Hurricane Harvey’s historic deluge in Texas showed clear signs of human-caused climate change, with rainfall at least 15 percent heavier and the likelihood of such a calamity tripled, according to two independent studies published Wednesday. The new findings are part of a surge of research suggesting that communities need to revisit their vulnerability to extreme weather in a warming world.
“Climate change made this event more likely and heavier,” said Karin van der Wiel, a researcher at the Royal Netherlands Meteorological Institute who co-wrote one of the papers and is among 25,000 people attending the fall meeting of the American Geophysical Union here.
Hurricane Harvey hit the Texas coast near Corpus Christi on Aug. 25 after it intensified rapidly in the Gulf of Mexico. The storm then stalled, as if plugged in mud, and dropped record rains for the better part of a week on Southeast Texas before finally drifting north and dissipating. The storm flooded Houston and much of the region and was one of several hurricanes that impacted the United States during a volatile 2017 season, including Hurricane Irma in Florida and Hurricane Maria in Puerto Rico.
Van der Wiel and her colleagues concluded that a deluge such as Harvey would have occurred in the region once every 2,400 years in the pre-warming period but that it is now a 1-in-800 year event — and is becoming more likely. Her research found that warming likely increased the intensity of the storm’s precipitation by approximately 15 percent.
The second study, from researchers based at the Lawrence Berkeley National Laboratory, said Harvey’s precipitation was likely boosted by at least 19 percent. And that’s just the lower bound; the team’s best estimate, which acknowledges a great deal of uncertainty, is a 38 percent increase. The report found the likelihood of a Harvey-like event had probably increased tenfold but at the very least had tripled.
“Thirty-eight percent seems pretty big,” Columbia University research professor Suzana Camargo, an expert on extreme weather, said when she looked over a poster outlining the research of Michael Wehner, senior staff scientist at the national laboratory in Berkeley. She said this kind of research points to the inadequacy of today’s flood maps, which need updating as the climate changes: “They have not been improving the maps as they should. They’re treating that as static.”
The AGU meeting is a sprawling affair in a convention center that extends for what seems like a full mile along the Mississippi River. The annual meeting is a scientific wonderland, with experts on hurricanes, earthquakes, volcanoes, climate change, Mars, Jupiter – you name it.
The meeting has numerous sessions Wednesday devoted to late-breaking research on hurricanes Harvey, Irma and Maria. Scientific research usually takes longer to cohere, but 2017 was an astonishing year of natural disasters and many people dropped what they were doing to tease out early findings about the hurricanes and other tumult, including western U.S. wildfires.
The Dutch and Berkeley research teams worked independently and used different methods – for example, examining different geographical areas, different time periods during the week that Harvey struck Texas, and framing their findings with different standards of certainty. Though their numbers are not identical, the scientists on the two teams emphasized that each study bolsters the other, with strikingly similar conclusions and lessons for the future.
“We have two independent efforts with essentially the same answer,” said Wehner. “There’s a clear human fingerprint. The numbers will undoubtedly change as more researchers look at this with different techniques, and perhaps different data sets and different methods. But our numbers are kind of big.”
He added: “We were stunned.”
“Attribution” research, as it’s known, seeks to find a climate change signal amid a weather event, which has always been problematic. There’s a truism: Climate is what you expect and weather is what you get. Weather events emerge from chaotic forces and elements, and there is variability from place to place and year to year. Scientists, and journalists reporting on weather, have typically framed their reports with caveats and cautionary notes, relying on the generalization that climate change makes events more likely but doesn’t necessarily cause them.
That’s changing.
“There is a large, new body of literature about attributing human influence on individual extreme events,” Wehner said. “It’s no longer appropriate to say scientists can’t say anything about these individual events.”
The textbooks declare that for every degree Celsius increase in atmospheric temperature there should be 6 to 8 percent more moisture in the air. That’s roughly the degree of global atmospheric warming in the past century. Wehner said he guessed, when he started his research, that Harvey might have dropped about 6 to 8 percent more rain than an identical storm would have dropped in 1950. But both the Dutch and Berkeley teams found the actual rainfall to be much higher than expected.
Van der Wiel said that indicates that there is some factor, perhaps involving the dynamics of hurricanes, that results in additional precipitation – beyond what you’d expect for the greater atmospheric moisture. As she put it, “There’s another extra thing on top of it.”
This is not the first time scientists have said an extreme weather event has a signal of climate change. Wehner said the 2010 Texas drought was an event twice as likely due to climate change. And floods in September 2013 in Colorado came after rainfall that was 30 percent heavier that should be expected, he said.
“In 2017, climate change slammed the U.S. hard,” he said. “But it’s not the first time it’s happened.”
