an incorrect prediction that the Himalayan glaciers would vanish by 2035. Because of such flubs, the IPCC has been repeatedly attacked by outside critics -- one of whose favorite epithets is calling the panel "alarmist."

Ironically, perhaps precisely because of all that criticism, it isn't.