Plurk paste

Preface
If you are reading these words right now, you are most probably not dead. At least, not yet.
You most likely consider this to be a good thing (and so are authors and book publishers, because dead readers constitute a notoriously bad market). Being alive might not seem to be anything exceptional, as most rational, reasonable, mature adults believe that they are generally cautious, place value on safety, and would not purposely expose themselves to danger. Rational people would not drive through a red light, would not operate a hair dryer while in a bathtub, would not drink bleach, would not pet a twenty-foot-long alligator, would not walk their dog while holding a lightning rod in the middle of a thunderstorm, would not stab a horse in the butt while standing behind it, or perform other stunts that would compete for the Darwin Awards (bestowed to those who kill themselves in creatively stupid ways and, in doing so, by the virtues of natural selection, significantly improve the human race by progressively eliminating the moron-gene from its DNA pool).1 Most rational humans do take sound actions and make sound decisions when facing the risk of immediate death or—at best—death in the short term.
What is more interesting, and intriguing, is that most rational adults typically do little, or sometimes nothing at all, to protect themselves from things that could kill them years from now, and particularly from events that could suddenly and simultaneously kill a large number of people.
The world is full of such looming disasters. Yet the further away in the future that deadly event might be, or the lower its probability of occurring, the less is done. As if death a few years or decades away is not a “less imminent danger,” but rather a “nonexistent danger”—at best, a blurry possibility that might be acknowledged with the best of intentions, but not necessarily acted upon.
Why is that?
To what extent is this the case and how can this foretell the future of our civilization?
It is the purpose of this book to answer these questions.
This is done by exploring how our societies deal with lowprobability high-consequence events, by bringing together in counterintuitive (and hopefully entertaining) ways information from history, science, engineering, economics, risk, politics, and human behavior, addressing how knowledge, interpretations, reality, misconceptions, resourcefulness, and beliefs may condition certain actions, inactions, decisions, indecisions, preparedness, and responses—effective or not—to cope with these events.
This exploration proceeds by providing in-depth perspectives on the causes of disasters, on how humankind copes and learns from successive disasters, and the implications of this process on how civilization will face other future hazards and existential threats. This linkage between the past and the future can be done because, be it for natural hazards or anthropogenic ones (that is, made by humans), for current threats or existential ones, the patterns of human behavior before and after a disaster are the same. Therefore, those looking at the future with ecoanxiety, existential dread, and other fears calling them to action can gain valuable insights by reviewing how humanity has addressed or ignored other hazards.
With that mindset, a working title for this book was “Earthquakes Are Good for You—If You Survive.” “Earthquakes,” in that context, is a metaphor for a large number of life-changing events. The book’s title could just as well have been “Earthquakes, Hurricanes, Tornadoes, Floods, Volcanic Eruptions, Terrorist Attacks, Technological Catastrophes, and All Kinds of other Disasters Are Good for You—If You Survive,” but this would have required the cover page to be typeset in a font too small for marketing purposes. Alternatively—and more bluntly—it could also have been titled “Evolution by Disasters and Tombstones.” But “The Blessings of Disaster” paints a crisper picture, because, in the end, catastrophes serve as wake-up calls for survivors, who then improve things for the better. Sadly, a disaster is a necessary rite of passage to a next step along the ride up to a more robust and resilient society.
The journey through this world unfolds here in three steps.
The first part of the book focuses on hazards that have created past disasters—and that will continue to do so for the foreseeable future. Not only to understand the physics at play in each case, but also to witness how mundane it can be to live on a volcano, to work in a building behind a levee, to retire in hurricane country, to trust technology that is certain to fail, and more—until a disaster happens, of course. This is important because coexistence with disasterprone conditions is universal. To put it plainly, there are no disasters without humans. Disasters “R” Us. When it comes to the simple matter of preparing, not preparing, and coping with various hazards, human behaviors range from baffling to amazing. The second part of the book seeks to explain why John and Jane bought a bungalow straddling the San Andreas Fault, Jim and Janet retired to a beach villa that will fly away in the next hurricane, Julio and Juliet reside on the slope of Mount Vesuvius, and Jack and Jill went down the hill and built their dream home there, in a flood zone. It may not make sense, but at the same time, it makes perfect sense.
Confusing predictions and statistics, imperfect building codes, disinterested politicians that are simply mirrors, more pressing priorities, and the brain’s biases in interpreting reality and beliefs, all play a part—with potentially deadly consequences.
The last part of the book is an extrapolation of the previous observations that makes it possible to address the issues of monetary collapse, climate change, overpopulation, nuclear holocaust, and other similarly joyous topics. In other words, it seeks to answer the question: “Are we doomed?” Examples provided throughout the book span over a range of disasters. If it seems that points are made by referring to earthquakes more often than for other disasters, it is because in a world with fifty shades of disasters, earthquakes provide the perfect black-and-white contrast needed in case studies: they are sudden and devastating, but their threat is rapidly and conveniently forgotten during their long periods of quiescence.
As a word of warning, this book is not intended to be a rigorous treatise. By any scientific standards, it is not a dissertation resulting from years of research in public policy, sociology, psychology, economics, philosophy, theology, or witchcraft. According to his official bio, the author is recognized for his decades of work in disaster mitigation from the perspective of . . . earthquake engineering. Oops!
Admitting to be an engineer can be an effective repellent in a party, so this is maybe where some will close the book for good. Yet, an earthquake engineer may be good company here, because, beyond the fascination with the extreme power of earthquakes, everybody wants to be reassured that our infrastructure will not collapse and that we will not suffer or die following a big one—as if for once, we could be protected from the inescapable flaws of human nature.
For those who did not close the book, it is hoped that the pages that follow will provide an enjoyable ride—a roller coaster one that will start from the world of earthquakes and loop (roll, dive, twist, and corkscrew unpredictably) through many other similarly captivating types of disasters and domains of human nature. However, as you proceed forward, be warned that you are taking a risk. This is best explained using an end-of-chapter capper—a format that will be used throughout the book to insert (under the heading “On the Disaster Trail”) some of the author’s personal anecdotes that could be skipped by those who prefer their books written in a third-person point of view.
Finally, two realities must be acknowledged before going any further.
First, this book does not promise to change you or make you better. In fact, if anything, it is the exact thesis of this book that human nature cannot be changed easily—or, arguably, in some aspects, cannot be changed at all, barring a disaster. Those rummaging through this book in search of a life-altering experience will not find such a secret recipe (there are religions, cults, and drugs better suited for that quest). At best, this book may lead some to reflect a little bit on certain perspectives and interpretations of reality, and maybe spark a few ideas. It may lead to some introspection and some New Year resolutions, but the chances that it will change anything over the long term are slim. It takes a disaster to create the conditions leading to long-lasting changes for the better (which, again, is the thesis presented here). However, if the book can offer a glimmer of hope and show that eventually things come out right, and sometimes better, for the lucky ones who survive earthquakes and other disasters, then that is already a decent accomplishment.
Second, it is important to acknowledge that there are thousands of professionals that have been working—and continue to work—behind the scenes to improve things, such that future disasters are lesser disasters for each subsequent generation. Sometimes they succeed, sometimes not, but they are tirelessly championing the goal of achieving a safer world, and it is largely because of these invisible heroes that many people will actually survive disasters, and do so facing a lesser hardship than would have been the case otherwise. Engineers, scientists, policy makers, and many more have performed this thankless job —and are exceptionally thanked here for that. People rarely get medals or rewards for averting a disaster because disasters that have not happened are unknown. Yet safety is not magically happening by itself. Some parts of this book will briefly—superficially at best—highlight how these heroes have contributed to make the world less disaster prone, safer, and more resilient, but providing a thorough review of all these successes, and of the sustained and coordinated efforts that have been necessary to get there, is way, way, way beyond the scope of this work. As a consequence, some professionals with decades of achievement in the many disciplines that have contributed to abate the risk of disasters, or that have provided deep academic treatments to relevant subjects, may find that this small book does not fully recognize the extensive contributions of their disciplines to the subject at hand, and is thereby only a crude and overly simplified view of the bigger picture. They may feel that scholarly works on many relevant topics and subtopics have not been given due credit. These omissions, simplifications, and shortcuts will hopefully be forgiven, with the understanding that this book is not targeting a limited academic or specialized audience but rather is intended to connect with the broadest possible audience in a direct and thought-provoking way.
Therefore, when all is said and done—and read—it is the author’s sincere hope that the book will increase awareness, trigger fruitful discussions, and broaden the conversation beyond the realm of academia, to embrace all constituencies and lead to a better future. Beyond that, all other concerns that may arise might be best appeased by sharing a glass of wine.
ON THE DISASTER TRAIL
Who to Trust?
Some years back, I spent a few weeks in Taipei as part of an international, collaborative, structural engineering research project investigating the seismic performance of steel plate shear walls—which is the long and formal way to say I was having a good time doing full-scale experiments destroying huge chunks of cleverly designed steel structures with a good friend and colleague at the National Taiwan University.
One evening, back at my hotel room after a long day of testing in the lab, while busy answering emails, the hotel started to sway. “Earthquake!” shouted the little seismograph hardwired into my brain (I am getting good at recognizing those). The whole room started to move back and forth in space and the first period of vibration of the entire building was driving the response. Severely. My first reaction usually when exposed to earthquake shaking is to try to guess the magnitude of the earthquake and the distance to the epicenter (it is a stupid game and I have explained how it works elsewhere, in a different context).2 This time however, the motions kept amplifying, the sway was becoming more severe, and I realized that I was on the tenth floor of a reinforced concrete building in a country where some of those have spectacularly collapsed in past earthquakes. My first thought was: “Damn it! If it is the ‘Big One’ for Taipei, I hope the structural engineer who designed that building knew what he/she was doing.” Evidently, I survived—and largely, in this case, because it was far from being the “Big One.” However, the point of that story is that this thought that arose at the start of an earthquake should exactly match your thought right now, at the start of this book: “Damn it! I hope the structural engineer who designed this book knew what he was doing.”

MEET (SOME OF) THE HAZARDS
Which Little Pig Are You?
A MATTER OF RELATIVITY In response to a moderate earthquake that had struck a foreign country and damaged a significant number of engineered bridges and buildings, the government had dispatched a team of experts to perform earthquake reconnaissance activities. The team’s mandate was to travel across the affected region, to document the extent of the damage, to determine to the extent possible what caused the extensive damage suffered by the infrastructure, and, most importantly, to report on whether such a disaster could happen at home.
Sad but true, there is always much to learn from disasters that kill and injure thousands of people and produce billions of dollars in damage and losses. Teams of engineers have conducted earthquake reconnaissance visits after damaging earthquakes all over the globe in the past decades, so this was by itself not an unusual thing to do.
However, this time, it was different. Not only had this earthquake caused $40 billion in damage, but it did so by striking a country that considered itself a leader in the development and implementation of modern seismic design codes and standards; a nation that was part of that elite club of players whose members represent more than 60 percent of the world’s net wealth; and, most significantly, a friendly nation with whom the government had longstanding close ties. On that account, there was potentially much more to learn than usual from the earthquake damage, and the findings from the earthquake reconnaissance mission were sure to be valuable.
So, the government got its report.
What happens once a technical report is printed and submitted very much depends on the political forces at play.
