The life and death stakes of climate adaptation
Infrastructure built for the pre-climate change world is dangerous
When I was an undergraduate civil engineering student, a practicing engineer from a Boston area firm spoke to one of my classes. He asked us this question: “What is a civil engineer’s primary job?” I don’t remember our responses, perhaps because his answer was so resoundingly correct I never forgot it: To ensure the safety of the public. The main responsibility of a civil engineer is to design structures and systems that will keep the people who rely on them safe. That Americans can generally get away with not thinking about the safety of our infrastructure speaks to how consistently engineers - and everyone else involved in building - meet this standard.1 But if we don’t rapidly upgrade our built environment in the face of rising climate impacts, this may start to change.
The strength of a manmade structure is in relation to its function and the conditions it needs to withstand, including the weather-related extremes it will face. Hypothetically, if the most rain you will ever get is, say, 6-8 inches in 24 hours, you can get away with one kind drainage system. But if you now might realistically get as much as 12 inches in 24 hours, you need a different drainage system with a significantly larger capacity and more redundancies. The same is true of road beds, bridges, power plants, landslide prevention infrastructure, wastewater treatment plants, and much, much more.
This scenario is only semi-hypothetical. Because hotter weather drives more water evaporation, and warmer air holds more water vapor than colder air, our warming planet will generate ever more powerful rainstorms until we reach net-zero greenhouse gas emissions. (As Bill McKibben memorably put it, “Warm air holds more water vapor than cold, and from that the main events of post-modern twenty first century will descend.) Drainage and water management systems need to account for this, but most were designed for pre-climate change conditions and can’t handle the volume of water produced by today’s strongest storms.
In Climate change means destabilization, not doom I described one recent example of the more intense rainfall we’re now experiencing:
For more than 37 years, from July 1967 through August 2004, the most rain that ever fell in a single hour in New York City’s Central Park was 1.58 inches. Then in September 2004, 1.76 inches fell in an hour. This record stood until August 21, 2021, when 1.94 inches of rain came down in sixty minutes. The two previous New York City records for rainfall in an hour had lasted for 37 and 17 years, respectively, but this one would last just 11 days. On September 1, 2021, over the course of one hour, Hurricane Ida dropped 3.15 inches of rain on the city. A record broken twice by 0.18 inches was obliterated by 1.21 inches. The deluge caused widespread flooding, in which eighteen New Yorkers died.
We’ve observed the inadequacy of drainage systems over and over again. We’ve seen it in New Orleans, Houston, Vermont, St. Louis, Kentucky, Tennessee, Central Massachusetts, and scores of other places in the U.S., not to mention in countless international floods.
—
When engineers design infrastructure, they consider the most extreme conditions the system or structure will need to handle and then add in a margin of safety.2 In a stable climate this approach was generally sufficient and effective. Now, however, climate change is intensifying weather dynamics, especially rainstorms, to such an extent that they semi-regularly exceed the margin of safety factored into drainage systems and other infrastructure.
Once climate change shifts the conditions infrastructure was designed to handle beyond some hard-to-pinpoint threshold, we won’t be able to confidently depend on the built environments around us. This doesn’t mean everything will fail all at once - it won’t. But it does mean we won’t know which components of our infrastructure are still safe and reliable and which aren’t. How close are we to that point? That is the crucial question leaders at all levels need to ask and try to answer.
To make this less theoretical, consider two events this summer. First, in the Libyan city of Derna on the Mediterranean coast, extremely heavy rains in September caused massive quantities of water to build up behind two dams, which we can see on this excellent Google Earth map from NBC News (I always hyperlink maps so you can click on them and be taken to their source).
After midnight local time on September 11, 2023, the larger upper dam failed, followed shortly by the lower dam. The dual failures sent a massive cascade of flood water crashing into the city, destroying hundreds of buildings and killing more than 3,000 people. This happened in the middle of the night and the enormity of the devastation is hard to grasp. In addition to those confirmed dead, as many as 10,000 of Derna’s approximately 100,000 residents are still missing, and the death toll is expected to rise. It is among the worst catastrophes in recent memory.
Derna’s context is relevant and worth understanding, as it likely contributed to these dams being poorly maintained. In the wake of the flood, widespread anger over neglected maintenance has led to the recent arrest of Derna’s mayor and several officials from its water department. The poor upkeep may have been facilitated by Libya’s political turmoil:
Libya is divided between the internationally recognized government based in Tripoli, the capital, and a separately administered region in the east, including Derna — where the main power broker is the Libyan National Army and its commander, Khalifa Hifter, a longtime militia leader.
“Libya for the past 10 years has gone through one war to another, one political crisis to another,” said Claudia Gazzini, a senior Libya analyst for the International Crisis Group. “Essentially this has meant that, for the past 10 years, there hasn’t really been much investment in the country’s infrastructure.”
Dysfunctional government and civil fracture are obviously risk factors for inadequate maintenance of infrastructure. But if we step back from the details we can see this calamity was caused by infrastructure that had performed its function for many years, but couldn’t handle the volume of water produced by a climate change-intensified rainstorm. The key lesson in this tragic story, in other words, is about the dams and the connection between their sufficiency and the safety of the public. If the dams were well maintained and designed to handle current conditions, the thousands of people who died in Derna would still be alive.
—
It’s not just politically dysfunctional countries that need to worry about water control systems. In early 2023, the Vermont chapter of the American Society of Civil Engineers (ASCE) issued a report that gave the state’s 1,000+ dams an overall grade of C, and said a third of them were in “poor” condition. A 2022 report from the state Auditor’s office reached a similar conclusion. The Auditor’s report subtitle pulls no punches: “Some High Hazard Dams, Including State-Owned, Linger in Poor Condition for Years, Risking Human Lives.”
