Hurricane Harvey and its remnants have quickly become one of the worst natural disasters in US history. The unprecedented duration and intensity of the storm has sparked a heated debate about how much climate change is to blame. The short answer is that we don’t really know yet. But attempting to answer that question will help us prepare for the future.
Image: NASA
Many folks have rightly pointed out that Harvey is the type of extreme storm we expect to see more of in a warming world, with epic rainfall rates and surges worsened by the rising seas. It’s also clear that the impacts of Harvey were worsened by sprawling coastal development, and policies that paid little heed to flood control. It’s vital that we plan future development not only with the present-day risk in mind — the Gulf Coast is hurricane prone, after all — but the added risk that climate change will bring.
But it is also important to note that the degree to which climate change worsened this storm is unsettled — was it a little, or a lot? The relationship between climate change and tropical storms is incredibly complex, and our historical data is limited. We are just starting to draw quantitative linkages. The good news is, that means there’s room for our knowledge to grow, and for new tools such as weather attribution to help us manage future risks.
“People naturally want to know, when an [extreme weather] event has just happened to them, what really happened?” Martin Hoerling, a research meteorologist at NOAA’s Earth Systems Research Laboratory in Boulder, Colorado, told Gizmodo. “Attribution allows us to say, this is how we’re gonna fix issues in our tools and build better early warnings.”
So far, we can make some simple — but powerful — statements about the relationship between severe storms and climate change. We know that warmer sea surface and air temperatures can drive more water vapour into the atmosphere, yielding more extreme precipitation. The heaviest downpours are getting heavier in many parts of the world. And we know rising sea levels can worsen storm surges. When asked whether Harvey was made worse by climate change, several experts contacted by Gizmodo said “yes” or “probably”, citing one or both of the above factors.
Kevin Trenberth, a climate scientist at the National Center for Atmospheric Research, noted that the Gulf has been running roughly 2 degrees warmer than normal recently. Combined with the fact that it’s late summer, that means “tons of warm water to fuel storms”. Trenberth thinks Harvey was “a bit more intense, bigger and longer lasting” than it would have been in the absence of climate change. Richard Alley, a climate scientist at Penn State University, zeroed in on the impact of sea level rise on storm surge. “Mostly, this is ‘weather’ — big, dangerous weather, but still weather,” he said. “But, because of global warming the ocean is a little higher than it otherwise would be, and that made the storm surge higher.”
Other experts were reluctant to make definitive statements about the relationship between a specific extreme weather event and climate change without attribution studies. And at least one expert, climate scientist Kerry Emanuel at MIT, is unsure that warming played a significant role in this deluge. As Emanuel noted in a preliminary analysis this week, the temperature rise Texas has seen doesn’t fully account for the uptick in severe rainfall events in recent years. He suggested a recent decline in steering currents — the atmospheric currents that drive tropical storms toward the land, where they die out — might instead be to blame.
Alley also pointed out that the answer to whether climate change impacted Harvey — or any storm, really — depends on how you ask the question. Are we starting with the null hypothesis that everything is natural, and only changing our minds if we can be highly confident humans played a role? In that case “we’d probably accept that this could have been all natural”. One the other hand, if we start by recognising that the world has warmed because of us, and that warming has consequences, “then what happened here is fully consistent with us having had an influence”. Hoerling made the same point in a different way. “The fact that 40 inches [102cm] of rain fell in 1978,” during Tropical Storm Amelia, “suggests you don’t really need climate change to produce such prodigious amounts of precipitation. But maybe, in 2017, rather than 40 inches of rain we get 45.” And that isn’t a trivial distinction — as far as flooding and other human impacts go, even a few centimetres of additional rainfall can make a big difference.
One way we might arrive at some firmer, more useful answers is through weather attribution, a new field that addresses how the statistical frequency and intensity of extreme weather changes with climate change.
“Some people will say you never can identify a single cause, and that’s true,” Hoerling said. “Extreme events always bring multiple factors together at the same time. But the analysis can lead to quantification.”
Hoerling pointed out that last August, following a bout of record-setting rainfall that led to flash flooding in Baton Rouge and surrounding areas, climatologists rapidly assembled data to perform an attribution study. They found that climate change increased the likelihood of the no-name storm by 40 per cent, and increased its rainfall totals by around 10 per cent.
There are big error bars on those estimates, and there’s plenty of debate within the scientific community about how reproducible they are. Attribution is based on using models to attempt to recreate historic weather records — different models and assumptions give different answers. But many see attribution as a start toward quantifying, for instance, the increased risk of extreme rainfall events along the Gulf Coast due to warming. And when drawing up development plans for the future, that’s going more relevant to Gulf Coast cities than global averages.
“Models used by the IPCC,” an international scientific body that assembles reports on the latest climate science, “are much coarser in their resolution,” Hoerling said. “The Louisiana attribution model was running at scales of 10 to 12 kilometers,” instead of hundreds. “Knowing what caused Harvey will matter down the road as people wonder, should I be staying here, should I be building differently?”
There are still key gaps in our understanding of tropical storms and climate change that need to be filled. Will we see an uptick in tropical storm frequency, both globally and in specific ocean basins? (We haven’t, and some models project an overall decrease.) Will we see a clear uptick in the most intense tropical storms? We expect to based on most climate models, but “the trend signal has not yet had time to rise above the background variability of natural processes,” according to a recent draft section of the National Climate Assessment. Will we see more “stalled” storms like Harvey? How will the balance of forces shaping tropical storms add up?
These aren’t just esoteric academic questions. What we know — and suspect — about the relationship between extreme weather and climate change should cause alarm bells to ring. We know as our world gets hotter, more crowded, and more urban that some risks become greater. Preparing for storms was always important, and the harrowing fallout from Katrina, Sandy, and Harvey remind us it’s now more important than ever.
“Houston is already naturally at risk for hurricanes and coastal flooding,” climate scientist Katharine Hayhoe told Gizmodo. “It’s one of North America’s largest cities, and it’s located near the Gulf of Mexico, where hurricanes are common. It’s also already at risk of significant inland flooding due to heavy rain as a result of both natural and human factors,” including impervious surfaces and outdated infrastructure. “We care about a changing climate because it exacerbates the risks we already face today.”
Climate science will never be able to predict the weather. But it’s starting to contextualise our windy, rainy, messy, crowded planet, and in the future it will do so even more. It’s up to us to take heed of its insights.