Countless reports “rest in peace” on library shelves; less fortunate others are “filed” in bankers boxes buried in archives. Yet, when the stars align and the timely words fall in receptive ears, lucky reports serve their purpose and can lead to changes in building codes, enhance design specifications, or fulfill some other noble purposes mostly invisible to the public. In some rare instances, parts of these reports that have had an impact are quoted by politicians.1 For whatever the reason may have been, in one such moment, an elected government official2 deemed it appropriate to reassure the public and commented on the many bridges that collapsed during the earthquake in that other country. He emphasized that the way bridges were designed there was different than at home and that, contrary to what had been observed in that other country, the nation’s bridges were safe. “Safe!” That elected of
f icial was a representative of the Japanese
government commenting on the damage from the January 17, 1994, magnitude 6.7 Northridge earthquake that struck at 4:30 a.m. near Los Angeles and during which many bridge spans and overpasses either fell off their support or collapsed due to column failures. Most significantly, one of those collapsed spans occurred along the busiest freeway in the U.S. (Interstate 10), which, as a result, was closed to traf f ic for almost three months.3 These collapses could have been of dramatic consequence had the earthquake not happened so early that day, but instead during rush hour with cars zooming by at more than seventy miles per hour.
The Japanese official’s confident statement undoubtedly reassured the population at the time. Most unfortunately, he was wrong.
Exactly one year to the day after the Northridge earthquake, the Great Hanshin earthquake struck near Kobe at 5:46 a.m. on January 17, 1995. Beyond demonstrating that, statistically speaking, large earthquakes have a propensity to strike on a January 17 before business hours (one can really make statistics say anything), this magnitude 6.9 earthquake destroyed many bridges across the region,4 including large segments of the elevated railway for Japan’s iconic bullet train—which could have been of dramatic consequence had the earthquake not happened so early that day, but instead during rush hour with trains zooming by at more than two hundred miles per hour. Also noteworthy, the elevated Hanshin Expressway that ran though Kobe suffered massive damage over nearly its entire length, and a segment of it collapsed in such a spectacular manner that photos of it made the front page of magazines and newspapers worldwide—and, not to forget, provided sensational opening footage for the television evening news. Beyond bridges, the earthquake also damaged more than one hundred thousand buildings,5 triggered about three hundred fires that burned for days, and debilitated Kobe’s entire infrastructure. And Kobe’s port, one of the world’s busiest at the time of the earthquake, was ravaged by the earthquake and never recovered its stature.
So much for “safe.” However, that was not the worst of it. This earthquake was also embarrassing for the Japanese government for a number of other reasons. First in the list of embarrassments6 is the fact that the prime minister learned that the Kobe earthquake had occurred, not from government agencies, but rather from broadcast news. Unlike some other countries, where the president wants to be (or already is) the most powerful man/woman on earth, in Japan, the prime minister only aspires to be a good leader for his nation—sometimes acting as a manager of timid transformational forces, most often plenty happy to tend to the day-to-day business of presiding over the government. Nonetheless, it remains that the prime minister is the top decision maker in the country.
7 It is the prime minister’s office where the proverbial “buck” is supposed to stop, so from the head-of-state perspective, being the last one to know amounted to losing face—a particularly painful thing in many parts of Asia. Comparing the prime minister to a father (or mother, but Japan has not yet elected a woman to that position), it would be like having dad watch television in the den upstairs and learn on the news that a few hours earlier, somebody broke into the kids’ bedroom in the basement, thrashed the place and set the curtains on fire, while the kids huddled in their closet. In 1995, the World Wide Web did not exist, but emails, telephones, CB radios, telegraphs, and smoke signals all existed; the fact that the prime minister’s office did not receive an of
f icial notification of the earthquake and of its severity before news crews could be dispatched illustrates how local authorities were taken by surprise and the extent of their disarray following the earthquake. Part of this confusion can be explained by the fact that, in spite of evidence to the contrary, the general belief was that future large earthquakes were certain to strike soon in the Kanto area of Japan, close to the more than thirty million people that lived in and around Tokyo, but certainly not so in the Kansai area, close to the more than twenty million people that lived in the contiguous towns of Kobe, Osaka, and Kyoto (see end of chapter). Second subject of embarrassment: the post-earthquake response was aggravatingly slow. For example, it took nine hours before the military was ordered to assist, and, thereafter, army vehicles spent hours stuck in massive traf f ic jams created as the population either tried to leave or to return to Kobe, winding around town to circumvent the collapsed bridges. In many instances, when firefighters finally reached burning homes (and sometimes entire burning neighborhoods), it was often to no avail as waterlines had been ruptured. To top it off, as a matter of national pride, the government reflex was to decline international assistance until shamed to do otherwise. The US offer of a nearby aircraft carrier that could have provided a floating two-thousand-bed medical facility was declined, and Swiss dogs specially trained in post-earthquake searchand-rescue operations were held for days in quarantine at the Kansai airport.8 In fact, the government’s response was so poor that the Yakuza (which is the Japanese mafia) reportedly took it upon itself to provide food and water to residents in some neighborhoods (which is, after all, not totally surprising given that a large part of organized crime’s activities are about providing services that the government does not). Part of the problem was attributed to the fact that there was not an equivalent to the US Federal Emergency Management Agency (FEMA) in Japan at the time—although FEMA has had its problems with disasters too, as will be addressed later.
Third embarrassing matter: recovery was uneven. On the positive side, repair of bridges progressed at an impressively brisk pace; in fact, train service throughout Kobe was completely restored in four months, which is impressive considering that Kobe’s three separate train lines (JR, Hanshin, and Hankyu), in addition to the Shinkansen (bullet train), each suffered damage over more than 20 miles of elevated tracks. However, the number of gas lines failures was so large that it took approximately three months to fully restore service to all customers. Three months can be an untenable delay for those who depend on gas for heat and cooking; contrary to transportation, where there are alternate roads and/or means of transportation that can be taken, there is only one line supplying gas per residence, and no alternative. Furthermore, because of earthquake damage or fires, a lot of people became homeless overnight. In rural settings, people can literally camp in their backyards, but in one of the most densely populated countries in the world, life is far from pastoral, and a lot of people were displaced. As a gauge of the problem, seventy thousand people spent two months or more living in temporary shelters, and some temporary housing units provided by the government remained needed for over five years. Of course, some optimistic spirits have professed that the post-earthquake recovery went smoothly considering the circumstances, but had this truly been the case, maybe Kobe’s deputy mayor in charge of reconstruction would not have doused himself in kerosene and set himself on fire fourteen months after the earthquake,9 becoming in the process one of the many victims of post-disaster stress—although not all those so af f licted have set themselves ablaze.
Now, at this point, it is important to emphasize that the Japanese were not ignorant of their exposure to damaging earthquakes. Quite on the contrary, and this is best explained by a department store analogy.
In Japan’s crowded urban environment, department stores are typically multistory buildings often located at major train stations (not coincidentally, since some of the stores were originally owned by railway companies).
Although the practice has tended to disappear in recent years, some of the bigger stores still employ young “elevator girls,” dressed in the company uniform, whose purpose is to cheerfully greet customers entering the elevator, graciously thank them as they leave, and in-between call out the services provided on the floors at every stop. Going up, they would announce:
Basement: Food department—the quintessential Japanese grocery store, providing everything one may desire, from blemish-free perfectly shaped fruits (a national obsession) to live octopus.
Ground level: Cosmetics and beauty products—yes, that, by itself can fill an entire floor.
First floor: Women’s fashion.
Second floor: More women’s fashion—yes, that, by itself can fill more than an entire floor.
Third floor: Men’s fashion.
Fourth floor: Sporting goods.
Fifth floor: Home furnishing—from water purifiers to waterbeds.
Sixth floor: Kids clothes, stationery, and toy department —the only place in the world where one can buy an Ultraman figurine (not your typical superhero—readers will have to Google it to appreciate).
Top level: Restaurants of all kinds, which display, next to their entrance, plastic replicas of their main menu items —most convenient for the locals or international visitors who cannot quite decipher the Kanji, Katakana, and Hiragana symbols that combine to create the Japanese written language.
So, Japan, in a nutshell (and arguably in its collective subconscious), is a department store of disasters, where one would find:
Ground floor: Floods—for example, thousands died in the summer of 1953 when dikes along rivers failed due to heavy rain.
First floor: Landslides—another consequence of downpours, and a widespread problem given that 73 percent of Japan is covered by mountains.
Second floor: Earthquakes.
Third floor: Earthquakes—yes, that by itself can fill more than an entire floor.
Fourth floor: Fires and conflagrations—for example, 143,000 people died in Tokyo in 1923, when the fire caused by the earthquake caused more destruction and death than the earthquake itself. Note that, given the predominance of timber in Japanese residential construction, other cities there burned down on their own without the triggering effect of an earthquake, such as Hakodate, Hokkaido, in 1934.
Fifth floor: Tsunamis, which are massive waves triggered by off-shore earthquakes. It is reported that 21,959 people died during the 1828 Sanriku earthquake and tsunami that struck Tohoku, the largest to hit Japan until the 2011 Tohoku earthquake and tsunami during which fewer people died (15,896), but which is considered to be the costliest natural disaster in recorded history (some estimates reach $360 billion),10 and that created a meltdown at the Fukushima Daaichi Nuclear Power Plant complex.
Sixth floor: Typhoons and storm surges, which go handin-hand—more than 19,000 people died in the 1828 typhoon that made land-fall on Kyushu with 180-milesper-hour winds.
Seventh floor: Volcanoes—the Japanese islands are dotted with 110 active volcanoes, and while Mount Fuji near Tokyo has been dormant since 1707, a 2018 government study11 indicates that a repeat of that event would paralyze the economic capital of Japan indefinitely.
Eighth floor: Nuclear Bombs—Hiroshima and Nagasaki are the only two cities ever leveled by nuclear weapons.
On August 6, 1945, 70% of Hiroshima’s buildings were destroyed, and over 100,000 people died instantly or in the subsequent months from the effects of radiation.
Top floor: Godzilla—this is actually where the collective subconscious part comes into the picture. As world record holder for the longest continuously running movie franchise, with thirty-one Japanese full-length features since 1954, a bad actor in a latex Godzilla suit, scaled as needed to always be taller than Japan’s tallest building of the day, has trampled and destroyed countless scale models of Japanese cities, as the embodiment of the subconscious conviction that the country is perpetually on the brink of being annihilated by forces beyond its control. In short, Godzilla means: “If you live in Japan, beware that some disaster is always lurking around the corner.”
No one spending time browsing through the Japanese department store of disasters can miss the fact that disasters—including earthquakes—have consistently occurred throughout Japan’s history. Hence, prior to the 1995 Kobe earthquake, everybody in Japan knew that they lived in earthquake country. How they acted in the years prior to that earthquake very much depended on their perception of the risk of it happening soon, in their very own backyard, and of the possible consequences of that event.
For each organization and each individual, the level of preparedness very much depended on the complex juggling of relative priorities that takes place in the human psyche.
And that is not unique to Japan, but rather universal.
Which brings up the Three Little Pigs.
There have been many versions of this classic nursery rhyme, from 1853’s England to today’s YouTube (with Disney’s Silly Symphony adaptation about halfway), so everybody is presumably familiar with the story. In a nutshell, a first Little Pig builds a straw hut and the second one a house of sticks. This is expeditious and leaves plenty of time for frivolous play and to ridicule the third Little Pig who labors to build a brick home. Then, out of the blue, as if nobody saw it coming, a “Big Bad Wolf” shows up and blows away the first two huts. Depending on the age of the audience, the first two pigs either find refuge in the fortress of the compassionate third Little Pig or end up in a delicious pulled-pork sandwich. The moral of the story is that hard work pays off and—evidently, here—that the third Little Pig is a better engineer: It pays to build a more resilient structure.12 Indeed, the Three Little Pigs story is most relevant here, when dealing with extreme events and circumstances, because not all houses are created equal and some will suffer more damage than others during any disaster.
However, one of the most important points of the story, but one that is not stated even though it is at the root of countless decisions, is that if no wolf ever came, the first two Little Pigs would have won, so to speak, with more free time to enjoy life and dollars to spare—which is essentially counter to the lesson underscored by the nursery rhyme.