On July 10, 2023, Vermont was hit with a powerful storm that dumped many inches of rain across wide swathes of the state. The intense rain caused flash floods and river flooding, which devastated towns and destroyed or damaged countless homes and businesses. Fortunately, the most critical dams held up under the extreme conditions. But there were instances of reservoirs overflowing and some near misses, as well as dams that were damaged by the storm. From the independent Vermont Digger:
[F]or many Vermonters, rising water levels made for a nail-bitingly anxious time. In the Wrightsville Reservoir, water rose to within a foot of its spillway, threatening to exacerbate the damage in already-flooded Montpelier. Ultimately, the reservoir did not reach the spillway, but state officials said that between three and four more inches of rainfall could have brought water pouring over the top.
…In Chester, water overflowed the Chester Reservoir dam and eroded its sides, raising fears — ultimately unrealized — that it would fail. After the flooding, officials in Williamstown also expressed anxiety that a privately owned dam could breach, sending floodwaters into town, according to the Times Argus.
After water receded, state inspectors fanned out to examine the conditions of more than 350 dams in Vermont. Inspectors found defects in at least 60 dams, which “one could presume” occurred amid the floods, according to Kamman.
Five of those dams were classified as “high” hazard, which means a “probable or certain” loss of life downstream in case of failure. Twenty-two were “significant” hazards — like the Hands Mill Dam — meaning failure could cause “major or extensive” property loss.
The precarious state of so many dams is a kind of looming, latent danger. In Vermont and elsewhere, rainstorms in the coming years will be worse than in the past. Not every community has yet experienced the full force of climate-intensified rainstorms or other extreme weather events. If you live somewhere that hasn’t yet, count yourself lucky. Eventually, every city and town will get hit by something stronger than the strongest storm of the past. And whether it’s a significant disaster like Vermont experienced - tons of property damage, months and years of clean-up ahead, and one fatality - or something a couple notches closer to the heart-wrenching cataclysm in Derna, will depend in large part on infrastructure quality.
—
In an interesting wrinkle, Vermont’s July flooding actually demonstrated the promise of upgrading infrastructure. Back in 2011, Hurricane Irene slammed into the state, causing serious flooding and destroying countless bridges, roads, and structures. In its post-Irene rebuilding, Vermont understood its new infrastructure needed to be able to withstand increasingly fierce weather and they planned and built accordingly. The flooding this summer put the newly ruggedized infrastructure to the test, and it generally performed well. From the Vermont Digger again:
When Tropical Storm Irene hit Vermont in 2011, one piece of wreckage stood out: the state’s waterlogged office complex in Waterbury, which ironically housed Vermont’s emergency operations center.
At a cost of $130 million, the largest state construction project in history, the complex was rebuilt with future floods in mind. Flood-prone buildings were removed, and those that were reconstructed were elevated to survive a 500-year-flood. It worked: State officials have reported no damage at the complex during last week’s catastrophic flooding, which once again inundated Waterbury.
“When you look at the aerial imagery of that space, you’ll see both the road and the parking lots were inundated with water — but the buildings were fine,” Jennifer Fitch, the commissioner of buildings and general services, said Tuesday.
Other than the important point that a building can only be as resilient in a storm as the roads people use to reach it, Vermont’s post-Irene rebuilding has a lot to teach us. We need to build infrastructure for the world we’ll have in 50-75 years, not the world we knew before or even the one we see today. In addition, Vermont’s experience should illustrate the imperative of upgrading our infrastructure before it gets overwhelmed and destroyed, not after.
There’s a classic liberal bumper sticker that reads, “If you think education is expensive, try ignorance.” It’s a good line, both sticky and true. A similar slogan for the climate change era might be, “There’s nothing more expensive than cheap infrastructure.”
In the climate change era, everyone in positions of leadership should be asking a critical set of questions:
What infrastructure that I’m responsible for, that affects my constituents, is most vulnerable to climate-amplified extreme weather?
How should we prioritize our infrastructure ruggedization to keep people safe?
How do we build the will and access the resources to upgrade these structures and systems before the damage happens rather than after?
This goes not just for governors and mayors, but hospital administrators, CEOs, university presidents, and others with responsibility for infrastructure.
Homeowners need to learn to think this way, too. Just because your home has functioned well and remained safe and dry in the climate of the last 20 years doesn’t automatically mean it will fare as well in the years ahead. In what ways is your home vulnerable to climate impacts? We all need to develop this climate lens for assessing our homes’ potential challenges and prioritizing ruggedization investments (roof, enhanced drainage, tree hazards, electricity, windows, etc.).
—
I’ll end on a high note by pointing out that President Biden’s Bipartisan Infrastructure Law and Inflation Reduction Act were huge steps toward ruggedizing infrastructure in the U.S. These laws are too expansive to summarize here, but just to flag one Bipartisan Infrastructure Law section among many:
The law also directs the Secretary [of Transportation] to designate 10 regional Centers of Excellence for Resilience and Adaptation and a national center to coordinate the regional centers. These centers will receive grants to advance research and development that improves the resilience of regions of the United States to natural disasters, extreme weather, and the effects of climate change on surface transportation infrastructure and infrastructure dependent on surface transportation.
Biden has jump-started the process of infrastructure ruggedization, and that’s very good news. But this undertaking is enormous and even in the best case scenario will take decades.
There’s no time to lose, public safety depends on it.
There are exceptions, of course, and their infamy is inversely proportional to their frequency.
The factor of safety used to be explicit - you would take the maximum conditions and then apply a multiplier (the number 3.1 sticks in my head, but I can’t remember why). In the more automated way engineering design works today, the factor of safety is embedded in the standards civil engineers rely on.