The Duck of Minerva

Sandy and Climate Change: Part II

1 November 2012

In my last post, I noted that Hurricane Sandy has finally triggered a vigorous discussion about climate change, prompting much speculation about whether climate change had some role in causing the storm.

I expressed some doubt whether this is the right question to ask, as it frequently has triggered feuds among climate analysts about the science of climate change and diverted limited attention from mobilizing people to act. But, let us suppose it would be politically useful to have some greater certainty that extreme weather events like Sandy are related in some fashion to climate change.

But what do we really want to know?

Causal, but How?
David Roberts of Grist magazine writes that what we really want to know is, among a number of “distal” causes, how “proximate” a cause is climate change of the extreme weather event we are witnessing.

I think that what we’re also interested in is relative causal weight, how important is climate change as a cause of the phenomenon in question. Though the connection between climate change and extreme weather events are physical consequences, this is the same kind of debate we’ve had with respect to climate change and a numnber of social-political outcomes, including conflict (see this exchange between Tad Homer-Dixon and Alex de Waal on Darfur) and famine (see my posts on this blog about the Somali drought of last year). How important a cause is climate change, given everything else?

Andrew Freedman uses the metaphor of a court case and suggests “If this were a criminal case, detectives would be treating global warming as a likely accomplice in the crime.” My comments from yesterday notwithstanding, I think we may need do better than this and found out if climate change is the culprit.

How Unusual was Sandy?
As hurricanes go, Sandy was not the most intense. At its peak, it was at most a Category 2 hurricane. However, as meteorologist David Nolan noted, it is unusual for hurricanes at this time of year to come ashore in the North Atlantic:

Sandy represents the confluence of a modestly strong hurricane coming up from the Caribbean and a fairly strong but not unusual continental weather system — what meteorologists call a “trough” or “dip” in the jet stream — that has been traversing the country over the last few days.n the vast majority of cases, these weather systems weaken hurricanes while at the same time pushing them quickly out to sea.

Jeff Tollefson in Nature echoes this perspective.

Hurricane Sandy is potentially unprecedented, officials say, because of the meteorological context within which it has developed. First, before reaching land, it was feeding off unusually warm surface waters in the Atlantic Ocean. Second, whereas such storms tend to skirt the US coast before drifting to the northeast and dissipating at sea, Sandy has been influenced by a high-pressure system off Greenland that has forced it inland. In doing so, the storm has merged with a winter system moving in from the west, putting forecasters in the unusual position of having to issue snow advisories for a tropical-hurricane system. Finally, the effects of the storm may have been enhanced by a full Moon, which generally means higher than average tides.

Water levels at the Battery in New York City. On Monday night, the storm tide (combination of surge and high tide) reached 13.88 feet above normal, a record. (NOAA)

As a consequence, Sandy broke a number of records, for precipitation, storm surge, and storm pressure among other indicators. The nearly 14 ft storm surge at Battery Park, New York broke a record dating back to 1821. It was more than 9 feet above the average high tide!

Sandy also broke a number of previous one-day rainfall records, at Reagan National, Dulles, Philadelphia, Atlantic City, and Philadelphia. Sandy’s storm pressure of 940 millibars (mb) was one of the lowest the region had ever seen (lower pressure generally means a more intense storm).

Is there a Connection Betweem Climate Change and Hurricane Sandy?
Until a couple of years ago, scientists were loathe to make any claims that a particular weather event could be said to be caused by climate change. The New Yorker‘s Elizabeth Kolbert echoed this trope:

As with any particular “weather-related loss event,” it’s impossible to attribute Sandy to climate change. However, it is possible to say that the storm fits the general pattern in North America, and indeed around the world, toward more extreme weather, a pattern that, increasingly, can be attributed to climate change.

Forensic Climate Science
The science has actually developed to permit researchers to now say something about individual weather events and their association with climate change. In the past few years, a sort of forensic science of disaster attribution has matured, allowing researchers to examine how much more likely a particular weather event was as a result of climate change (this kind of research has been applied to the Thai floods of 2011 and the Texas drought of last year as well as the European heat wave of 2004). However, as Climate Central’s Andrew Freedman noted, those studies can only be done after an event and take months to analyze the data and write up.

Climate Change and Hurricanes

Hurricane Sandy’s ability to develop may have been enhanced by warmer sea temperatures, in part enhanced by climate change. Here, Andrew Freedman usefully writes:

Water temperatures off the East Coast were unusually warm this summer — so much so that New England fisheries officials observed significant shifts northward in cold water fish such as cod. Sea surface temperatures off the Carolinas and Mid-Atlantic remained warm into the fall, offering an ideal energy source for Hurricane Sandy as it moved northward from the Caribbean. Typically, hurricanes cannot survive so far north during late October, since they require waters in the mid to upper 80s Fahrenheit to thrive.

Andrew Revkin notes that a recent paper in the Proceedings of the National Academy of Sciences found, using the tide gauges of the Atlantic Coast, that warms years were had more cyclone activity than cold years. He quotes the study:

We find that warm years in general were more active in all cyclone size ranges than cold years. The largest cyclones are most affected by warmer conditions and we detect a statistically significant trend in the frequency of large surge events (roughly corresponding to tropical storm size) since 1923.

Revkin, however, then downplays the potential connections between hurricanes and climate change, noting that “on longer time scales, the situation is murky because so many factors shape the formation and growth of tropical cyclones.” He then cites a number of studies that suggests that over the last 3,000 years, there have been cycles of more intense storms, suggesting that Sandy could just be a product of natural variability.