Likewise, when it comes to earthquakes, investments in earthquake protection measures may never actually provide any return on investment in the lifetime of the investor if no damaging earthquake occurs. The same is true for all other extreme hazards or life-impacting conditions. Therein lies the dilemma. One can invest resources and energy in hope of maybe reaping a benefit in some distant future. However, even when fully aware of the risk, betting on the probability that no disaster will occur, to spend time and money instead on things providing immediate rewards, is always an option.
And a most attractive one at that.
As such, at any point in time, depending on circumstances and timing, everybody can be any one of the three Little Pigs, which makes preventing disasters an uphill battle.
This is what is explored at length in this book. ON THE DISASTER TRAIL
Kobe or Tokyo?
After a long day spent visiting a number of research labs in the Tokyo area, some Japanese engineers and I were in a small family-owned restaurant—the type where a television is on a shelf along the wall behind the chef. We were enjoying a glass of sake while waiting for the sushi, when all were jolted.
“Earthquake. Nice!” “Hai!” they replied.
The shaking was mild and the entire room moved for only a few seconds, with a bang, as if a small truck had rammed the building. It was not scary but it is always impressive to feel an earthquake—and fun too when they are small and there is no damage.
“You haven’t seen anything yet,” they added, smiling, looking at their watch. “Minutes.” Sure enough, within three minutes, on the television, a news anchor interrupted the ongoing programming to report that a magnitude 4 earthquake had just happened, showing on a map the epicenter where it occurred and a map of the area over which it was felt. Wow. Being able to report in such “near-real-time,” in mid-1994, was state-of-the-art, and as impressive as the earthquake itself.
I mentioned to them that I was going to be in Kyoto during the first six months of 1995, with my family, and that I hoped they would also get the chance to feel such small earthquakes and live a similar experience. They laughed and replied, “There are no earthquakes in Kyoto. That is in the southern half of Japan. If you want to experience earthquakes in Japan, you have to be near the Tokyo area, in the northern half of Japan.” Of course, six month later, on January 17, 1995, the magnitude 6.9 Great Hanshin earthquake shook the entire Kansai area that includes Kobe-Osaka-Kyoto (in the southern half of Japan), killing more than five thousand people and creating over $100 billion in damage. It was the largest Japanese earthquake since the 1923 Kanto earthquake that devastated Tokyo and the surrounding prefectures.
The “Big Bad Wolf” had visited Southern Japan.

Earthquakes Happen Because . . .
THE DNA OF AN EARTHQUAKE Those who fundamentally believe that the earth is flat, or that the 1969 moon landing is an elaborate hoax filmed in a Hollywood studio, or that earthquakes are deliberate actions of God (or gods) to destroys sinful cities that have attained the “Sodom and Gomorrah” elite status, are likely to find this section offensive and can easily skip it without detrimental effects on their reading enjoyment (however, note that requests for pro-rated refunds of the book’s cover price are non-receivable).
Early on in the history of humanity, the best brains of the world were quick to figure out the mechanisms that, every now and then, unpredictably, made the solid bedrock of all civilizations shake, rattle, and roll frantically for a few terrifying seconds. When these great thinkers experienced shaking strong enough to move all frames of reference into an infernal dance and frightening enough to violate their beliefs in the immovability of the world, when the surge in adrenalin amplified the acuteness of their senses and distorted their perception of time and space, they understood it all. They were able to explain what had happened and thus re-stabilize a world that had been tilted out of balance for a moment, although unaware that various nations had established similarly accurate, competing models of how the universe triggered such tremors.
In India’s model, the Earth was supported by four gigantic elephants standing on the back of an even bigger turtle, itself standing on a cobra quite sizeable in itself.
Understandably, any twitch by any one of those pillars of the world created terrifying ground motions.
The Japanese model, of superior simplicity, understood that all of the nation’s islands rested on the back of Namazu,1 the giant catfish pinned at the bottom of the sea by a demigod holding a stone on its head. Whenever the demigod got tired—as all demigods do every now and then —the pressure released and the catfish managed to wiggle, generating all those vibrations that wreaked havoc at the surface.2 Today, in a society where cute characters that look like childish-artwork (known as Kawaii)3 are ingrained in culture and marketing, cute little catfishes are still used to symbolize earthquakes, for example in the logo of the 2020 World Conference in Earthquake Engineering4 and of the Japan Meteorological Agency Earthquake Early Warning System.5 In the latter logo, a worried (but cute) yellow catfish is broadcasting warning waves from its antenna-like (but cute) tail.6 It goes without saying that Japan and India were the more advanced civilizations. Had all of the world’s fine thinkers met at some Antiquity’s International Conferences, the Japanese and Indian delegates would have had quite a laugh at the Siberian model of the earth resting on a giant god-driven dog-sled in which earthquakes occurred when dogs scratched their fleas, as well as other equally misguided models that relied on giant frogs or dragons. Eventually, it became obvious to the sages and savants that the gods did not need fishes or dogs, and could perfectly wreak havoc on their own. Some were baby gods, in Mother Earthquake’s womb, creating earthquakes every time they kicked her tummy, as any pregnant woman knows; others were full grown gods who carried the earth in their arms or on their shoulders and had a bad day or a fit of anger every now and then.
Other earthquake-creating gods had more complex personality issues, like the Scandinavian god Loki. There, as in all Norse mythology, the narrative gets complicated: in a nutshell, for having murdered his brother, Loki was sentenced to spend the rest of eternity tied to a bunch of big rocks deep within the earth, which so annoyingly happened to be located right under a poison-dripping snake.
With each abrupt twist by Loki to avoid a falling drop of venom, came—what else—an earthquake.7 At the other end of the spectrum, for simplicity, were the super-specialists, like Rūaumoko, the Mauri god whose main job description is to create earthquakes and volcano eruptions, which he triggers by merely walking around New Zealand. Yes, that country is quite seismically active: one earthquake per Rūaumoko step to be exact.8 Eventually, some civilizations came to their senses and realized that gods were not needed to create earthquakes.
Even the Greeks—who long held that earthquakes were caused by Poseidon striking the ground with his trident— eventually accepted Aristotle’s more rational and elegant explanation that earthquakes were nothing more than the results of crazy winds trapped in underground caverns, struggling to escape.9 Yet, for most nations, moving away from the view that turtles, dogs, frogs, fishes, dragons, and annoyed gods were responsible for creating earthquakes was a hard transition— particularly when the prevailing theology wholeheartedly embraced the due rewards and punishments of divine justice. Which brings 1755 Lisbon into the fore.
At that time, the capital of Portugal was flush in gold imported from its Brazilian colony, and flush with religious orders. As ecclesiastic establishments of the era generally excelled at enriching themselves, their growing wealth attracted more devotees, and by 1750, Portugal counted more than 200,000 priests, nuns, and friars, and no less than 409 monasteries and 129 convents of 12 monastic orders (ranging from the Franciscans, with approximately 200 monasteries, to the Carthusians with only two).10 All of that while Portugal’s entire population was approximately two million people,11 and its size half of that of today’s Florida. To say that Lisbon was the capital of what considered itself to be a pious nation is, by all counts, an understatement. With Franciscan, Bernardine, Augustinian Dominican, Jesuit, Benedictine, Carmelite, Paulist Hieronimite, Loios, Trinitarian, and Carthusian orders running the show, Catholicism had a monopolistic control of the Portuguese faith.
Presumably holding the record for largest number of prayer candles burning per square miles (outside of the Vatican), to the eyes of all these devout Catholics, if a city had to be in God’s good graces, it had to be Lisbon. Or so they might have thought, until that fateful All Saints’ Day in 1755, when many of Lisbon’s quarter of a million people were in church celebrating the spiritual bridge between heaven and earth provided by all of the Church’s known and unknown saints.
Suddenly, in the Basilica de Santa Maria, chandeliers began to swing madly, the walls of the cathedral rocked, and the congregation rushed out to the streets. From the square in front of the church, where they were joined by other residents fleeing from all nearby buildings, some reported seeing the spires of the Basilica wave “like a cornfield in a breeze.” Then it stopped. Terrified, most stayed there and prayed, asking for God’s protection. The response came promptly. Shaking resumed, with more power than before. The shock was so severe that the heavy masonry façades of the Basilica and of adjacent grand buildings collapsed toward the square, burying everybody in stone rubble—or, symbolically, in piled up tombstones.
Forty minutes later, a tsunami hit Portugal’s shore; it flooded the harbor and the downtown area. To top it off, throughout all that violent shaking, some of the burning votive candles uncontrollably spilled their vows all around, igniting all nearby combustible materials, including the tapestries ubiquitous in those days to both insulate and decorate churches. In other buildings, collapsing wood floors and roofs sometimes landed in burning fireplaces, which added more fuel to the problem. From there, it grew into a conflagration that burned for three days. It is estimated that 85 percent of Lisbon’s buildings were destroyed and that 15 to 30 percent of Lisbon’s population died from the earthquake-tsunami-fire triple whammy.
12 All these deaths and damage did not make the Lisbon earthquake that different from other similar catastrophes prior. What was remarkable, rather, was its timing: smack dab into the Age of Enlightenment.
It did not matter that Catholicism focused on afterlife, with heaven and hell being the essential outcome as reward or punishment for present life, providing a meaning for the current existence. Those embracing a life of religious devotion were hard pressed to explain why a God of mercy and justice would decide, out of the blue, to unleash fire and fury on a city that was considered to be a bastion of the faith, to destroy its sacred temples, and to wipe out disciples by the thousands. Crushing them under collapsing basilicas, to make it worse. So, to some thinkers, either the faithful of Lisbon had been wrong all the time to believe in a merciful God, or earthquakes had to be caused by something else altogether that had nothing to do with God. Before the earthquake itself, in the early stages of the scientific revolution, philosophical optimists, such as Leibniz, seeking to reconcile scientific understanding and religious beliefs, had advanced that God had created the best possible of all worlds.13 They believed that the universe created by God was a Celestial Clockwork14 that had to be, maybe not perfect, but the best of all possible outcomes— the premise being that since only God can be perfect, the universe could only be “nearly perfect.” Leibniz, a worldrenowned mathematician,15 claimed to have logically proven this. As such, the philosophical optimists had painted themselves into a corner: Either God had created the best possible world, or God did not exist. As such, the Lisbon earthquake severely shook and fractured the foundations of that philosophy. The widespread destruction and death throughout the city could not possibly be understood as being the outcome of a near perfect divinely ordered clockwork, created by a God of goodness and mercy. Only an uncaring and dark force could have so harshly swept away so many Christian lives and churches—crushing the philosophical optimists in the process.
The irony of the Lisbon earthquake was not lost on the critics of religion. As one of the most highly regarded satirical polemicists at the time, Voltaire had a blast. The Lisbon earthquake led him to write the novel: Candide: or, The Optimist—his magnum opus. Not surprisingly, the book was of f icially banned by the Church, on account of religious blasphemy and political sedition, which must have made for great publicity, as it became the best-seller of 1759, and is alleged to have been the fastest selling book ever at that time.
While the contentious matters of philosophy and religion attracted all the attention then—as is often the case now— behind the scenes, a more scientifically inclined thinker looked at the 1755 Lisbon earthquake from a different perspective. Since earthquakes were not anymore, literally, acts of God, it made sense to try to understand what they are. By definition, this can effectively be considered the birth of seismology, even though the word itself did not exist at the time. Based on his studies of the Lisbon earthquake, the English geologist John Michell published in 1760 the first scientific article attempting to explain what created earthquakes.16 Like most scientific publications, this one “flew under the radar,” was not banned by the Church, and was most probably not a best seller. Evidently, it would have been a miracle of science for the first study on such a complex and difficult topic, on a phenomenon that cannot be directly observed or measured, to reach conclusions that were flawless. So, not surprisingly, he was totally wrong when he stated that earthquakes are caused by boiling water from nearby volcanoes, but it was a nice try, nonetheless. However, he hit a bullseye when he stated that earthquakes create shock waves and that, like any other traveling wave, the location where the earthquake was triggered could be determined by triangulation, knowing the time when the waves were felt at various locations—a brilliant insight given eighteenth-century technology. Note that it took almost a century after Michell—until 1846— before the words “seismology” and “epicenter” were coined by Robert Mallet.17 Mallet was an engineer who detonated underground bombs, like mini earthquakes, seeking to measure the speed of shock waves in various kinds of soils.
