The Texas floods are the latest in a disturbing pattern » Yale Climate Connections

As we reported on July 28, the terrible flash floods in Central Texas over the July 4 weekend killed at least 138 people. This marks the largest toll from a U.S. flash flood in nearly 49 years, since the Big Thompson River disaster in Colorado on July 31, 1976, during the state’s centennial weekend.

Unfortunately, this isn’t the first time in the last couple of decades that an extreme U.S. weather event has caused loss of life unseen from that particular type of weather hazard in generations. These are the types of tolls one might have wrongly assumed we’d never see again from a single hurricane, tornado, wildfire, or flash flood. So what’s going on? Is it a fact of life in our changing climate, or could warning and communication improvements reduce these death tolls? I posed this question to a roundtable of experts, but first, let’s take a look at some of these recent deadly disasters. It was 20 years ago this week that the first of these catastrophes arrived.

The deadlist hurricane in 77 years | Hurricane Katrina, 2005 

When Hurricane Katrina struck the upper Gulf Coast during the last week of August 2005, it had been many decades since a hurricane making landfall on the U.S. Gulf or Atlantic coast had killed more than 1,000 people. The Okechobee hurricane of 1928 swept from West Palm Beach into central Florida on September 17: a massive storm surge swamped communities along Lake Okechobee made up largely of migrant farm workers, killing at least 2,500 people and perhaps more than 3,000. The nation’s modern hurricane watch-warning system was established in the 1950s, and few death tolls after that point exceeded 100 until Katrina devastated southeast Louisiana and southern Mississippi.

The National Hurricane Center’s official death toll from Katrina is 1,392, although, as with many hurricanes, the indirect death toll was considerably greater (one study estimated Katrina’s indirect deaths at more than 500). Going beyond the U.S. Gulf and Atlantic coasts into U.S. territories, one independent research team’s best estimate of the direct and indirect death toll from 2017’s Hurricane Maria was 2,975, which has been accepted as the official death toll by the Puerto Rican government.

The deadliest tornado in 64 years | Joplin, Missouri, 2011

A fast-moving F5 tornado struck several communities in the Texas Panhandle and northwest Oklahoma on April 9, 1947, killing at least 184 people. A longtime ban on the U.S. Weather Bureau mentioning tornadoes had been lifted in 1938, but there was still no organized way to provide tornado warnings through radio or telephone. A pair of tornadoes that struck Oklahoma City’s Tinker Air Force Base one year later helped catalyze the modern watch-warning system.

No subsequent U.S. tornado took more than 120 lives until an EF5 twister struck Joplin, Missouri, on May 22, 2011, resulting in 158 deaths.

The deadliest wildfire in 100 years | The Camp Fire, 2018

The catastrophic Camp Fire that decimated the town of Paradise, California, on November 8, 2018, took a total of 85 lives. That made it the state’s deadliest wildland fire on record and the nation’s deadliest in a century, since the Cloquet-Moose Lake fire complex in October 1918. Dozens of fires tore across northeast Minnesota near Duluth in that 1918 disaster, which killed 453 people. It struck near the end of an era in which massive fires raged across heavily logged parts of the Midwest, and just before complete fire suppression was adopted as national policy for decades.

Wildfire swept through Lahaina, Maui, on August 8, 2023, taking even more lives than the Camp Fire did. More than 2,200 buildings were destroyed, and 102 people perished.

Some thoughts from three experts at the intersection of disasters and society

The set of deadly disasters above would be a major concern even without climate change having influenced these 21st-century threats in the following ways:

As I pondered these “deadliest in generations” events, I found myself wondering: Could it be that our warning systems – coupled with the recent omnipresence of cellphones that put weather information at our fingertips (and even push it to us) – aren’t keeping up with how people are living, traveling, and consuming (or discounting) information; how local, state, and federal officials are responding (or not responding) to official warnings and other data; and how weather disasters themselves are shifting in a human-altered climate?

As it happens, the past 20 years have also seen an explosion of research into the social science of extreme weather warnings, so I pitched my question to several leaders in this burgeoning field.

Jen Henderson, assistant professor of geography and the environment, Texas Tech University. Henderson studies how risk and uncertainty emerge in decision-making processes amid weather and climate extremes, aiming to help those left invisible in disasters to be seen by policymakers, organizations, and agencies in power.

