Climate science is confusing. For decades, scientists have said that more CO2 means higher temperatures, longer dry spells, and bigger storms. But ask them whether global warming caused a Midwest heatwave, the California drought, or a New York hurricane, and they'll explain ad nauseam how hard it is to untangle whether any single weather event is due to natural variation or climate change.
Hard, but not impossible. A NOAA report released November 5 looked at how (or how not) climate change affected 28 different global events in 2014. Each study attempts to quantify how much climate change affected the duration, geographic extent, and severity of the weather event in question. But some numbers were easier to crunch than others. For some historically well-recorded events—like heat waves—scientists can quickly and reliably calculate the impact of climate change. Take a rare and complex storm like a tropical cyclone ... and, well, things get complicated.
Fifty years ago, Lyndon B. Johnson received a warning about the effects of climate change, the first of many issued to US presidents. Since that day, climate study has been focused on the large-scale effects, current and predicted, of greenhouse gases. Those predictions—higher temperatures, stronger storms—are familiar now. But strangely missing was any formalized way to look at climate change in the opposite direction: to identify a particularly nasty weather event and ask whether it was caused by human emissions.
Event attribution began with a 2003 Nature paper that explored whether deductive tools from epidemiology could be applied to events like heat waves. “In epidemiology they look at things like, to what extent does cigarette smoking increase the chance of getting lung cancer," says Heidi Cullen, chief scientist at Climate Central (Disclosure: This author interned for Climate Central briefly in 2013), and co-author of one of the papers in the report. In 2004 came the first real example of an event attribution study (also published in Nature). "They found that climate change likely doubled or quadrupled the European heat wave of 2003," says Cullen.
That methodology was pretty simple. Set up two models: one of the world without fossil fuels, the other one chuggin' and burnin'. Then you run the scenario thousands, millions of times in each model, and eventually get a statistical likelihood for the heat wave happening in the world without humans versus the world as it exists today. In fact, it's so simple that some people in the field think these heatwave-climate change connections could soon be as rote as your local meteorological forecast. That makes sense: Warmer temperatures are what atmospheric greenhouse gases do best.
But drawing the connection between warmer temperatures and more complex weather events like drought, flooding, hurricanes—or even the expansion of the Antarctic ice sheet—is way more difficult. "For events like that, it's not clear that we have an established agreement in the scientific community," says Stephanie Herring, a NOAA climate scientist and editor of the report. The problem is complexity. Or maybe the problems are complexities. Anyway, it's confusing.
The key metric in these models is called fraction of attributable risk. That’s a measurement of how much stronger, higher, swirlier the thing being tested for—rain, temperature, wind—was in the emissions-heavy model. "For example, in a world without greenhouse gases a heat wave might come in at 5 degrees above normal," says Herring. "With climate change, you add an extra 1.5 degrees."
Flooding is only slightly more difficult to suss out than heat waves, because scientists have a pretty good idea of how heat and moisture interact in the atmosphere—but still, the models don't always agree. With a drought, the complexity continues to increase as variables like soil moisture, snow pack, and deforestation come into play. In these studies, authors determined that several of the droughts they looked at had no discernible links to climate change.
Tropical cyclones are the trickiest to study. This is partly because they are so complex to begin with. Also, despite their news value, cyclones are pretty rare. The satellite record only goes back to the 1970s. Prior to that, the sample size is limited to storms humans or their weather stations sat through first hand. "Because the hurricane data record is so short, there's a lot of uncertainty," says Cullen.
Considering that stronger storms are the marquee extreme weather event predicted by climate scientists, that uncertainty is a little frustrating. But revealing those blind spots is one of the goals of these studies.
Cullen is leading a project at Climate Central called World Weather Attribution to make these weather/climate connections faster and better. The ultimate goal, she says, is to have this information readily available to politicians, leaders, and city planners. That way, they can build a world that's prepared for the worst climate change can dish out—or even better, stop those man-made disasters before they happen.