From there on, the most broadly known achievements of seismology were to develop: (1) the Mercalli Intensity scale to measure the severity of the earthquake by correlation to the damage it produced, and (2) the Richter Magnitude scale to measure the inherent power of an earthquake as a physical phenomenon by itself, independently of the damage it produces (more on that later). Development of the knowledge to understand what geophysical process actually produces earthquakes took much longer, and that story follows a tortuous path, because before getting there, one first had to recognize that the surface of the earth was not static—a fact that was hard to swallow. In other words, a map of planet earth purchased three billion years ago at the Galactic Hitchhiker’s store is nowadays completely obsolete, because every continent and many other parts of the earth’s surface have moved over time.18 Once upon a time, all of the earth’s continents slept together, as in a hippie commune. Africa was spooning with South America, and North America embraced both at the same time.19 This consensual intimate relationship between all partners lasted for a while, but eventually, like many similar polyamorous relationships where equilibrium is dif f icult to maintain, it did not last. Since their “break-up,” the continents have been moving away from each other, like partners from a violent divorce. In other words, not only is the earth shaking every now and then, but it does so because its continents are traveling. Some take their time, moving about half an inch per year, while others race through at four inches per year.
20 The idea that continents were once snugging together can be traced back to 1596, when the Dutch cartographer Abraham Ortelius—who is credited for having created the first modern world atlas—noticed how well South America and Africa would fit together if they were two pieces of a worldwide jigsaw puzzle. He even went as far as suggesting that they might have been torn-away by earthquakes and flood.21 History does not say if any of his contemporaries took the idea seriously, or dismissed it as the senile ramblings of a sixty-nine-year-old dinosaur in an era when life expectancy was about forty years. Irrespectively, many respected individuals reflected on this over the course of the following centuries, and most notably through the nineteenth century,22 with the geographer Antonio SniderPellegrini going as far as showing in an 1858 map how Africa and South America could have fit together. Yet, while some other geographers probably thought in similar terms, it is the thirty-two-year-old German meteorologist Alfred Lothar Wegener who apparently first coined the term “continental drift” in two papers published in 1912.
Even though a 1912 meteorologist who publishes in scientific journals is more credible—albeit sometimes less well paid23—than one who presents the weather forecasts during the evening news, in the eyes of geologists that credibility was apparently not worth much. To put it mildly, Wegener’s theory was not met with much enthusiasm. He certainly had supported his theory with evidence of matching geologic structures and fossils where South America and Africa would have been once glued together, but he had no explanation as to what force could have propelled the continents apart from each other.
Yet, over the fifty years that followed in the twentieth century, it was found that the movement of continents over time was the only possible explanation to a number of other scientific observations. For example, when rocks are formed, some of the minerals that are embedded in the rocks lock in the orientation of the earth’s magnetic field. When rocks of different ages were compared, the orientation of the magnetic field was different, so either the magnetic north pole drifted over time, or the continents did. If the magnetic north pole had drifted while all continents remained where we know them to be today, then at any given point in time, the magnetic pole location would have been recorded to be the same by all rocks on all continents. It was not the case, so only the drifting and rotation of continents in different directions could explain this discrepancy in the data.
Likewise, early studies that surveyed the deep ocean floors provided tons of evidence that the seafloor of the Atlantic Ocean was spreading in both directions from the midoceanic ridges,24 expanding the width of the ocean at those locations. Imaging techniques decades later showed that, in the Pacific Ocean, a layer of the oceanic crust was pushed under other layers, reducing the width of the ocean there. This makes sense since the circumference and diameter of the earth, as a sphere, was not observed to grow.
Rapidly, it was possible to map the boundaries of all the moving “plates” at the surface of the earth, where either “new earth” was pushed out at the bottom of the oceans, or where “old-earth” was pushed in.25 This rapidly became the plate tectonics revolution.26 Furthermore, the birth of spaceage satellite-based geodesy in the late twentieth century made it possible to measure the drift of each of these individual plates using the Global Positioning System (GPS).27 As a result, plate tectonics is nowadays a widely recognized and accepted phenomenon.
The plate tectonic revolution was also timely, as it tied together nicely with “on the ground” observations following earthquakes. This all helped explain what creates earthquakes—finally, getting to a credible explanation.
Thirty miles northwest of San Francisco, the San Andreas Fault crosses the Point Reyes National Seashore. The long and narrow Tomales Bay at the northeast edge of the National Seashore makes it clear on a map where Point Reyes is sawed away by the San Andreas Fault. This straight line cut, clearly visible from aerial photographs, is inconspicuous at ground level, particularly when the fault is inland. Hence, when an Olema Valley farmer built a picket fence, years prior to 1906, unbeknown to him was the fact that it was perpendicularly crossing the fault—and that he had built a monument that would decades later belong to the federal government. All because when the 1906 earthquake hit, the land on the west side of the San Andreas Fault moved north. The part of the fence sitting on that side of the fault came along for the ride, sheared off from the rest of the fence, and traveled a grand total of sixteen feet (a road crossing the fault further north close to Tomales Bay shifted twenty feet).28 A replica of the west and east remnants of that fence are now part of the Earthquake Trail in the National Seashore, as a reminder of the 1906 earthquake.29 Geologists knew that earthquakes occurred along faults, and the San Andreas Fault had already been identified eleven years earlier.
30 Events such as the 1906 earthquake made it possible for them to witness the slippage that developed along the fault during that earthquake and formulate theories—such as the elastic rebound theory—to explain those locally observed permanent shifts. In essence, the elastic rebound theory can be explained as follows. If the surfaces that meet at the fault line were perfectly lubricated, like two pieces of polished steel sprayed with 10W40, the relative movement between both sides of the fault would occur continuously, at a slow rate imperceptible to the human eye. However, rough rocks are not polished, but rather locked together along surfaces going miles deep in the ground and hundreds of miles horizontally. They do not slip easily. The continental plates may be stuck together at their edges (that is, at the fault line), but that does not prevent them from continuing to displace relative to each other. Like two dog owners moving away from each other while their dogs dig their paws determined to stay put together. If the dog leashes are rubber bands, as the owners move further away and progressively more energy accumulates in the stretching leashes, Fido and Fluffy are setting themselves up for quite a rebound.
The energy locked in a major fault is enormous, and when that energy is released, earthquakes are produced by the rocks along the fault grinding on each other during that rebound. The sheer magnitude of this energy makes nuclear bombs look puny. For example, the 2011 Tohoku earthquake released energy equivalent to 9.32 million megatons of TNT, which is roughly six hundred million times the energy of the “Little Boy” nuclear bomb that exploded over Hiroshima in 1945.31 To be clear, the plan hatched by the zany sociopathic villain in one of the worst-ever James Bond movies, which consisted of blowing up a single bomb inside the San Andreas Fault to wipe out Silicon Valley, was ludicrous, a severe case of bad math, and bad movie making.
It takes time to build up all that energy along a fault before it releases. If a fault always slipped twenty feet during each earthquake, and the relative movements between two plates were recorded to be one inch per year, this would imply that there would be an earthquake every 240 years. Unfortunately, there are at least two reasons why this will not happen as simply as that. First, seismic faults are not Swiss Cuckoo clock parades of wooden characters recurring at perfectly regular interval. Across a segment of fault miles deep and hundreds of miles long, rocks not only lock together in a random fashion, but there is no way to see how they are locked together. Second, humans have only been recording useful information about specific ground motions for a few hundred years—or less in some cases.
Earthquakes that have happened before human settlements are not forever lost though. Seismologists have dug trenches across faults in search of geological evidence of prehistorical earthquakes in the rock layers, which has made it possible to determine that the San Andreas Fault near San Francisco has ruptured at an average interval of two hundred to three hundred years.32 Over a million years, the fact that the San Andreas Fault moves at a relative rate of roughly one inch per year33 adds up to a total displacement of one million inches, but in fits and starts. At that rate, given that the fault slices California north to south, all the way down to Southern California, Los Angeles will become a suburb of San Francisco in only twenty-four million years. By the way, it is always great fun to see the bewilderment of those who watch an accelerated video animation of that phenomenon and think that it will happen in twenty-four years because they fail to see the “x106” in small print next to the numbers on the counter (which is a mathematical shortcut to say 24,000,000).
Nonetheless, while everybody can visit the sixteen-foot offset in the fence at Point Reyes, there is no way to know if the 1906 earthquake released all of its locked strains, or only some of it, leaving a lot of it temporarily locked and accumulating more stresses for another imminent release of energy. This means that the next “Big One” there could be anytime (and, to make it more exciting for San Franciscans, keep in mind that the city is surrounded by other equally dangerous faults expected to rupture sooner).34 All of that is to say that, bottom line, earthquakes have happened for as long as the earth has existed, and most probably will continue to do so for as long as it will exist.
This process cannot be stopped.
Might as well learn how to live with them.
CAN’T ESCAPE THEM IS BAD; CAN’T PREDICT THEM IS WORSE DISCLAIMER: Some scientists underscore that there exists an important difference between a prediction and a forecast. Some argue that a prediction must define at what specific time and location an event will occur, whereas a forecast is expressed as probabilities over a short time window.
35 Others argue that a forecast relies on calculations using past and present data to extrapolate into the future, whereas a prediction can be based on facts, judgment, experience, opinion, or a crystal ball.
36 As interesting a nuance as this may be, to the general public (and throughout this book), the two words are used interchangeably and context will be suf
f icient to determine
whether the words refer to hours of calculations by an egghead using complex models running on a cluster of supercomputers, or an opinion from Grandpa after snif
f ing
the air for a few seconds. Anyone who believes that scientists will soon be able to provide a warning of exactly when and where an earthquake of a specific size will strike has been watching “way too many” bad Hollywood disaster movies. Barring an explosion of amazing scientific or engineering breakthroughs (each of the E = mc2 jaw-dropping kind), this is not likely to happen in this lifetime. Pessimists will argue that funding research focused on short-term earthquake prediction is like throwing dollars into a money pit, with little expectation of return on investment. Hard-core pessimists will add that it is more of a black hole than a money pit, sucking resources into oblivion.
Optimists (and most definitely those who thrive on this kind of research funding) will argue that breakthroughs are imminent and that the benefits that society will reap by being able to make such predictions will far outweigh the costs of the research. Whether or not such predictions really can be of any value will be examined later, but it remains true that the inability to make short-term credible predictions has been frustrating to seismologists, particularly because it often seems to be the only thing that people want to know about earthquakes.37 In the long history of failed earthquake predictions, the “Parkfield Experiment” is a special example of optimism blown to pieces. Parkfield is a world famous, minuscule village located near the middle of the 750 mile-long San Andreas fault with a population of eighteen persons (in 2007).38 World famous in the seismological community that is. It happens to be located along a part of the San Andreas Fault that creeps at a constant rate of roughly an inch per year.