Rebecca Morss, senior scientist, Weather Risk Analysis and Decision-Making Group, NSF National Center for Atmospheric Research. Morss studies how weather forecasting systems intersect with risk communication, decisions, and societal outcomes, with a focus on hurricanes, floods, tornadoes, and other high-impact weather.

Adam Rainear, associate professor of media and culture, West Chester University. Rainear’s research utilizes technology – such as social media, robotics, and artificial intelligence – to understand how individuals access information and communicate about risks such as weather, climate, and environmental hazards.

Each of these experts has spent many hours mulling the complex stew that can turn extreme weather events into excruciating disasters. Here are a few of their reflections, edited slightly for space.

YCC: Do you think it’s purely a coincidence that all these “deadliest in generations” events have happened in the last two decades, or do you think there might be factors in the warning/info environment for extreme weather — or in society more broadly — playing a role?

Henderson: Perhaps there’s a bit of coincidence, but mainly, I see the problem as an issue with warning information and societal issues. Warning information is often hazard-focused rather than people-focused, and our systems are biased toward individual hazards rather than compound or cascading hazards, like many hurricanes, wildfires, or floods tend to be. If there is an extreme event like a historic flood, then simply going on vacation to a cabin or camp near a river, as many people did in central Texas, exposes us to both potential weather and any inadequacies in information and warning systems.

The most important factor to my mind is vulnerability and its complexity. Those who are less resourced, who live more precariously – such as the unhoused, children, and the elderly, or those who are new or transient to a particular area – are more consistently vulnerable. There are also larger structural causes at play, whether it be system-wide inequities, historical or political contexts, technological failures, or the like. It’s difficult to point to coincidence in extreme weather events that involve deaths of individuals who live in poverty, for example, or unsafe housing, or who are in areas neglected by national and local resilience policies. 

Rainear: Our communication has gotten better and worse simultaneously. There are many important people who have evolved our risk and warning communication before, during, and after weather events. Unfortunately, alongside that, there exists a whole set of companies, pages, influencers, groups, and beyond also vying for that attention (and ultimately, money). It has severely muddied the waters and probably made it harder to find the information one desires when they are in a risky situation. AI fits here too, as it’ll just generate “more,” and that more is going to be harder to sift through to get to your desired information.

Morss: When these types of events happen, it’s often because multiple circumstances combine in such a way that the extreme weather manages to surpass the many things we’ve put into place to try to avoid them. In the Texas floods, despite decades of floodplain management, awareness based on history and education, National Weather Service forecasts and warnings, emergency management structures, etc., a combination of factors led to disastrous outcomes. Along with the extreme rain and flooding, there were points of failure in the warning and response system, combined with exacerbating factors such as the overnight timing, a holiday weekend, and summer camps in session.

The good news is that every year we see many extreme weather “close calls” where the situation could have been much worse — and would have been much worse 100 or more years ago. Yet every once in a while, an event manages to overcome our risk reduction efforts in a particular place and time, and we see a death toll that we thought (hoped) we’d never see again. Even in those stories, however, there are success stories where the outcomes could have been much worse. For example, in the recent Texas floods, there are communities nearby where the loss of life was less.

YCC: Is it possible that it’s getting tougher for people to distinguish ahead of time, or in real time, the “pretty bad” events from the “pretty close to worst case” events?

Henderson: I don’t know if it’s getting tougher, but it is tough. In an everyday context, we are living our lives, making choices about what to do and where to go, hoping that we – and our loved ones – will be safe, happy, and healthy.  While the weather is an important feature of our decisions, we have other priorities like working, caring for others, and conducting the menial activities of existence. These keep us understandably focused on our immediate concerns.

Morss: My view is that these events keep happening because extreme weather happens, and every so often, that is going to overcome the risk management that we (individuals and society) have put into place, leading to disaster. Sometimes these events are an extreme example of a threat that is already known in the area (at least to some), like the Texas floods or Hurricane Katrina. Sometimes they are a threat that was exotic or not well known in the area, like the Marshall Fire in Colorado suburbia in winter. Of course, what seems like a known or unknown risk depends on one’s perspective, because in all of these events, many of those affected said it went beyond what they could have imagined. By continuing to improve our understanding and our ability to predict and manage these risks, we can reduce how often these deadly events occur and their devastating impacts. But I’m not sure we can ever avoid them. 

