Catastrophic Blizzards, Heat Waves and Floods: Global Warming or Just Crazy Weather?

Catastrophic Blizzards, Heat Waves and Floods: Global Warming or Just Crazy Weather?
Should we be talking about extreme weather events as evidence of global warming? The experts weigh in.

Officially it's not even winter yet, but freakish 60-mile-an-hour blizzards and frigid temperatures have already walloped the Pacific Northwest, while record a-foot-in-a-day snows hit the Northern Plains and Rockies. With improbable weather becoming routine, forecasters may be in for another wild ride this winter.

There's no way of knowing exactly when or where extreme cold or heavy snow is going to hit during the next three months, but the forecast does call for a 100 percent chance of someone -- most likely a Republican who wants to gut environmental regulation -- seizing on such weather as proof that the planet isn't warming.

Last February, as the snow just kept piling up in Washington, D.C., Sen. Jim Inhofe, R-Okla., and his family built an igloo down the street from the Capitol and labeled it " Al Gore's new home." And the Virginia Republican Party ran TV ads telling viewers to call legislators who supported climate-change legislation "and tell them how much global warming you get this weekend. Maybe they'll come help you shovel."

Of course, the "Snowmageddon" jokesters had no way of knowing at the time that within less than five months the snowshoe would be on the other foot. By early July, with much of North America broiling in record-shattering heat, most climate-change holdouts were keeping a low profile. (But not Inhofe, who defiantly lectured a sweating ABC News team, "We're in a cycle now that all the scientists agree is going into a cooling period.")

Meanwhile, some environmental groups pointed to the blistering July temperatures as confirmation that we've headed over the cliff of global climate change. The National Wildlife Foundation rushed out a report supplement titled "Extreme Heat in Summer 2010: A Window on the Future," filling it with pictures of sweating city-dwellers and hot-colored charts. The Natural Resources Defense Council put out a press release in which its climate-center director announced, "Welcome to what might be termed 'the dark side of climate change.'"

Back in the 1990s, the major environmental groups made a decision not to highlight extreme weather events as early signs of global warming. Should they have stuck to that policy? When heatwaves, droughts and floods are exhibited as evidence of climate disruption, does it make it that much easier for people like Inhofe to whip up more confusion the next time a winter storm hits? Is there any cool-headed way to talk about the crazy climate of recent years?

The odds on odd weather

Given the complexity of climate science, it's not surprising that the best way to get the attention of the media and the public is to talk about exceptional weather that's happening right now rather than the bigger threat of long-term climate disruption. But that makes life difficult for those who study climate for a living.

One such researcher is Katharine Hayhoe, an associate professor at Texas Tech University. She sees a clear necessity to come back hard against fatuous arguments that, she says, go something like, "Well, you know, temperatures are cooling in the month of September in Erie, Pennsylvania, so how can the planet be warming?" But, she warns, climate scientists have to be careful themselves not to go beyond the data: "It is very tempting to seize on a single dramatic event, but we have to stay true to what we know, to stick to terms like 'consistent with' and 'risk of.'"

In public statements, most climate scientists are indeed careful to stress that we cannot draw conclusions from individual extreme weather events. But Michael Mann, director of the Earth System Science Center at Penn State University, is now thinking such caution may have gone too far.

He says many of his colleagues, weary of being attacked by people like Inhofe, tend to "jump too soon," starting their responses to reporters by discounting the relevance of individual weather events. But by understating the links, he believes, they are "erring in the opposite direction," which in itself can be misleading. "Statistics show very much that these events really are part of bigger trends," he stresses.

To illustrate, Mann uses a metaphor popular among climate researchers these days: "Suppose you're betting on dice, but that someone had replaced the five on this particular die with a second six. If you don't know this, you might get cheated out of a lot of money. But when you demand your money back, can you point to any one roll of that die that you can prove lost you money? No, you can't." But, he says, it's a fact that the size of your losses is a direct result of the change in the die.

Similarly with the Earth's atmosphere, he says, statistics tell us that shifts in climate have contributed to extreme weather: "As the numbers start piling up, you can say that the numbers have been shifted by climate change. A 1,000-year event becomes a 30-year event."

To follow year-to-year climatic trends through the centuries before weather data were being recorded, scientists like Mann can use indicators like tree rings. But it is usually impossible, he says, to detect trends in shorter-term climatic extremes that occurred in the distant past.

