Last week, millions of people across the eastern coasts of the United States and Canada faced a frightening prospect: landfall of a major hurricane to rival Sandy, or perhaps even Katrina. At noon on October 1st, Hurricane Joaquin churned over Samana Cay, the largest uninhabited island in the Bahamas, and perhaps the first land glimpsed by Columbus in 1492. Maximum sustained winds of 205 kilometres/hour made the storm a category three on the five-category Saffir–Simpson Hurricane Wind Scale. Just over an hour later, it had strengthened to a category four. At the time, every major American “model” that simulated the storm’s trajectory predicted that it would make landfall anywhere between South Carolina and New York.
Not two months earlier, I had moved to Washington, D. C., to take up a position as professor of environmental history at Georgetown University. I designed and currently teach a course that explores connections between climate change and our species, from the beginnings of human evolution to the looming catastrophes of the future. On the morning of October 1st, I prepared to lead my students through a discussion about the collapse of Mayan civilization, which coincided with a series of severe droughts around the year 900 AD. At the same time, I obsessively searched for the latest updates on the trajectory and growing power of Joaquin.
I concluded - very unscientifically - that there was about a one in two chance that the storm would hit Washington directly. Given the near record heat of the waters fuelling Joaquin’s rise, I guessed that there was a smaller, but still substantial, possibility that the storm would make landfall as a major hurricane. Georgetown rests on a hill, and my home does, too. Yet most of Washington was built across the low-lying terrain of a former swamp. The storm surge that would accompany a hurricane would probably cause widespread flooding, power outages, and loss of life. Scholars have long predicted that such a flood will happen in Washington, sooner or later.
It was a surreal experience: buying survival supplies while planning a lesson on a disaster that afflicted another society, long ago, during another period of dramatic climate change. On this site, I have described the jarring psychological impact of the storm-shattered landscape, and reflected on the teachable moments that hurricanes can give us. Until October 1st, I had not grasped that moving to Washington might, sooner or later, give me an entirely new perspective on these issues. Perhaps I should have: I just reviewed Stuart Schwartz’s Sea of Storms: A History of Hurricanes in the Greater Caribbean from Columbus to Katrina. It’s a compelling book, and one that explores the environmental history of hurricanes that occasionally tear into the American coast.
That afternoon, I devoted the first half of my class to the teachable moment provided by Joaquin’s pending arrival. My students are a bright and diverse bunch, and there isn’t a history major among them. They had a great deal to say, from many fascinating perspectives, but our conversation was tinged with anxiety. Would we be back next week?
Of course, by October 2nd it was clear that Joaquin would veer east, into the Atlantic. In the great cities of the American Northeast, the threat passed with a whimper. Not so in South Carolina, which is still struggling with catastrophic floods that were fed by rain driven north by the hurricane. The people of the Bahamas suffered, too, as did the crew of the giant cargo ship, El Faro, which succumbed to winds, waves, and faulty engineering.
Even for the millions who were more fortunate, there are lessons to be learned from the storm that didn’t come.
First, it is increasingly clear that cities previously at little risk from hurricanes now face the possibility of enormous destruction from storms fuelled by unusually warm water. Such storms can happen even in climates cooler than our own, but they are certainly more likely now. Scientists have reached that conclusion by modelling relationships between oceanic warming and hurricane frequency and severity. These connections are very complex, and some scholars have argued that hurricanes, cyclones, and typhoons might actually moderate warming at tropical latitudes, while accelerating it elsewhere. Others maintain that records of hurricane landfalls do not go far enough back in time for scientists to reliably find connections between warming and the storms.
However, two recent studies successfully reconstruct the deep history of hurricane frequency using “proxy” sources that have responded to changes in hurricane activity. In August, Michael Burn and Suzanne Palmer published an article in the journal Scientific Reports that reconstructs a millennium of hurricane landfalls using lake sediments. More recently still, lead author Christian Ercolani has published an article in the journal Quaternary Science Reviews that uses overwash deposits from Naples, Florida, also across a thousand years. Burn, Palmer, Ercolani, and their colleagues all find that hurricane landfalls were significantly more common in warm climates than in chillier regimes. Scientists are still debating whether modern warming is raising the frequency of all hurricanes, or “merely” severe hurricanes. Nevertheless, it seems clear that warmer waters are raising the chances of a catastrophic storm in the Caribbean, and the likelihood that northbound storms will remain hurricanes even in high latitudes.
Second, the National Hurricane Center needs a better supercomputer, and fast. Even on October 1st, the European Centre for Medium-Range Weather Forecasts predicted that Joaquin would veer out to sea. Its forecast provided some comfort for me on October 1st. I knew that the European supercomputer behind the forecast was more powerful than the one responsible for the National Hurricane Center’s Global Forecast System model, the best available in the United States. Because the European computer is more powerful, it can predict weather with a greater geospatial resolution – or precision – than its American counterparts. This means that the Europeans are often much better at predicting even severe weather that affects the United States. It’s been that way for a while. In 2012, the European model was days ahead of its American competitors in forecasting Hurricane Sandy’s disastrous turn west. A year later, American models predicted a giant blizzard in Washington that never hit, but the European model was right.
In fact, both continental European and British models are better than their U.S. equivalents. Yet European meteorological technology and knowhow are not somehow superior. The National Hurricane Center could simply replace its current supercomputer with one more powerful than its European rivals. More importantly, American researchers could work to improve how their computers gather and assimilate meteorological data to build an accurate model of changes in the atmosphere over time.
But that would require money, and here the problem will be familiar to scholars of American disasters. It is relatively inexpensive to fund preventative or predictive programs before a disaster, which greatly diminish the costs of that disaster. It is outrageously expensive to finance recovery after a disaster, especially when that disaster was not adequately predicted or prepared for. However, in just about every society, it is politically dangerous to oppose recovery funding – as Senator Lindsey Graham is now discovering – yet politically simple to reject funding for science that superficially has no acute need.
Here on the East Coast, we were lucky to avoid another hurricane Sandy last week. Yet we received yet another reminder that a warming climate is placing us at risk, and that we need better ways of predicting that risk. Hopefully, this time we’ll listen.
Harry Enten, "Why Hurricane Joaquin Is So Hard To Forecast." FiveThirtyEight, September 30, 2015.
Jason Samenow, "What the European model ‘win’ over the American model in Joaquin means for weather forecasting." The Washington Post, October 6, 2015.
Marshall Shepherd, "Some Perspective On European Vs American Weather Models After Hurricane Joaquin." Forbes, October 3, 2015.
Nate Cohn, "Hurricane Joaquin Forecast: Why U.S. Weather Model Has Fallen Behind." The New York Times, October 2, 2015.
Christian Ercolani et al., "Intense Southwest Florida hurricane landfalls over the past 1000 years." Quaternary Science Reviews 126 (29015): 17-25.
Kerry Emanuel "The Hurricane—Climate Connection." Bulletin of the American Meteorological Society 89: ES10-ES20.
Margaret C. Stewart, "The dynamic role of social media during Hurricane #Sandy: An introduction of the STREMII model to weather the storm of the crisis lifecycle." Computers in Human Behavior, 22 July 2015.
Michael Berlemann, "Hurricane Risk, Happiness and Life Satisfaction. Some Empirical Evidence on the Indirect Effects of Natural Disasters." Jahrestagung des Vereins für Socialpolitik 2015: Ökonomische Entwicklung - Theorie und Politik, 2015.
Michael J. Burn and Suzanne E. Palmer, "Atlantic hurricane activity during the last millennium." Nature 5 (2015).