39 In fact, the ends of a bridge built in 1936 across the fault40 have continuously moved laterally with respect to each other, bending the bridge by more than five feet since.41 Moderate earthquakes of roughly magnitude 6 are known to have occurred there previously in 1857, 1881, 1901, 1922, 1934, and 1966. At some point in the early 1980s, someone brought up the fact that not only did this correspond to an average of twenty-two years, but it also seemed to happen like clockwork about every twenty-two years—as long as the geological clock can be forgiven for not having Swiss-watch accuracy, because the 1934 event clearly fell out of line by a substantial margin. Nonetheless, this unusual regularity in observed seismicity, coupled with the fact that seismographic records (available only for the last two in that series) showed identical epicenter locations along the fault, was alluring. If Mother Nature was to throw a wild earthquake party in Parkfield twenty-two years after 1966, seismologists surely did not want to miss it—as far as an earthquake can be an exhilarating proxy for sex and booze when it comes to a seismological wild party. As such, in 1985, the US Geological Survey (USGS) issued a bold but of f icial prediction that the next magnitude 6 earthquake in Parkfield would occur in 1988—the very first such prediction by the USGS. The USGS is a serious agency of the federal government whose mandate is to “provide science about the natural hazards that threaten lives and livelihoods.”42 It not only published this prediction in Science magazine,43 but also of f icially informed the governor of California—evidently catching the media’s immediate attention. Wisely recognizing that the creeping motion of the San Andreas Fault did not operate as perfect clockwork, it qualified its prediction by stating that there was a 95 percent chance that a magnitude 6 earthquake would occur there in the tenyear window between 1983 and 1993. It was a fairly safe bet to make, since it called for a relatively small earthquake in an area where only a few ranchers lived, thus avoiding the risk of a panic and economic depression due to fleeing residents—in case anyone there trusted the government’s prediction. Given that the 1985 prediction was already encroaching, two years into that ten-year window, $20 million44 were spent in a rush to “pepper” the Parkfield area with instruments, such as seismometers, strain-meters, creepmeters, GPS receivers, and other state-of-the-art gadgets to record everything physical and chemical that could occur in rocks, soil, and ground water, during and after the event.45 Then, 1988 came and went without any magnitude 6 earthquake occurring in Parkfield. So did 1989, 1990, 1991, 1992, and, finally, closing the window with a big disappointment, 1993. In short, Parkfield could have been to seismologists what Woodstock was to hippies, if the main attraction had actually showed up to perform—but it did not.
A magnitude 6 earthquake eventually occurred in Parkfield in 2004, which prompted some to victoriously proclaim that the predicted earthquake had occurred after all, “as predicted,” only a bit later than originally foreseen.46 Those likely to find solace in that argument are probably the same ones who would not mind if the groom or bride showed up sixteen years late to their wedding, expecting guests to have waited in the aisles.
As a consolation prize, all those still dreaming of an earthquake-Woodstock can travel to the Parkfield café, rated by TripAdvisor as the best (and only) restaurant in Parkfield,47 to take a selfie with the café’s big outdoor steel water tank where is inscribed in vivid letters: “Earthquake Capitol [sic] of the World—Be Here When It Happens.” Admittedly, the best and the brightest might have been seduced by the simplicity of the measured constant creeprate along the fault and the perception of a steady return period in Parkfield. Critics have described the assumptions underlying the prediction as simplistic and possibly politically motivated.48 In fact, Parkfield is nothing but one more unsuccessful experiment. Actually, hundreds of millions of dollars have been spent by various government agencies and research funding agencies throughout the twentieth century in support of research on earthquake prediction—with, arguably, nothing to show for it.49 This was often done in spite of repeated testimony by engineers that it would be more pragmatic instead to fund research on how to design buildings to perform better during earthquakes.
Funding for research on earthquake prediction peaked in the 1960s and 70s, and proponents of this research used a battery of scientific approaches and measurement techniques to prop their optimism in this wild goose chase.
Many felt they had to catch up to the Russians and most particularly the Chinese who claimed great successes in earthquake prediction.
Indeed, in the mid-1970s, the Chinese government claimed that they had nailed the science of earthquake prediction. In one such success story, they had issued an of f icial warning at 10 a.m. on February 4, 1975, predicting that an earthquake was to occur soon in the Liaoning Province.50 The fact that this is a large province covering 56,332 square miles—being more than 350 miles at its widest point51—somehow got lost in the official narrative.
When a magnitude 7.3 earthquake actually occurred nine hours later, near Haicheng, located right in the middle of that province, wild claims were made to the effect that the scientific foresight precisely had pinpointed the location of the earthquake ahead of time. It was also claimed that the government accordingly had ordered the city to be evacuated, but no such evidence has ever been produced.
In any case, the earthquake killed 1,328 people in Haicheng —hardly a success story by any standard. The truth of the matter, though, is that hundreds of foreshocks occurred over a three-day period before the large devastating earthquake hit, including hundreds of shocks in the early hours of February 4—foreshocks being small earthquakes that damage nothing but usually scare people enough to make them sleep outside out of precaution, particularly when foreshocks come in swarms. Therefore, while it is true that the Chinese seismologists made a successful prediction, anybody with an IQ above 50 might have also successfully called this one.
The problem is that most large earthquakes are more sneaky than that and do not provide the luxury of foreshocks. That was the case for the tragedy of the Tangshan earthquake that hit on July 28, 1976. In spite of the Chinese claims (emboldened by the “success” of the Haicheng prediction eighteen months earlier) that seismologists there had nailed the science of earthquake prediction by monitoring water levels in wells, concentration of radon and other chemicals in groundwater, inclination of the land using tilt meters, changes in magnetic fields, temperature anomalies, and changes in the behavior of dozens of species of animals, the plain fact remains that nobody—absolutely nobody—foresaw the Tangshan earthquake. By the official count of the Chinese government, it killed two hundred forty thousand people, making it one of the deadliest earthquakes in recorded history. Although some unofficial estimates report six hundred fifty-five thousand victims, the true number will likely never be known because one plus one equals whatever the government needs in some regimes.
In an entire textbook devoted to the topic of earthquake prediction written by the German-Chilean-Mexican renowned seismologist Cinna Lomnitz,52 each scientific method that has once been considered valid or promising for earthquake prediction is reviewed, explained, and—more importantly—debunked. As stated loud and clear in the book’s concluding chapter, earthquake predictions are about as successful as predictions of economic recessions or the outcome of elections. Simply put, earthquakes cannot be predicted.
Today, posted on the USGS website is the definitive statement: “Neither the USGS nor any other scientists have ever predicted a major earthquake. We do not know how, and we do not expect to know how any time in the foreseeable future. USGS scientists can only calculate the probability that a significant earthquake will occur in a specific area within a certain number of years.” On that last point, the USGS typically does so to provide the values (and maps) of seismic parameters used for the design of buildings and other infrastructure, using statistical modeling to forecast over a fifty-year period.
In summary, none of the prediction methods proposed in the past work. Still, facts have never prevented colorful characters from making claims to the contrary.
QUACKS AND CLOWNS, OR GENIUSES?
As the geologist Jim Berkland used to say, “It’s really very simple to predict earthquakes, and anyone could do it. The hard part is being right.”53 Following the principle that most broken clocks can give the right time twice per day, if one predicts earthquakes often enough, a successful prediction is likely to occur at some point.54 Since people are inclined to remember only successes, the red carpet is usually ready for charlatans, pseudo-scientists, prophets, and attention seekers.
Sometimes it is all about packaging. For example, there is apparently a guru out there advertising his services to expecting parents, claiming that he has the mind power to control the universe such that they will get a kid of the gender of their choice. In fact, he is so confident in his special “skill” that should his intervention fail in their specific case, he will fully refund them his $5,000 fee. With such an ironclad money back guarantee, what is there to lose? Inevitably, the guru is bound to be right half the time, and thus to keep his fee half the time, which makes for a pretty lucrative business. Earthquake prediction by quacks, clowns, and prophets can be somewhat similar in concept, except that it is not always clear if there is any monetary gain to be made in this case. In this era of social media, some have been ingenious in turning prediction websites into dollars, but most seem motivated by love of the limelight or by an eagerness to save lives using their sixth sense, extrasensory perception, psychic gifts, or superpowers. Many have also been skillful in getting free publicity by attracting media attention.
In a world where “infotainment”55 is king and “information” is not, clicks and likes and eyeballs matters more than substance. What does it say when all international media outlets considered it newsworthy to report the story of a ten-year old grilled cheese sandwich that sold for $28,000 on eBay because the burn patterns on its toasted bread purportedly showed the image of the Virgin Mary?56 Although some argued that the face looked more like Greta Garbo or Marlene Dietrich than the Virgin Mary,57 it made perfect sense to hard-core believers that only a miracle from Jesus’s mother could keep a ten-year old sandwich free of mold. Some have ventured scientific explanations as to why a ten-year old grilled cheese can be mold-free,58 some have taken advantage of the frenzy to explain the brain’s tendency to formulate familiar images in clouds, ink blots, or other random patterns (a phenomenon called pareidolia),59 and others have simply called it a hoax.60 Not to forget those who took advantage of the news coverage to advertise toasters that can burn a perfect image of Jesus on the morning daily bread.61 If that entire circus is considered legitimate news, it then stands to reason that any atypical or free-spirited earthquake prediction will get its fifteen minutes of fame.
Examples include the prediction by a self-proclaimed Dutch “quake-mystic” that an 8.8+ magnitude earthquake was to strike California on May 28, 2015 (it did not happen),62 and (broadening the horizons a bit) that a “megaquake” of magnitude 8 or greater was to occur somewhere on the planet between December 21 and 25, 2018 (it did not happen), and that a magnitude 7 earthquake was to occur on the planet on April 12, 2019 (it did not happen either).63 Considering that at least one magnitude 8, eighteen magnitude 7, and 150 magnitude 6 earthquakes occur each year on the planet,64 lowering the size of the predicted earthquake with each prediction is an excellent strategy to eventually hit the jackpot. Another brilliant strategy would consist of enlisting 365 Nostradamus-minded friends to each pick a day of the year and post on Facebook their prediction of a magnitude 8 earthquake somewhere on the planet on that day; one of them is sure to be crowned a genius by all media when the yearly magnitude 8 happens. Even more so if it happens near a populated area rather than in the middle of nowhere as is more frequently the case.
Also, since making a prediction on a hunch costs nothing, might as well aim high: a Pakistani visionary with a “sixth sense” alerted the governments of seven Asian countries on November 2017 that a massive tsunami was expected within two months in the Indian Ocean (it did not happen).65 This benevolent warning forced all the emergency response agencies in those countries to be on alert just in case—not because anyone believed it, but rather because nobody wanted to be blamed if by sheer coincidence such a tsunami did happen that month.
What matters for a prediction method to be credible is repeated successes under close, independent, expert scrutiny, and predictions by “free-spirited theorists” have not scored well on that account. Predicting earthquakes by looking at the shape of clouds, lights, electric signals,66 vapors, snowpacks, aching bunions, and other unrelated phenomena has happened consistently over the years and has—not surprisingly—been consistently ignored by the scientific community.
67 It is amazing (and possibly disturbing) that brash forecasts based on nothing more than hunches or pseudoscience receive massive media attention and have even prompted many to stock-up on supplies and brace themselves for the upcoming jolts, while forecasts anchored in real science often fail to generate excitement—or often any response at all. Or maybe it is not so surprising, given that some people pay fortune tellers to “read” their future in tarot cards (cartomancy), lines in the palms of their hands (chiromancy), crystal balls (crystallomancy), shapes of clouds (nephelomancy), numbers (numerology), tea leaves (tasseomancy), and even the wrinkles, cracks, bruises and dimples on butt cheeks (rumpology).68 For instance, in 1968, clairvoyant Elizabeth Steen predicted that a giant earthquake would destroy San Francisco that year, and moved with her family to Spokane, Washington69—followed by thirty-five other families equally moved by her mystical prophecy. Evidently, nothing happened in 1968, but others built on that prediction after the fact. Combining it with an earlier prediction by another “prophet” and a fictional account in a book of California falling into the sea in 1969, some people became convinced that the fateful event would occur in April 1969. Arguably, these were the “groovy” sixties, when hard drugs facilitated prophecies and conspiracy theories were plentiful—after all, if the government lied about JFK’s assassination, the moon landing, and flying saucers in Area 51, why not earthquakes? This might explain why hundreds left the state, sometimes incited by preachers leading the flock to safety, in spite of all the public announcements and newspaper articles attempting to counter the rumor by providing factual information. Again, nothing cataclysmic happened—San Francisco made it through the year.