On the other hand, daily weather data that have been recorded for 50, 100, or more years in many places can tell us a lot about extremes. And, says Mann, "When extreme events that actually occurred are the ones you'd expect based on climate-change models, you have a lot more confidence."

"If we were seeing a lot of longer, more intense cold periods, we'd all be scratching our heads. But when you confirm what the hypothesis proposed, you have an increased degree of confidence."

And confidence is increasing. Last year, 15 climate scientists published a paper in the Proceedings of the National Academy of Sciences that contained this straightforward statement, based on the most recent Assessment Report by the Intergovernmental Panel on Climate Change (IPCC): "It is now more likely than not that human activity has contributed to observed increases in heat waves, intense precipitation events, and the intensity of tropical cyclones."

Furthermore, even on a warming planet, regions with traditionally cold winters will still have plenty of below-freezing weather; when that cold air combines with moist air masses (in a generally warmer atmosphere that's able to carry more water vapor than it used to), a lot of moisture can suddenly get dumped in the form of snow.

Hayhoe helped write a June 2009 report published by the U.S. Global Change Research Program finding that the share of precipitation falling as snow rather than rain is increasing in the northeastern United States. Furthermore, said the report, "Heavy snowfall and snowstorm frequency have increased in many northern parts of the United States." Six months later, with the northern and middle Atlantic coast paralyzed by record snowfall, Hayhoe and her co-authors could be even more confident that the trend they had observed was not a mirage.

But if climate models project with some assurance that present and future emissions of greenhouse gases will lead to rapid warming of the atmosphere and extreme weather, is it even necessary to continue digging into past climate and weather records for evidence of change? Do historical studies add any useful information as we plan for the future?

Sixteen scientists who contributed to a 2008 report for the U.S. Climate Change Science Program stressed that while the past does not hold all the answers, it is important to learn what it can tell us. The first step, they explain, is "detection" -- establishing that changes have occurred in some type of extreme, say heat waves, over time. Then comes the second step, "attribution"

   Attribution further links those changes with variations in climate forcings, such as changes in greenhouse gases, solar radiation, or volcanic eruptions ... Attribution often uses quantitative comparison between climate-model simulations and observations, comparing expected changes due to physical understanding integrated in the models with those that have been observed.

Such a comparison can be seen in graphs plotted by University of Oklahoma researchers. In them, an index including several kinds of extreme climate events can be seen increasing over recent decades in a way that can't be explained by natural variation. But when greenhouse emissions are included in climate models, observation and theory align very well.

When extreme heat becomes routine

If last summer's brain-cooking heat seemed to be unusually persistent, it wasn't just your fevered imagination. Weather records show that occurrences of two-day and three-or-more-day-long runs of exceptionally high temperatures have been increasing steadily since 1960. The trend toward more extra-hot days, and especially extra-warm nights, has been strongest, as we'd expect, in urban areas where concrete and asphalt trap heat and vegetation is sparse. But rural areas and suburbs have also seen increases.

A half-century ago, record high and record low temperatures occurred at approximately the same rate in the United States. Now record highs are happening at least twice as often as record lows, and the ratio might be as high as four-to-one.

One link between heat waves and human-induced warming of the atmosphere is simply a matter of statistics. Daily temperatures are distributed like most phenomena, in a bell-shaped curve, with most readings heaped up in the middle, near the average for the date, and the rarer extremes tapering away in both directions as "tails." As the earth warms, that curve tends to shift to the right, toward higher temperatures, with its right tail leading the way.

Even if the bell curve stays exactly the same shape as it moves, a small shift can lead to many more heat waves. Notes Michael Mann, "The one-degree Celsius increase we have seen in average temperature, for example, appears to be leading to a doubling of the rate at which record-breaking temperatures occur." That happens because as the curve moves right, the "fatter" part of the tail moves into "extreme" territory. The graph below, by IPCC via the U.S. Centers for Disease Control, illustrates the effect of the shift:

heatwavecurvemean
But there may be more to the story. The monster heat wave that killed tens of thousands of Europeans in 2003 was off the charts -- impossible to explain by a simple shift in the bell curve, according to a Swiss climate team. The group reported a few months after the disaster that even considering the increase in average temperatures in Europe from 1990 to 2003 -- but assuming no change in the shape of the curve -- a heat wave like that of 2003 could be expected to occur only once every 46,000 years.

