Dr. Hansen’s climate change paper: scarier than any horror story I have ever read

Decided to look at this due to this NYT article: Scientists Warn of Perilous Climate Shift Within Decades, Not Centuries. I actually started talking about this on twitter, but decided to go with blog because it looked to be getting lengthy and I want to quote.

Bless Justin Gillis, who wrote this article, by the way. Because:

Virtually all climate scientists agree with Dr. Hansen and his co-authors that society is not moving fast enough to reduce emissions of greenhouse gases, posing grave risks.

And:

Yet many of the experts remain unconvinced by some of the specific assertions that were made in the draft paper, and they have not all been persuaded by the final version.

“Some of the claims in this paper are indeed extraordinary,” said Michael E. Mann, a climate scientist at Pennsylvania State University. “They conflict with the mainstream understanding of climate change to the point where the standard of proof is quite high.”

Among Dr. Hansen’s colleagues, some of the discomfiture about the new paper stems from his dual roles as a publishing climate scientist and, in recent years, as a political activist.

Looking at the actual field controversy instead of the manufactured bullshit not-controversy presented by deniers that’s normally sought out for balance. (And also fair to note that Dr. Hansen’s activism does make him stand out. Personally, I’m all for what he does in that arena.)

And it’s fair, because Dr. Hansen is making a pretty big claim. You can get his paper here: Ice melt, sea level rise and superstorms: evidence from paleoclimate data, climate modeling, and modern observations that 2 °C global warming could be dangerous

The abstract alone is well worth reading and I think understandable by non-scientists. (Though if you’re not sure about something, feel free to ask.) The big claim he’s making is right here:

Continued high fossil fuel emissions this century are predicted to yield (1) cooling of the Southern Ocean, especially in the Western Hemisphere; (2) slowing of the Southern Ocean overturning circulation, warming of the ice shelves, and growing ice sheet mass loss; (3) slowdown and eventual shutdown of the Atlantic overturning circulation with cooling of the North Atlantic region; (4) increasingly powerful storms; and (5) nonlinearly growing sea level rise, reaching several meters over a timescale of 50–150 years.

None of the points are revolutionary in terms of current climate science, as far as I know. What makes this a big deal and controversial is the timescale that they’re citing.

Our study germinated a decade ago. Hansen (2005, 2007) argued that the modest 21st century sea level rise projected by IPCC (2001), less than a meter, was inconsistent with presumed climate forcings, which were larger than paleoclimate forcings associated with sea level rise of many meters.

That’s the rub, this disagreement with the IPCC assessments. The IPCC reports basically represent the consensus on climate change, so going outside of them is arguing with the consensus. That doesn’t mean it’s wrong, just that (as Mann is quoted as saying) there’s going to be a really high standard of proof. Here’s what the most recent IPCC report (AR5) said about sea level rise:

Global mean sea level will continue to rise during the 21st century (Table 2.1, Figure 2.1). There has been significant improvement in understanding and projection of sea level change since the AR4. Under all RCP scenarios, the rate of sea level rise will very likely exceed the observed rate of 2.0 [1.7–2.3] mm/yr during 1971–2010, with the rate of rise for RCP8.5 during 2081–2100 of 8 to 16 mm/yr (medium confidence). {WGI SPM B4, SPM E.6, 13.5.1}

Sea level rise will not be uniform across regions. By the end of the 21st century, it is very likely that sea level will rise in more than about 95% of the ocean area. Sea level rise depends on the pathway of CO2 emissions, not only on the cumulative total; reducing emissions earlier rather than later, for the same cumulative total, leads to a larger mitigation of sea level rise. About 70% of the coastlines worldwide are projected to experience sea level change within ±20% of the global mean (Figure 2.2). It is very likely that there will be a significant increase in the occurrence of future sea level extremes in some regions by 2100. {WGI SPM E.6, TS 5.7.1, 12.4.1, 13.4.1, 13.5.1, 13.6.5, 13.7.2, Table 13.5}

So basically, at 8-16mm/year, the IPCC is predicting something more like +0.8-1.6m of sea level in 100 years. Dr. Hansen’s “several meters” is a big jump from this. (The findings in AR5 are generally much more moderate relative to all of Dr. Hansen’s points.)

Moving on to the rest of the paper…

Good that they used the IPCC forcing data, since that does make their simulation results much more readily comparable to prior results. The fewer factors you change between simulations, the better the comparison showing the factors that were changed.