There is, in principle, no limit to the possible escalation of media madness and nonsense driven by a “precise” earthquake prediction. This was well illustrated by the dozens of national and international media news crews that drove satellite and radio vans to New Madrid, Missouri, on December 3, 1990, waiting for the earthquake predicted by Dr. Iben Browning to happen. Dr. Browning’s PhD was in the field of zoology, with minors in genetics and bacteriology.
70 He became a business consultant in various scientific fields.
His prediction of a magnitude 6.5 to 7.5 earthquake to strike in New Madrid on December 3 was based on the expectation that the alignment of the Moon and Sun with the Earth on that date, combined with a few other astronomical factors, would create record tides last seen 179 years ago. Given that a series of large earthquakes of up to magnitude 8 had occurred in the New Madrid area from December 16, 1811, to February 7, 1812 (that is, 179 years before 1990), it apparently all added up. To put things in perspective, Browning predicted that bridges across the Mississippi River would collapse and that a third of the buildings in Chicago would suffer damage from the upcoming earthquake. No matter how many experts discredited the prediction, nor how many scientists explained that none of it made sense, media could always find someone to make comments that left enough uncertainty in the air.
Prior to the D-day, members of emergency response agencies in Missouri and nearby states attended seminars to learn about earthquakes and made sure all their equipment was ready to deploy. Many of these agencies conducted earthquake drills and prepared emergency shelters equipped with food, water, and medical supplies.
More than a thousand emergency responders participated in a mock disaster drill in St. Louis. The Arkansas National Guard performed a drill for a scenario that assumed a magnitude 7.6 earthquake with 4,950 dead, 25,097 seriously injured, and 98,020 left homeless. At least sixteen hundred members of the Kentucky National Guard were on call. Leaflets on how to mitigate losses from earthquakes were widely distributed.
Schools around the New Madrid area in Arkansas, Illinois, Indiana, Missouri, and Tennessee closed on December 3, some for three whole days just to be safe. Some businesses closed, including an aluminum plant with fifteen hundred employees. Some people decided it was best to leave town for that day. Christmas parades and seasonal concerts were canceled. Religious activities abounded, and a preacher used a van equipped with megaphones to preach about— why not?—the end of the world. Of course, for many people, December 3 turned out to be a great outdoor event, with rock bands, “quake-burgers,” “Shake Rattle and Roll Tshirts,” and all the stuff typically found in a tailgating party or a holiday celebration.
As expected, December 3 passed, the earthquake did not happen, and the news crews went back home. All that remains from that temporary insanity is some old T-shirts, maybe some survival toolkits, and a fascinating USGS report that was produced in an attempt to understand how such an unfounded prediction attracted so much attention.71 The report contains nearly two hundred pages of press clipping (as a small but representative fraction of the total number of news articles published) that “built-up” the frenzy by promoting the prediction and informing people on what actions to take to prepare, survive, and live after the earthquake. The report concluded that, as one of its many findings, “what began as an interesting story, but one not given particular credibility, evolved into a contagion that promoted Browning’s prediction. Many educated and otherwise sensible people were caught up in believing that perhaps Browning was right. New York’s finest ad agency could not have done it better.” Once all is said and done, there is only one method to predict earthquakes that is 100 percent infallible. It is known as the Richter Law of earthquake prediction—from the same Richter who invented the earthquake Richter scale and the seismograph, no less. This best-kept secret states that every earthquake takes place within three months of an equinox,72 and to this day, no earthquake has ever occurred outside of this range (it usually takes a few seconds to do the math on that one).
More seriously, the only possible way to predict the future occurrence of earthquakes is to review information from all sound scientific bases— fault-slip rates, average return periods of past earthquakes, and other seismological data—and provide a probabilistic assessment. Notably, for the San Francisco–Oakland–San Jose Bay Area, a region crisscrossed by multiple faults that have been extensively studied for decades by the best minds in the field, the of f icial prediction is that there is a 72 percent chance for one or more earthquakes of magnitude greater than 6.7 to occur between 2014 and 2043.73 That is informative in some quantitative way, and hopefully helpful to incite people to prepare, but it is probably no more useful than if the Three Little Pigs were told that there is a 72 percent chance that the Big Bad Wolf will show up some time in the next thirty years.
ON THE DISASTER TRAIL
Make It Stop Three days after the Christchurch earthquake in New Zealand, I was awakened at 3 a.m. by a rather strong aftershock. There had been a few more during the night strong enough for me to open an eye, think “aftershock,” and fall back to sleep, but that one was much stronger. It also lasted longer. It said, “are you sure it was a good idea to fly down here, dummy?” I got out of bed and walked to the window. Across the street, the masonry building that had collapsed the day of the big shock was there, in the dark, as a grim reminder of what happens to those poorly equipped to face the forces of nature. I remembered that a good friend once told me, “Some people chase butterflies, others chase earthquakes.” Not quite. You cannot drive around to “catch” an earthquake when it happens, like the storm chasers who follow dark clouds to catch tornados, but you can walk through the rubble it creates. Actually, more importantly, you can walk in the buildings still standing up to diagnose why they suffered specific types of damage. In research labs, we spend hundreds of thousands of dollars to tests specimens that are parts and pieces of buildings—or even to test scale models of entire buildings on shaking tables—in well controlled experiments, to investigate how things will behave during future earthquakes. Here, nature gave us all the results of many full-scale experiments, leaving damaged “specimens” everywhere. All that is needed is to figure out what the actual properties of the tested specimens were—as a sort of “reverse engineering” problem.
Staring at the collapsed building in the dark, I remembered another evening, almost twenty years earlier, in Los Angeles following the Northridge earthquake, in another hotel whose architecture suggested that it was built with a decent amount of reinforced concrete walls and was therefore likely to be safe for a few nights during a postearthquake visit. In the penthouse restaurant of that other hotel, an aftershock started to shake the room. A woman at a table nearby started to shout: “Not again! Make it stop, make it stop!” She was quite frazzled. She might have been one of the twenty thousand people whose house got damaged, forcing her to stay in a hotel. There had been more than $20 billion worth of damage to single-family homes north of Los Angeles, in Northridge—only half of it insured.74 Here in Christchurch, thousands of homes were similarly damaged (the newspapers mentioned one hundred thousand homes needing repairs, ten thousand to be demolished and rebuilt),75 but most of it was insured to some degree.
Looking at the collapsed masonry building across the street. Its damage. So predictable. Everybody knows how unreinforced masonry buildings collapse during earthquakes. Not much new to learn there. So predictable.
As the saying goes, it is not a matter of “if,” but only a matter of “when.” The big “when” question for which there might never be an answer.
That was when, lost in thought, I remembered the expert from NASA whom I had met fifteen years earlier. Without doubt, a credible scientist. I had completely forgotten about him. He had told me then—with a certain excitement—that NASA had totally figured out how to aggregate data from various satellite-based sensors to be able to predict earthquakes accurately. This breakthrough discovery was soon to be announced. Of course, such an announcement from NASA never came in the end. NASA had not predicted any of the world’s earthquakes in the fifteen-year span since our encounter either—Christchurch included. For sure, if anybody ever finds the Holy Grail that makes earthquake prediction possible—and especially NASA—it will not be kept secret.
Earthquakes to Tsunamis
JAPAN—A REDEEMING VIEW The National Railroad Passenger Corporation, known as Amtrak, receives roughly $1.5 billion per year in subsidies from the federal government (corresponding to approximately 30 percent of its total income)1 to provide passenger train services to the nation.2 In return for this small investment, Amtrak’s customers have the pleasure of playing the popular game “Guess what time your train will arrive.” The Amtrak “report card” sets the punctuality goal as follows: If trains running on a given line arrive within fifteen minutes of their scheduled arrival time at least 80 percent of the time, it is a success.3 To be clear, this means that if a train scheduled to arrive at 8:00 a.m. actually arrives at 8:14 a.m., it is considered to be “on time.” However, for trains running on lines for which Amtrak does not receive subsidies from states (in addition to its massive federal government subsidies), the passing grade is set at 70 percent. On the basis of these generous passing-grade definitions, in 2019, seventeen of the twenty-eight statesupported Amtrak routes failed to achieve the self-imposed 80 percent standard, and fourteen of fifteen of the other routes failed to achieve the 70 percent standard.4 According to Amtrak’s Of
f ice of Inspector General, in 2018, passengers traveling to Atlanta on the Crescent line arrived “on schedule” (recall that this is defined as meaning no more than 15 minutes late) 3 percent of the time. These were the lucky folks; everyone else suffered an average delay of 124 minutes, and 46 percent of the trains were delayed by more than two hours.5 Further north, Canada’s ViaRail does not fare any better.
On some regional lines, federal government subsidies amount to approximately $600 per passenger, for arrivals late by two hours on average.6 On January 20, 2019, passengers traveling from Montréal to Halifax (775 miles away) arrived at their destination thirty-three hours later than scheduled,7 for a total travel time of roughly fifty-five hours (equivalent to fourteen miles per hour, the average speed of someone riding a bicycle—but still an impressive four times faster than if riding a donkey at steady pace).8 In both countries, delays are largely because passenger trains travel on corridors owned by freight railroads. Amtrak owns only 3 percent of the tracks on which it travels;9 ViaRail none. A train with three hundred containers generates $600,000 of income to those who own the tracks, while ViaRail pays them $25,000 to run a passenger train on the same line.10 It does not take a degree in accounting to figure out why cows on a freight train to the slaughterhouse have priority over folks on their way to the daily grinder.
Strangely, this state-of-things seems to be considered normal—or, at least, not enough of a concern nationwide to trigger riots in the streets by angry mobs of passengers demanding change.
In Japan, the picture is a little bit different. At a typical suburban train station, on a weekday at rush hour, where trains are scheduled to stop by every twenty minutes, there will be nobody on the railway platforms. No adults, no kids, no dogs. Nobody. That is, until two minutes before the train’s scheduled arrival time, when suddenly a mob storms the station and fills the platform, packing it elbow to elbow.
Two minutes later, as expected, the train arrives on time, the doors open, and the mob rushes in—squeezing everybody already on board a bit more in the process. The doors close and departure is on time.
Seeing a deserted station become so crowded at once may be disturbing to the casual tourist, but it makes complete sense. With the train arrival and departure times being known to the minute—and reliably so—what is the point of waiting on a railway platform when there are more enjoyable or important things to do in life?