On this point, Revkin received considerable pushback from some climate advocates, who thought he was muddying the waters at a time when decisive action was needed.

Brad Plumer of the Washington Post noted how hard it was to trace the connections between climate change and hurricanes at a more aggregate level:

And yet trying to attribute specific hurricanes to changes in global temperature remains quite difficult. In its big report on natural disasters last year, the Intergovernmental Panel on Climate Change (IPCC) said it had “low confidence” that humans were currently affecting tropical cyclone patterns.

The IPCC report noted that it was “likely” that tropical cyclones in some areas would get stronger, with faster winds and heavier rainfall, as the world warmed.

Most of the research I’ve read suggested that while climate change may not affect the number of hurricanes, it may make them more intense (see page 485 in my 2008 Security Studies piece for a review of some of the evidence, Chris Mooney’s 2008 Storm World is another).

Source: NOAA

Climate Change, Storm Surge, and Sea Level Rise
But, hurricanes aren’t the only climate-related mechanism potentially operative in this case. As Freedman noted, sea-level rise is another means by which climate change may have made Sandy worse, enhancing the storm surge that accompanied Sandy:

Katharine Hayhoe, a climate researcher at Texas Tech University in Lubbock, said manmade climate change likely contributed to the storm surge at The Battery in Lower Manhattan, with 1 foot 3 inches of long-term sea level rise recorded at that location, the result of manmade sea level rise, sinking land, and ocean currents. She said the manmade contribution to the storm surge may have been a small amount.

As Plumer noted, the fact that Sandy hit at high tide in a full moon magnified the storm surge, but the fact that sea level rise had already raised sea levels also made the situation worse. Sea level rise from climate change will surely pose a greater threat in the future since most of the expected sea level rise will occur over the course of the century. As Mireya Navarro wrote in the New York Times:

With higher seas, a common storm could prove as damaging as the rare big storm or hurricane is today, scientists say. Were sea levels to rise four feet by the 2080s, for example, 34 percent of the city’s streets could lie in the flood-risk zone, compared with just 11 percent now, a 2011 study commissioned by the state said.

“Blocked” Weather/Hybrid Storms
Another means by which climate change may have affected Sandy is through a “blocked” weather pattern, which brought this storm on-land. These “blocks” or what the Center for Climate and Energy Solutions (C2ES) called a “traffic jam” may be related to Arctic sea ice loss. Andrew Freedman explained:

The upper air flow over the Atlantic Ocean was temporarily jammed by a powerful area of high pressure near Greenland and a storm system in the Central Atlantic, leaving the storm no escape route away from the U.S. Such patterns are known as “blocking” events, and they have occurred with increasing regularity and intensity in recent years. Blocking patterns have been linked to several noteworthy extreme weather events, such as the deadly 2010 Russian heat wave and Pakistan floods, the 2003 European heat wave, and the March heat wave of 2012 in the U.S.

C2ES explained the potential linkage with Arctic sea ice loss:

Recent research has shown that these blocking events and fall cold outbreaks are related to sea ice loss in the Arctic. In short, open water in the arctic helps break down the barrier between high- and mid-latitude weather, which increases the risk of cold outbreaks and blocking events. Hurricane Sandy seems to have tracked into the middle of one of these unusual meanders in the jet stream. While this is an evolving field of research and these conditions could have occurred in the absence of climate change, there is growing evidence that climate change is increasing the risk of extreme atmospheric arrangements.

Andrew Freedman has a great discussion of the potential connection to Arctic ice melt as well. However, as he concedes, the science on this point is not settled.

Kerry Emanuel, one of the leading climate scientists on the association between climate change and hurricanes, was asked by Foreign Policy‘s Blake Hounshell on the links between Sandy and climate change. He said:

I am not sure we scientists understand the link all that well. North Atlantic hurricane power has been tracking tropical Atlantic sea surface temperature for over a century of record, and we think the recent increase in Atlantic SST (and hurricane power) is related both to increasing greenhouse gases and decreasing sulfate aerosol pollution, both man-made trends.

Sandy is an example of a hybrid storm, drawing energy both from the evaporation of seawater, the source of energy for tropical hurricanes, and from horizontal temperature contrasts, the source of energy for winter storms. But we have not done a comprehensive climatology of hybrid storms, so we do not know whether there are trends in their incidence, nor have we studied how they change in climate model simulations. Thus we really have no basis for saying anything useful about their relationship to climate.

Does it matter?
I think the upshot of all this is that for pieces of the puzzle, such as sea-level rise and storm surge, we know that climate change is and will increasingly have an effect on extreme weather events. For issues like hurricanes, the “blocking” weather pattern, and hybrid storms, the connections are not yet clear. The questions remains how much certainty do we need?

It may actually be important to know with some greater certainty. There are some adaptation strategies like improved sea defenses that we may want to invest in whether or not climate change had a hand in making Sandy the Frankenstorm it was. Because of rising populations and infrastructure on the coast, these may be important investments in any case. This is certainly Roger Pielke, Jr’s argument in the Wall Street Journal (which I’ll talk more about in my next post).

But, whether or not we do climate mitigation, whether or not we decisively and expensively break our addiction to fossil fuels, may hinge on how convinced we are that climate change is not just an accomplice in the unfolding extreme weather events we are seeing but one of the major culprits. In my next post, I’ll talk about how and why the public is coming around to this view and what should be done.