We use these concepts in discussing evidence that most ocean models, ours included, are too diffusive. Such excessive mixing causes the Southern and North Atlantic oceans in the models to have an unrealistically slow response to surface meltwater injection. Implications include a more imminent threat of slowdowns of Antarctic Bottom Water and North Atlantic Deep Water formation than present models suggest, with regional and global climate impacts.

If true, that’s really not good. (Basically, modeling climate response is incredibly hard already because there are so many moving parts. But if this is true, that would indicate a huge inaccuracy in current models, which Dr. Hansen’s group claims means that all projected responses to factors like freshwater input are way too slow.)

Skimmed all the nitty gritty about the simulation specifications because to be honest, it went over my head.

The big driver of their conclusions is how they’re modeling meltwater:

Freshwater injection is 360 Gt year^-1 (1mm sea level) in 2003–2015, then grows with 5-, 10- or 20-year doubling time (Fig. 5) and terminates when global sea level reaches 1 or 5 m. Doubling times of 10, 20 and 40 years, reaching meterscale sea level rise in 50, 100, and 200 years may be a more realistic range of timescales, but 40 years yields little effect this century, the time of most interest, so we learn more with less computing time using the 5-, 10- and 20-year doubling times. Observed ice sheet mass loss doubling rates, although records are short, are 10 years (Sect. 5.1). Our sharp cutoff of melt aids separation of immediate forcing effects and feedbacks.

We argue that such a rapid increase in meltwater is plausible if GHGs keep growing rapidly.

(Note: GHG=greenhouse gas)

There’s the rub. Buying any or all the conclusions means you have to buy the exponential meltwater projection.

Temperature change in 2065, 2080 and 2096 for 10-year doubling time (Fig. 6) should be thought of as results when sea level rise reaches 0.6, 1.7 and 5 m, because the dates depend on initial freshwater flux. Actual current freshwater flux may be about a factor of 4 higher than assumed in these initial runs, as we will discuss, and thus effects may occur 20 years earlier. A sea level rise of 5m in a century is about the most extreme in the paleo-record (Fairbanks, 1989; Deschamps et al., 2012), but the assumed 21st century climate forcing is also more rapidly growing than any known natural forcing.

Arguably not a bad assumption. I just about shit myself when I realized how much faster humans were putting carbon into the atmosphere than whatever mechanism (I’m still on #TeamGeologicalFart) caused the PETM, which is basically the geological poster child for atmospheric carbon-forced rapid climate change. (Rob did the math here, looks right to my eyeballing and I currently don’t have my papers to check.)

They did some simulations with all factors but freshwater input controlled to characterize the change. (Also to allow comparison to their Eemian dataset, since the Eemian had a steady amount of greenhouse gases.) Basically, looked at the amount of freshwater it would take to shut down the Atlantic meridional overturning circulation (AMOC–that thing that helps keep the temperatures in the UK mild) and how that would effect climate once that happened.

The AMOC shuts down for Northern Hemisphere freshwater input yielding 2.5m sea level rise (Fig. 10). By year 300, more than 200 years after cessation of all freshwater input, AMOC is still far from full recovery for this large freshwater input. On the other hand, freshwater input of 0.5m does not cause full shutdown, and AMOC recovery occurs in less than a century.

And this is a very scary thought here:

A key Southern Ocean feedback is meltwater stratification effect, which reduces ventilation of ocean heat to the atmosphere and space. Our “pure freshwater” experiments show that the low-density lid causes deep-ocean warming, especially at depths of ice shelf grounding lines that provide most of the restraining force limiting ice sheet discharge (Fig. 14 of Jenkins and Doake, 1991). West Antarctica and Wilkes Basin in East Antarctica have potential to cause rapid sea level rise, because much of their ice sits on retrograde beds (beds sloping inland), a situation that can lead to unstable grounding line retreat and ice sheet disintegration (Mercer, 1978).

Basically, even though they were simulating an influx of very cold fresh water, it actually caused heating in the deep ocean due to density differences–saltwater is denser than freshwater because of the salt. And if you make the deep waters warmer this could have bad effects on the stability of the remaining ice sheets.

Moving on to the storm predictions, this too requires that you find the exponential meltwater projection to be reasonable.

Our inferences about potential storm changes from continued high growth of atmospheric GHGs are fundamentally different than modeling results described in IPCC (2013, 2014), where the latter are based on CMIP5 climate model results without substantial ice sheet melt.