Japanese train conductors have a clock on the dashboard and specific checkpoints that must be met—almost to the second—along the route, and anyone can lean on the window behind the conductor and watch the process unfold (like clockwork, literally). This is critical given that, for Japanese railways, a train arriving one minute behind schedule is considered of
f icially delayed.11 This typically
happens only when an earthquake causes operation stoppage or when people commit suicide by jumping on the tracks in front of an arriving train—and when that happens, apology notes are immediately distributed to all passengers so that they can give these notes to their bosses to in turn excuse their tardiness showing up to work.12 In fact, in two instances, Japan’s Metropolitan Intercity Railway Company issued an apology because its Tsukuba Express connecting Tokyo and Tsukuba departed the station twenty seconds early (November 2017) and twenty-five seconds early (May 2018).13 The official apology14 stated, “The great inconvenience we placed upon our customers was truly inexcusable.” Such a fact may be mindboggling to the average North American (even more so when stuck in a US train that is hours late), but in a country where crowds are used to storming the train station seconds before the departure time, it all makes sense. Since 1984, the average delay over a year for the Japanese bullet train (Shinkansen) has exceeded one minute only twice: it was 1.1 minutes in 1985 and 1.3 minutes in 1990.15 As they do for their train system, the Japanese take the construction business very seriously. As such, it is worthwhile here to provide a balanced view of the Japanese’s relationship with earthquake preparedness, response, and recovery—good and bad. This relationship is valuable to appreciate because, nowadays, the Japanese viscerally know that their homeland is earthquake country— maybe more so than any other nation. Nowhere in Japan can one hide from earthquakes. In fact, every year since 2005— maybe in some ways to be more proactive and make amends for the Kobe earthquake—the Japanese government has released a colorful earthquake map indicating the probability of powerful damaging earthquakes occurring within the next thirty years in the various parts of the country. From red in Tokyo and Yokohama where the odds are 85 percent and 82 percent respectively, down to orange, dark yellow, and light yellow for the other progressively less risky regions. No green. The government is making it clear that there is no such thing as a region with zero-probability of earthquakes in Japan.16 For sure, as indicated earlier, the Japanese were caught off-guard by the Kobe earthquake, in part because the bulk of their attention and preparedness efforts were focused on the Tokyo region to prevent a dreaded repeat of the Great Kanto earthquake and fire of 1923 that killed more than one hundred forty thousand people.17 As the Kobe disaster unfolded, all other governments worldwide watched humbly, because none could have done any better if similarly caught with “their pants down” (which is arguably the correct scientific term for when people who should have known better look clueless after an earthquake, and must deal with the resulting total mess). Nowhere in modern times had such a catastrophic earthquake hit a major, highly developed urban city that had all the sophisticated, complex, interdependent, and intertwined infrastructures needed to support today’s fast-paced economic life.
Naturally, the outcome was not pretty. The graphic details of how this earthquake had wreaked havoc on Kobe were broadcast all over the world 24/7, underscoring the fragile dependence of urban life and global economy on the vulnerable built environment and its entangled transportation, gas, power, water, and communication networks, and all the ensuing problems that arise when these systems fail.18 For sure, 6,279 deaths, 136,000 housing units lost, 300,000 people turned homeless, and $100 billion in damage19 definitely helped make that point.
On the positive side, the Kobe earthquake gave the Japanese government a solid “kick in the butt” (another highly technical term used to describe the jump-starting of important, new initiatives). Earthquakes can be good to those that survive and many post–Kobe earthquake achievements are worth highlighting.
First, as mentioned earlier, many aspects of Kobe’s reconstruction were fast. “Bullet-train” fast. The bullet train runs though most of the length of Kobe in tunnels under the mountains along the north edge of the city, emerging at the Sannomiya downtown area for the length of a train station, and then again in Nishinomiya at the east end of the city, where it continues on elevated bridges all the way to Osaka.
There, the bullet train’s elevated lines suffered damage over thirty-six spans, totally collapsing in some places.20 Lightning fast, all these bridges were repaired and reconstructed within eighty-one days.21 In some cases, this involved picking up parts and pieces of the collapsed reinforced concrete spans, slapping them together and wrapping them in steel plates, as if reassembling the pieces of a jigsaw puzzle, and injecting concrete inside the steel boxes to fill-in where concrete was missing.22 It helped a bit that the Kobe earthquake happened at 5:46 a.m., because no trains were running that early. So no bullet train emerged out of the tunnel at 175 miles per hour to fly over where no bridge existed anymore. Otherwise it might have taken weeks (or months) to respectfully recover all the body parts scattered among the rubble—a typical Shinkansen train has sixteen cars and about thirteen hundred passengers.23 The odds of such a mess happening at rush hours would have been high, with a train running in each direction in as little as five minutes,24 and considering that emergency braking of a bullet train at full speed still takes one mile before full stop.25 In truth, not everything in Kobe was repaired that promptly, as was mentioned in a prior chapter, but things got done—which is more than can be said for some countries where the urban scars of an earthquake remain visible for years, sometimes decades. Nonetheless, in spite of the Japanese’s efficiency with some specific construction activities, it remains that Kobe is a narrow, densely built urban strip wedged between the sea and mountains and packed with 1.5 million people, which can impede effectiveness. As such, it took six months to remove all the rubble, just as long to restore all six train lines, two years to rebuild Highway 1, which crossed town along its length, and two years to rebuild the port.26 Second, the Japanese government took proactive steps to ensure that the collective memory would not forget.
Going beyond the usual commemorative plaque, statue, sculpture, trinket, or monument that governments often plunk down somewhere in sincere (or fake) contrition, the Japanese built a full museum to “prevent memories of the Great Hanshin-Awaji Earthquake” (as the Japanese call it) “from fading and to pass on to future generations the thoughts of the survivors and the lessons learned from the disaster.”27 The museum is also part of the “Disaster Reduction and Human Renovation Institute.” The term human renovation (not defined anywhere in its website, and possibly a typical case of Japan-glish) likely refers to the institute’s broader mission of training disaster managers and broadly sharing knowledge with specialists in many related disciplines, as well as providing assistance when major disasters occur. Strategically located next to Kobe’s Art Museum and a waterfront park, it offers multiple floors of exhibits, 3D multimedia simulations, and full-scale reproductions of damage scenes.28 Third, the country invested. Some of it, evidently, could be counted as the typical politically astute way to respond to a disaster—the classic recipe that calls for scattering money all over the place to appease the grumbling population after the earthquake rumble. Maybe so.
Nonetheless, the Japanese did not pull punches there either.
As one interesting example, the government funded the construction of the world’s largest shake table testing facility, and had it built at the northern edge of the Kobe Prefecture.
A shake table is a platform that can be moved in space by servo-controlled hydraulic actuators, pretty much like a flight simulator or some large amusement park rides—think of Star Tour in Disney World, for example.29 Except that instead of being a box carrying forty people with mouse ears through a CGI field of asteroids in a fake starship, a shake table is a large flat platform on which anything that is constructed will be moved the same way the ground moves during an earthquake—vertically and horizontally at the same time. Like a giant jukebox, one can “play” the “time signature” of any past earthquake for which a strong motion recording is available from the world’s database of ground motions, which makes it possible to investigate how anything placed on the shake table would have performed during that earthquake. However, whereas the Disney ride only has to move forty Star Wars fans collectively weighing maybe as much as twelve thousand pounds (in an extreme case), a building and its contents can typically weigh between fifty to two hundred pounds per square foot of floor, depending on the type of construction. For a small 50' x 50' floor area, that equals one hundred twenty-five thousand to five hundred thousand pounds per story times the number of stories. In other words, to move a five-story building, it takes roughly the power of more than two hundred Star Tour simulators— and probably the force of as many Obi-Wan Kenobi’s.
Not surprisingly, before the Kobe earthquake, the largest shake tables in operation in research facilities across the world were roughly 20' × 20' when capable of full 3D motion (including pitch, roll, and yaw), which means that they could only support scale-models of buildings or small parts of buildings. Somewhat larger tables that could only move horizontally existed in Japan. After Kobe, the Japanese government literally put half-a-billion dollars “on the table” to build a full motion 65' × 45' monster platform capable of testing structures weighing as much as 2.6 million pounds, with servo-controlled actuators capable of imposing 5 million pounds of thrust in each horizontal direction, and 14 million pounds of thrust vertically. This testing facility also has its own power plant to generate the energy needed to dynamically move that table such as to recreate the movements of the Kobe earthquake at the base of full-size structures.30 Named “E-Defense” (to defend against the attacks of earthquakes—presumably like one defends against attacks of Godzilla), the facility opened in 2005.
Having “built the monster,” the Japanese then proceeded to “feed the monster,” providing research funding to build and test nearly one hundred multiple full-scale buildings of up to seven stories over the subsequent decade.31 If what separates kids from adults is the size of their toys, in this case, the Japanese definitely built the biggest playground—a gigantic Star Tour, for the exclusive pleasure of earthquake engineering researchers, albeit without music, smoke, mirrors, lasers, and sound effects.32 Fourth, Japan increased earthquake awareness countrywide. If earthquake awareness existed in Kobe prior to the 1995 Kobe earthquake, it was at best somnolent. As mentioned earlier, the government’s response was late, emergency relief was inadequate, and many would have starved if not for the more rapid and more effective food distribution by the Japanese mafia.33 When an organization profits from extortion, blackmail, financial market manipulation, and protection rackets, then it makes sense that the parasite must ensure survival of the host—speaking of the Yakuza here, not of the government, although some would argue that the description best fits the latter in some countries. In such dire circumstances, it is remarkable that there was no reported looting following the earthquake.34 Some have argued that the media were encouraged by authorities to close an eye—or, actually, a lens—to the few instances of fistfights, looting, and price gouging that were witnessed, but these misdemeanors at least did not occur at wholesale level.35 Some Japanese experts have alleged that the Kobe disaster is not the consequence of a deficiency in seismic awareness prior to the earthquake, but rather the results of a smug overconfidence in the superior quality of Japan’s infrastructure compared to everybody else worldwide.36 Either way, the reality of the 1995 earthquake hit home and provided the impetus for the government and engineering community to revise the seismic design and detailing requirements of all standards, and to mandate more stringent inspection rules for construction.37 Evidently, awareness by itself does not solve all problems. For example, from 1988 to 1998, the $3.6 billion Akashi Kaikyo suspension bridge was being built between Kobe and Awaji Island. It was to become the longest suspension bridge in the world. At the time of the 1995 earthquake, only the towers and cables had been constructed. Unfortunately, it so happened that the Nojima fault that ruptured for over twenty-five miles during the Kobe earthquake ran right between the two towers. As a result of the offset created by that rupture, after the earthquake, the towers found themselves one meter further apart than before. Fortunately, pulling the ends of a hanging cable is not too dramatic. The plans for the supported spans of the bridge were redrawn to accommodate the new site geometry and construction resumed. When the bridge opened in 1998, it not only was the longest suspension bridge in the world (with a 6,532 feet center span), but it was one meter longer than originally planned.38 Arguably, the engineers must have been aware of the presence of the fault and possibly expected that the bridge could accommodate the motion at the cost of some repairs after a rare earthquake at some point in the distant future. They probably just did not expect the distant future in question to occur so soon. Therefore, awareness in this case would likely not have changed the outcome, because it was physically impossible to bridge from Awaji Island to Kobe without crossing the dreaded fault line.
However, better disaster awareness can make a big difference for government emergency agencies planning their response to future catastrophic events. And, this being Japan, it did not take too long for all this practice to be put to good use, as the next massive earthquake happened only sixteen years later.
THE GREAT EAST JAPAN EARTHQUAKE OF 2011 Having learned hard lessons from the 1995 Kobe earthquake, the Japanese were definitely more ready when a magnitude 9.0 earthquake struck on March 11, 2011, off the coast of the Tohoku region. Officially named the “Great East Japan earthquake,” it was said to produce damage over 3,800 square miles and in thirty-seven cities—an area six times larger than the one affected by the Kobe earthquake.39 Contrary to the government’s initial confusion and slow response after the Kobe earthquake, this time the government was immediately informed of the disaster as it struck, and Japan’s Self-Defense Forces were mobilized within minutes. The first troops arrived within hours and more than one hundred thousand were deployed to the disaster area within three days. While there had been no coordination between the Japanese government and the few nongovernmental organizations (such as the Red Cross) that volunteered to help after the Kobe earthquake, this time coordination was immediately engaged with more than one hundred recognized nonprofit, private organizations and half a million registered and prepared volunteers. And while the Japanese government had rejected all of the 70 international offers of international assistance in Kobe, it accepted all of the 170 offers received following the 2011 Great East Japan earthquake.
Total cost of damage from the Great East Japan earthquake was estimated to be between $200 and $300 billion. More than two hundred miles of rail lines and seventy-seven highway bridges were closed, and as was the case following the Kobe earthquake, much of the infrastructure suffered damage due to the ground shaking, but many aspects of response improved. Gas and electric utilities were reconnected in one week instead of the six months it took in Kobe; railway service and roads all reopened in fifteen days, instead of the seven months in took in Kobe.40 And so on.