YCC: Along these lines, Rebecca, in a paper for the Journal of Extreme Events with Lisa Dilling and Olga Wilhemi, you wrote:

We argue that surprise is an unavoidable component of weather and climate disasters — one that we must acknowledge, learn to anticipate, and incorporate into risk assessment and management efforts. In sum, although it may seem paradoxical, we should be learning how to expect surprise.

Morss: A few years ago, I read Isaac Cline’s description of the 1900 Galveston hurricane in Monthly Weather Review. I was struck by how horrifying his story was, but also how similar it was to more recent stories from people who find themselves in the midst of life-threatening extreme weather situations. Whether they were aware of the threat before it came upon them or not, they just could not have imagined it could get that bad. And when they find themselves facing the worst, having somehow made decisions along the way where a small change could have helped them be safe. Those situations look very different from the outside or in retrospect than they do to a person in real life and real time.

This connects to my own experience with the Marshall Fire in Colorado, where I watched myself try to catch up to a situation that I never imagined. My experience wasn’t immediately life-threatening. However, that and others’ stories have helped me understand how a person can find themselves in such a situation despite everything, and how challenging it can be.

Henderson: I have a three-month-old, and just a few weeks after he was born, we had family fly into town. This coincided with what would become a tremendous hailstorm in our town – lots of damage to roofs, siding, animals, and vegetation. In my sleep-deprived and harried state, I was worried more about nursing my baby and consoling my jealous toddler, about my visitors’ dinner and comfort, than “just another hailstorm.” But the storm turned out to be more than I expected – more than I could cognitively manage at the moment.  I (and my family) was more vulnerable than we might have been otherwise.

This wasn’t a deadly hailstorm for my neighborhood, thankfully, but the same storm did produce a brief and weak tornado just miles north of me. We were lucky that the storm weakened, that the tornado mainly damaged trees, that the hail was “only” the size of eggs and not the baseball size and larger that would hit other parts of Lubbock later that month. As the saying goes, for us, it could have been worse. Still, it was pretty bad.

Rainear: This is purely anecdotal, but I feel like social media has made folks consider risk in different ways (maybe, one would argue, being more comfortable with it because we’re exposed to so many things on there that might seem risky or scary). It feels as if, in weather risks, many are more willing to wait it out to see if things “end up as bad as they say,” which ultimately leads some to wait too long and get stuck. Maybe it’s also related to how over-warned we are with everything being “breaking news,” “alert days,” and so on and so forth that supersede what the official bodies (Storm Prediction Center, National Hurricane Center, etc.) communicate to us. Prior to social media, we used to seek out interpersonal sources after seeing something (weather on the news or something). I wonder if we don’t do that as much anymore, and social media just takes this intermediary step.

YCC: As we know, it’s tough to specify exactly how much climate change has boosted the likelihood of a specific weather event, especially a complex one. But at least some of these high-end events do share some DNA with known patterns of climate change. Any thoughts on how climate change awareness might be overlapping with how people respond to high-end weather threats?

Rainear: Being able to personally experience changes, particularly if they are subtle rather than large-scale differences, showcases how small changes can have drastic impacts. The wildfires are a great example. While wildfire risk was probably higher in many of those locations people chose to live in, the risk may be so much higher now that it could have shifted how they felt about living in that space.

Another example is farmers becoming aware of a differing pattern, such as temperatures remaining hot enough to grow longer into a season, forcing them to change their processes to best cultivate.

Alongside events that may be heightened because of climate change, there is also a heightened (one could argue even panicked) tone in communicating to the public in hopes people will pay attention. When an event is over-hyped, or when a warning is disseminated on a large scale but only affects a nuanced or rural location, that is likely having more of a negative impact on future response than awareness of any climate change connection.

Postscript: As made clear by the experts above, there’s no easy way to untangle our co-evolving worlds of climatic, communications, and societal change. Maybe a good first step in facing the specter of a “deadliest in generations” event is just to recognize how difficult a task that can be when there are only hours or mere minutes to act. That’s the case for savvy experts and seasoned weather watchers alike – much less for the average person with so much else on their plate to deal with.