The conclusion is basically if the meltwater does what Dr. Hansen thinks and shuts down the AMOC, an increase in severe weather is strongly implied.

Increased baroclinicity produced by a stronger temperature gradient provides energy for more severe weather events. Many of the most significant and devastating storms in eastern North America and western Europe, popularly known as superstorms, have been winter cyclonic storms, though sometimes occurring in late fall or early spring, that generate near-hurricane-force winds and often large amounts of snowfall (Chapter 11, Hansen, 2009). Continued warming of low-latitude oceans in coming decades will provide a larger water vapor repository that can strengthen such storms. If this tropical warming is combined with a cooler North Atlantic Ocean from AMOC slowdown and an increase in midlatitude eddy energy (Fig. 21), we can anticipate more severe baroclinic storms.

Yowza. Predicted increase in severe winter storms rather than hurricanes/cyclones, as most people probably imagine when it comes to thinking about severe weather.

Section 4 of the paper is a discussion of the Eemian climate as a dataset. The Eemian wasn’t driven by atmospheric carbon input like today; what makes it interesting is that it had very similar temperatures to our modern world, but sea levels that were at maximum 6-9m higher than those of today–rapid rise due to freshwater input from melting glaciers. The geologic record can and does preserve evidence of storms, so…

The storm activity in the Bahamas during the Eemian dropped 1000t “megaclasts” (read: boulders)  onto the landscape of the time.

…some of the largest boulders are located on MIS 5e deposits at the crest of the island’s ridge, proving that they are not karstic relicts of an ancient landscape (Mylroie, 2008).

That’s… sure something. (Though in fairness, it has been argued that these boulders might be from a tsunami deposit, though Dr. Hansen obviously disagrees with that in his paper.)

The ability of storm waves to transport large boulders is demonstrated. Storms in the North Atlantic tossed boulders as large as 80 t to a height C11m on the shore on Ireland’s Aran Islands (Cox et al., 2012), this specific storm on 5 January 1991 being driven by a low-pressure system that recorded a minimum 946 mb, producing wind gusts to 80 kn and sustained winds of 40 kn for 5 h (Cox et al., 2012). Typhoon Haiyan (8 November 2013) in the Philippines produced longshore transport of a 180 t block and lifted boulders of up to 24 t to elevations as high as 10m (May et al., 2015). May et al. (2015) conclude that these observed facts “demand a careful re-evaluation of storm-related transport where it, based on the boulder’s sheer size, has previously been ascribed to tsunamis”.

brb, shitting pants.

Late Eemian sea level rise might appear to be a paradox, because glacial–interglacial sea level change is mainly a result of the growth and decay of Northern Hemisphere ice sheets. Northern warm-season insolation anomalies were declining rapidly in the latter part of the Eemian (Fig. 26a), so Northern Hemisphere ice should have been just beginning to grow. We suggest that the explanation for a late-Eemian sea level maximum is a late-Eemian collapse of Antarctic ice facilitated by the positive warm-season insolation anomaly on Antarctica and the Southern Ocean during the late Eemian (Fig. 26b) and possibly aided by an AMOC shutdown, which would increase warming of the Southern Ocean.

Oh, that’s interesting. Basically, dating of when temperatures began to fall by looking at the “insolation anomalies”–looking at when summer insolation was no longer enough to prevent ice sheet formation, indicating the move back into a glacial time period–looks like cooling and ice sheet generation (at least in the northern hemisphere) started while sea levels were still going up. Dr. Hansen interprets this as showing that the southern hemisphere was still getting hit with positive insolation, and coupled with freshwater-induced AMOC shutdown, that would keep the ocean warm in that area and promote further ice sheet destruction and melting.

I really recommend that everyone gives section 4.2.2 of the paper a read. It’s a very well-written explanation on the role that CO2 plays in the determination of the Earth’s climate.

Section 5 is where we get to the meet of Dr. Hansen’s argument as to why we should be modeling the meltwater input as nonlinear to exponential rather than linear.

Empirical analyses are needed if we doubt the realism of ice sheet models, but semi-empirical analyses lumping multiple processes together may yield a result that is too linear. Sea level rises as a warming ocean expands, as water storage on continents changes (e.g., in aquifers and behind dams), and as glaciers, small ice caps, and the Greenland and Antarctic ice sheets melt. We must isolate the ice sheet contribution, because only the ice sheets threaten multi-meter sea level rise.