The fact that the 2003 event really did happen led them to search for other explanations in greenhouse climate models. Those models, they discovered, predict a large increase not only in average temperature but in variability as well -- a flattening of the bell curve like the one illustrated in the IPCC graph below -- that would make killer heat waves much more common in, say, Switzerland.

heatwavecurvevar
Indeed, the Swiss scientists' models suggest that in Central Europe "toward the end of the century -- under the given scenario assumptions -- about every second summer could be as warm or warmer (and as dry or drier) than 2003."

While not as wildly unpredictable as Europe '03, the 2010 killer heat wave in Russia went well beyond anything else yet experienced and might also be an indicator of a flattening bell curve.

"The heat in Russia and the floods in Pakistan in the past year were not just weather flukes," adds Mann. Greenhouse models, he points out, projected that sinking dry air would migrate from northern Africa and southern Europe toward Central Europe and Russia in summer, and that moist air would move north from the tropical Indian Ocean toward subtropical Pakistan. "Those events were part of a larger circulation pattern," he says.

Extreme rains, extraordinary snow

In recent years, precipitation patterns appear to have gone haywire, not just in Pakistan but on every continent. Katharine Hayhoe has seen this up close: "In the five years I've lived in West Texas, we've had a 111-day rainless stretch -- the longest ever recorded -- and two '100 year' rainfalls" -- ones so heavy that such an event occurs only once per century on average.

"But," she says, "it all makes sense from a basic physics perspective. The atmosphere is holding more water vapor. Storm systems, when they come, have more to work with."

Warmer air is capable of holding more water vapor than is cooler air. As a consequence, the concentration of moisture in the atmosphere also has been increasing since the '60s, both in the United States and across the globe. A comprehensive 2007 study led by Katharine Willett, now at Yale University, concluded that the increases in humidity observed planet-wide can be attributed to human influence and that natural forces alone cannot explain the change.

With more moisture in the air, an increasing proportion of precipitation is coming in the form of more intense rainstorms around the world. Over the past 30 years, the southeastern United States has seen simultaneous increases in droughts, wet years, and strong rainstorms. These big swings in precipitation are related to the continuing rise in Atlantic Ocean surface temperatures and the increasing variability of those temperatures. The ocean temperature increase has been attributed to greenhouse emissions; the degree to which emissions affect variability in surface temperatures is less well understood.

Numbers of years with extremely high snowfall totals have declined since the 1950s over much of the country. That was to be expected, because with climate change, weather cold enough for snow is now even more rare in warm regions like the Southeast.

However, extremes follow a different pattern from totals. There has been a slight upward trend in strong snowstorms over the past century in the United States. What part of that trend you see depends on where you live. Warmer areas of the country are seeing fewer big snowstorms, but the upper Midwest and the Northeast are getting hit with more of them.

Meanwhile, that most media-friendly of all extreme weather phenomena, the hurricane, is probably also the least informative when it comes to the climate debate. Authored by a group of hurricane experts, a paper published earlier this year by Nature Geoscience ( subscription) examined all existing evidence of links between greenhouse emissions and Atlantic tropical cyclones.

In contrast to the IPCC report that had concluded it is "more likely than not" that humanity's emissions have influenced tropical cyclone activity, this study found that "despite some suggestive observational studies, we cannot at this time conclusively identify" a human fingerprint on the increasing intensity of tropical cyclones. However, "a substantial human influence on future tropical cyclone activity cannot be ruled out...."

No news is bad news

Whatever happens on the ground, in the sea and in the atmosphere in coming decades, it is very likely that public discussion of the climate will tend to focus on events like heat waves, floods and storms more than on the invisible, and ultimately more important, transformation of the planet-wide climate.

In a 2007 essay, Andrew Revkin of the New York Times explained why storms make headlines but climatic disruption does "not constitute news as we know it"

   ... the incremental nature of climate research and its uncertain scenarios will continue to make the issue of global warming incompatible with the news process. Indeed, global warming remains the antithesis of what is traditionally defined as news...Journalism craves the concrete, the known, the here and now and is repelled by conditionality, distance, and the future.

But could it be that the kind of energetic public discussion we saw following last winter's Snowmageddon can actually help remedy the situation, by introducing more people to the complex forces that are taking our climate on this wild ride? Do more people now know, for example, that if there is extraordinary weather again this winter, it can be entirely consistent with what we'd expect when living in a warmer, moister atmosphere? Will more of us see in next summer's heatwaves the roll of loaded dice?

When I asked Michael Mann those questions, he chuckled. "Well, yes, I hope the past year has provided a learning opportunity" for Americans.

But will we actually learn from it? On that question, Mann -- who makes his living estimating statistical confidence -- did not seem very confident at all.