But then I saw this cover on a New York magazine, and I suddenly felt like I had permission:
 |
| Businessweek cover story: "It's Global Warming, Stupid" |
I was also surprised to read Mayor Michael Bloomberg's endorsement for the 2012 election, which put climate change front and center, thanks specifically to Hurricane Sandy.
But was Hurricane Sandy made significantly worse by unnatural global warming? On the one hand, you have people like Mark Fischetti who boldly assert "yes." On the other hand, you have people like Anthony Revkin who insist that there is too much natural variability to know.
It seems to me that we have two questions to consider before jumping to a conclusion. First, what's the relative probability of a freakish hurricane like Sandy with and without global warming? Second, is there direct evidence that global warming made this specific hurricane more destructive?
COMPARING PROBABILITIES
Probability is difficult to speak about without falling into fallacies, so it's best to spell everything out with a formal argument.
We have two variables and want to know how they relate:
G = Man-made global warming is happening
H = We experienced a freakish hurricane
Of course, neither G or H is a binary variable, but for this purpose we'll assume they both are.
Probability notation:
P(A) means "the probability of A." This is a real number from 0 to 1.
P(A|B) means "probability of A being true given that B is true."
¬A means "A is false."
1. Does global warming make a freakish hurricane more probable? In probability terms, is P(H|G) > P(H|¬G)? Translating that to English: is the probability of a freak hurricane with global warming greater than the probability of a freak hurricane without global warming?
I don't think anyone doubts that the answer is YES. See here for explanation.
2. Is the occurrence of a freakish hurricane evidence for global warming? Or, is P(G|H) > P(G)? In English: is the likelihood of global warming given the hurricane greater than the likelihood of global warming independent of the hurricane?
This is the part where we have to be careful.
Bayes' Theorem says that P(G|H) / P(G) = P(H|G) / P(H). Meanwhile P(H) = P(H|G) P(G) + P(H|¬G) P(¬G), so we can write Bayes' Theorem as P(G|H) / P(G) = P(H|G) / [ P(H|G) P(G) + P(H|¬G) P(¬G) ]. To answer our question, we want to know whether the right side of that equation greater than 1.
Anthony Revkin argues that P(H|¬G) is not negligible, i.e., that we could reasonably have a freak hurricane without global warming. You see in Bayes' Theorem that P(G|H) gets smaller as P(H|¬G) gets bigger. So if you read arguments that boost the value of P(H|¬G), you might get the impression that the answer to Question 2 is no.
But Question 2 depends on the expression P(H|G) / [ P(H|G) P(G) + P(H|¬G) P(¬G) ], not P(H|¬G) by itself. We already established that P(H|G) > P(H|¬G) in Question 1, and with that information alone the answer to Question 2 is necessarily YES. The bigger the ratio P(H|G) / P(H|¬G) > 1 answering Question 1, the bigger the ratio P(G|H) / P(G) > 1 answering Question 2.
The bottom line: Don't get distracted by discussions of how Hurricane Sandy might have happened without global warming. Global warming makes hurricanes more likely be freaks, and in turn, the fact that we witnessed a freak is further support for the existence of global warming. That's logic, folks.
LOOKING FOR DIRECT EVIDENCE
Aside from the general probability of seeing freakish hurricanes, we now have data for the specific case of Hurricane Sandy. We can study its makeup for smoking guns from global warming. There were multiple factors that made Hurricane Sandy so horrific, including an unusually warm ocean, unusually high sea level, an unlucky jet stream, and an unlucky hurricane path. You could investigate each one to see if it was not just unlucky, but unnaturally unlucky.
The evidence takes time to process. It took Nature over a year to print an article linking global warming to the 2003 heat wave in Europe that killed tens of thousands of people.
In general, scientists have been very cautious about linking individual events to global warming. When journalists ask them about a weather disaster in the news, they tend to reply in heavily hedged probability-speak. But over the past few years I've noticed that more and more experts are willing to hedge less and less.
Furthermore, even if we could never prove a direct link, listen to Bloomberg's wisdom on this issue:
The floods and fires that swept through our city left a path of destruction that will require years of recovery and rebuilding work. And in the short term, our subway system remains partially shut down, and many city residents and businesses still have no power. In just 14 months, two hurricanes have forced us to evacuate neighborhoods—something our city government had never done before. If this is a trend, it is simply not sustainable.
Our climate is changing. And while the increase in extreme weather we have experienced in New York City and around the world may or may not be the result of it, the risk that it might be—given this week’s devastation—should compel all elected leaders to take immediate action.
And don't just look at New York. Look at the drought that has been devastating 80 percent of the United States for months now. Look at the record megafloods and heat waves that have struck all over the globe over the last few years. Look at the Arctic ice cap disappearing faster than anyone expected.
The climate is going to hell before our eyes. Don't mind if I panic.