Dr. Hansen goes on to argue that we should accept the data we have for sea level rise (pre-1900 estimates of 0.1-0.2 mm year^-1 rise, tidal gauges 1900-1990 indicating a 1.2+-0.2 mm year^-1 rise, satellites from 1993-present indicating 3 mm year^-1 rise) should be accepted as they are and understood to indicate a distinct acceleration in the rate of sea level rise. Here’s the figure from his paper:

Which is a scary looking line, to be honest, though my geologist sensibilities quail at calling anything less than a thousand years a trend. (Tough. We don’t have 1000 years.)

Dr. Hansen follows with the mass loss data for Greenland and the Antarctic, noting that both rates are accelerating as well, though noting there is currently not enough data (only 10 years worth) to infer a doubling rate or confirm exponential ice loss. (Which is what the most alarming findings of this paper hinge on.)

I’d also encourage everyone to read through the conclusions, since they act as a good summation of what Dr. Hansen’s said in the rest of the paper. But particularly chilling:

A fundamentally different climate phase, a “Hyper-Anthropocene”, began in the latter half of the 18th century as improvements of the steam engine ushered in the industrial revolution (Hills, 1993) and exponential growth of fossil fuel use. Human-made climate forcings now overwhelm natural forcings. CO2, at 400 ppm in 2015, is off the scale in Fig. 27c.

And

Our analysis paints a very different picture than IPCC (2013) for continuation of this Hyper-Anthropocene phase, if GHG emissions continue to grow. In that case, we conclude that multi-meter sea level rise would become practically unavoidable, probably within 50–150 years. Full shutdown of the North Atlantic Overturning Circulation would be likely within the next several decades in such a climate forcing scenario.

Jesus.

First, our conclusions suggest that a target of limiting global warming to 2°C, which has sometimes been discussed, does not provide safety. We cannot be certain that multi-meter sea level rise will occur if we allow global warming of 2 C. However, we know the warming would remain present for many centuries, if we allow it to occur (Solomon et al., 2010), a period exceeding the ice sheet response time implied by paleoclimate data.

And he makes a very good point here:

Second, our study suggests that global surface air temperature, although an important diagnostic, is a flawed metric of planetary “health”, because faster ice melt has a cooling effect for a substantial period.

The climate system is incredibly complex, and yeah, you could melt all the glaciers and it might actually depress global temperatures for a while, which would be a point of curiosity for everyone fleeing the drowned coastal cities, I’m sure. I suppose temperature has become the focus because it’s a solid, small, easy to understand number. (Though trust me, I’ve heard people expressing their confusion as to why 2°C is such a big deal because come on, there isn’t that much of a difference between 23°C and 25°C, etc.)

Third, not only do we see evidence of changes beginning to happen in the climate system, as discussed above, but we have also associated these changes with amplifying feedback processes. We understand that in a system that is out of equilibrium, a system in which the equilibrium is difficult to restore rapidly, a system in which major components such as the ocean and ice sheets have great inertia but are beginning to change, the existence of such amplifying feedbacks presents a situation of great concern. There is a possibility, a real danger, that we will hand young people and future generations a climate system that is practically out of their control.

Well, that paper was scarier than any horror movie I have ever seen.

I’d like to know what other objections scientists in the field have (modeling methods, maybe?), though the big sticking point is if you can buy that glacial melting is accelerating and looking to be exponential. (And if then, the big meltwater input is going to have the effect that these scientists modeled.) The biggest problem is that there isn’t enough data, and waiting around for that data would mean that by the time we have it, it’d be way too late. So that’s… great.

But it comes back to the point that we’re better off believing in this case that it could be way worse than we previously thought. I’d argue that overestimating the threat and over-responding would do less long-term harm to humanity and the planet than underestimating. We’ve already admitted that the way we keep dumping CO2 into the atmosphere is bad–but no one wants to do anything substantive about it. How much scarier does it have to look? How many more warnings do there need to be?

Dr. Mann, as quoted in the NYT, sums it up:

Even scientists wary of the conclusions of the new paper point out that Dr. Hansen has a long history of being ahead of the curve in climate science. As Dr. Mann put it, “I think we ignore James Hansen at our peril.”

(This was a long fucking paper and I’m a geologist with a little oceanography background, not a climate scientist. If I missed something important or got something wrong as I read, please tell me!)

Read this: Response to the paper at ICARUS. This very concisely goes over some of the major criticism.