This was largely due to the fact that, while the earthquake itself was one of the largest ever recorded, it occurred offshore. As a result, the ground shaking itself on shore was typically less than considered for the design of new buildings. Also, the smaller Sanriku-Minami earthquake that had occurred in 2003 in the same region a few years earlier, like a wake-up call, made it possible to better prepare. For example, a number of columns of the elevated Shinkansen train line in the Tohoku region were damaged during the 2003 event. A retrofit program was rapidly implemented to enhance the ability of that infrastructure to resist earthquakes. That work was completed prior to the 2011 earthquake and, consequently, these columns experienced no damage during the 2011 Great East Japan earthquake.41 Yet, the official reported death toll turned out to be 15,897, with 2,534 additional people reported missing42— more than three times the human losses that occurred in Kobe. Only 6 percent of all the casualties occurred in buildings that collapsed during the earthquake shaking itself43—still a large number, but not the dominant factor.
The big killer in this case was the tsunami created by the earthquake.
When it comes to depicting what a tsunami is, Hollywood somehow is stuck in this groove of portraying it as a single one-hundred-foot-tall wave crashing on a city (or a thousand-foot-tall one, in some of the worst movies). Maybe this artistic vision helps with the box-office receipts, but this is not how tsunamis work.
First, it must be acknowledged that waves hundreds of feet tall are possible, but this type of mega-tsunami only happens when massive landslides fall into water bodies. The largest wave ever recorded in history was generated by one such landslide when a magnitude 7.9 earthquake occurred in Alaska close to Lituya Bay, on July 9, 1958. Mountains up to 3,400 feet tall, with steep walls similar to what is found in fiords, are located at the head of that bay. During the earthquake, a piece of mountain measuring forty million cubic yards got loose and plunged from a height of two thousand feet above sea level. At roughly 2,500 pounds per cubic yard, that made for a 100-billion-pound rock hitting water at 250 miles per hour,44 like a mega-size pool bomb.
The resulting splash wave ramped up along the surrounding mountains with such velocity that it stripped them of all vegetation. The resulting exposed bare rock provided evidence of how high the water reached at various locations. On the mountain immediately facing the splash, the water rose up 1,720 feet. By the time the wave reached the narrow bay’s outlet into the Pacific Ocean, the run-up was “only” 75 feet (beyond that point, the wave rapidly dissipated, absorbed by the wider sea).45 Computer simulations were able to reproduce the phenomenon. Not to be outdone by Hollywood, the researchers produced a “Director’s cut” of the actual computer simulation46—except that this one was anchored in scientific models, not artistic license. Regular tsunamis born from fault movement in the ocean do not even come close to producing such insanely high waves.
The mechanism that gives birth to most tsunamis is displacement along a fault under water. In simplistic terms, it can be explained as follows. If both sides of the fault move sideways from each other, no wave is generated because water cannot be sheared. However, if one side of the fault moves up with respect to the other, all the water above the side that moves up is also moved up.47 That upward energy creates the tsunami wave. The bigger the earthquake, the bigger the vertical movement, and the bigger the size of the wave.
While a wind-generated wave typically moves at speeds ranging from five to sixty miles per hour, with a wavelength of three hundred to six hundred feet, a tsunami wave travels at five hundred to six hundred miles per hour over deep water, with a wavelength of sixty to three hundred miles, slowing down to about thirty miles per hour in shallower waters when approaching the coast. As such, tsunami waves (yes, plural, because they arrive in bunches) will arrive from 10 to 120 minutes apart from each other.
Because of those unique characteristics, when a tsunami wave travels across deep sea, it is barely noticeable. Once it gets close to shore, as it slows down, as all waves do, it starts to form a more noticeable wave. As the energy piles up water to create a wave, water is usually first drained away from the shore. This massive recess of the water from the beach is usually an unmistakable hint to start running the other way and uphill. Curious folks who decide to go seashell hunting and explore up-close the shore exposed by this “unexplained” sudden, extremely low tide unknowingly walk to their death as the tsunami wave will arrive shortly after. Amateur videos have captured the fate of such unfortunate curious strollers during the 2004 Indian Ocean tsunami.
When a tsunami wave arrives, it is typically in the form of a tidal bore rather than a giant wave;48 the water level rises within minutes, and it is the sudden inundation of the coast and the volume of water that comes with it, due to the long period wave, that creates the problem. It is like a flash flood rising from the ocean, engulfing the entire visible coast, and spreading miles inland while pushing tons of debris along the way. Considering that a cubic yard of water weighs almost a ton, when that cubic yard is filled with floating debris, it is clear that one can only go with the flow when caught by it. And hope to survive. Most do not. It is estimated that two hundred twenty thousand people died across fourteen countries during the December 26, 2004, tsunami created by the magnitude 9.1 earthquake that occurred in the Indian Ocean.
In response to that 2004 catastrophe, the United Nations helped coordinate international efforts to install an Indian Ocean Tsunami Warning System that became operational in 2006.49 It consists of a network of seismograph and tsunami reporting buoys that can detect a tsunami wave off-shore, so that an alarm can be issued ahead of its arrival to shore, making it possible for people in nearby countries to run to higher elevations—which had not happened in 2004.
Operating such a network seems sensible, but unfortunately, when a magnitude 7.5 earthquake struck offshore of Indonesia on September 28, 2018, no data was received from the twenty-two buoys near Indonesia because they had stopped being operational in 2012 due to vandalism and lack of maintenance (vandalism of buoys puts truancy in a league of its own).50 The government still issued a tsunami warning and evacuation order as a precaution, but since cell phone towers along the coast had been destroyed by the earthquake and there were no emergency sirens installed in the cities along the coast, nobody received the evacuation order. The tsunami struck Indonesia and killed twelve hundred people.51 Returning to Japan, the tsunami created by the Great East Japan earthquake of 2011 made quite a mess across the Tohoku region. Immediately after the earthquake, Japan’s Meteorological Agency issued a tsunami warning, estimating the height of the water run-up to reach eighteen feet in some regions along the coast. However, given that seawalls from fifteen to thirty feet tall had been constructed along the entire Tohoku shore decades before, half the population in some of these coastal cities apparently did some mathematical guesswork, concluded that the sea walls would provide the needed protection, and ignored the warning.52 This being the first major tsunami happening in the smartphone era, it became the most recorded one in history.
Hours of video are available showing the progressive run-up of water in various bays and shores, until all the tsunami defense walls built along the coast were overtopped by rushing water and drifting boats started to ride through the resulting waterfalls. One stranded driver even filmed the whole thing from inside his car as the street flooded and his car was swept away, thrashed about along with other debris —calmly running his windshield wipers through the entire ordeal.53 According to the Japanese agency tasked with the $250 billion reconstruction effort, as the ocean flowed inland— reportedly as far as six miles from the coast—more than 120,000 buildings were completely destroyed and over a million were half-or partially destroyed.54 In fact, when it was mentioned earlier how fast utilities, railroads, and roads returned to service, that did not include the infrastructure that was permanently lost due to the tsunami. Likewise, most of the casualties caused by this earthquake are directly attributable to the massive tsunami that followed.
To make matters worse, the tsunami also flooded the site of a nuclear power plant and created a nuclear crisis—as will be described in more detail later. A total of 470,000 people were evacuated from the coastal region and by 2020, 77,500 had yet to return,55 as the entire region that was affected by the tsunami was slowly being reconstructed. To make reoccupation of the region possible, one of the government’s first tasks was to remove more than thirtyone million tons of debris from the disaster area.56 By early 2018, new forty-one-foot-tall concrete seawalls had already been constructed along 245 miles of coastline, at a cost of $12 billion, to replace the shorter ones that failed in 2011, providing all coastal residents with a glimpse of what inmates experience, looking at a high wall all day long.57 Beyond giving the term “ocean view” a whole new meaning, the Japanese government also emphasized that the seawalls could not be expected to stop all future tsunamis, that evacuation plans must be in place and promptly followed in case of overtopping, that lower elevation coastal zones should preferably not be redeveloped, and that new structures should be built with the ability to resist a certain level of tsunami wave forces.58 And not surprisingly, a museum was built in Tohoku in commemoration of the devastating tsunami that followed the 9.0 magnitude “Great East Japan Earthquake” of March 11, 201159—right next to the rebuilt, taller seawalls erected to protect the region from future damaging tsunamis.60 More protections, newer constructions, enhanced awareness. All good things.
And yet, earthquake disasters are still looming in Japan.
GETTING READY FOR THE BIG GAME As damaging as the 1995 Kobe and 2011 Tohoku earthquakes were, these were small potatoes compared to the biggie that lies ahead for Japan. Yes, when it comes to disasters, Japan only plays in the big leagues.
The SwissRe reinsurance company reported that “the metropolitan area of Tokyo-Yokohama in Japan is by far the most earthquake-exposed community” of the 616 metropolitan areas it studied, worldwide.61 It also stated that “the cities most exposed to tsunami risk are in Japan.” Incidentally, beyond the earthquake-tsunami twofold blow, a 2014 report from the UN Office for Disaster Risk Reduction stated, “The exposure of the population in Japanese port cities to potential wind damage is extremely high. Tokyo ranks highest in the world for exposure to potential wind damage, with Osaka‐Kobe in sixth place.”62 With all hazards aggregated, Tokyo is ranked number one for the predicted number of people affected and loss of productivity due to future disasters—a fun fact that was probably not highlighted in the city’s bid to host the 2020 Olympics.
RMS, which brands itself as “The World’s Leading Catastrophe Risk Modeling Company,” has estimated that a repeat of the magnitude 8.2 1923 Tokyo earthquake alone— without a tsunami to boot—would cause insurance losses in excess of $100 billion today. Keeping in mind that not all the Tokyo infrastructure is insured, that implies a lot more than $100 billion in total damage—possibly even more than ever seen before. This is partly attributable to the fact that a lot of Tokyo is packed with old construction that is not expected to fare well in future earthquakes,63 including old residential homes of the type that were badly damaged during the Kobe earthquake.
Japan has not always been the prosperous high-tech giant known today for its high-quality high-reliability products—overlooking some shameful recent incidents where quality has not been up-to-snuff, deliberately or not.
Baby boomers will recall a postwar era where “made in Japan” meant cheap and of dubious quality, before the decades when SONY, Panasonic, Toyota, Honda, Canon, Nikon, Yamaha, Suzuki, and countless others “upped their game” and took over the world. In those early days, the country was hungry, raw materials were scarce, and the impact of this poverty was not only felt in low-quality exports. New buildings and infrastructure might be designed to the latest standards and equipped with the greatest technology—cost being no issue—but the overwhelming majority of the buildings, bridges, and other components of the Japanese urban landscape were designed and constructed in the 1950s–1980s. These are not necessarily “up to snuff” either. It was well demonstrated by the 1995 Kobe earthquake that older buildings and bridges were more prone to collapse and catastrophic failure. On the positive side, after the Kobe earthquake, massive efforts were undertaken to strengthen critical facilities, and particularly a lot of schools, but a study conducted after the Great Eastern Japan earthquake showed that while nobody was killed in the schools that had been strengthened prior to the earthquake—because none of them collapsed—they still suffered a lot of damage, had to be evacuated, and were to be demolished because repairing them would be too costly.
64 To the parents of the little kids who survived, that is good news. To the taxpayers or insurance companies who will foot the bill, it remains a painful outcome. And that is what was done for essential buildings deemed worth investing money in to achieve life safety—that is, life safety alone, nothing more. Far fewer old residential buildings in Tokyo have been strengthened in anticipation of the next large earthquake.
An old joke that poked fun at European stereotypes went like this: “Heaven is where the cooks are French, the lovers are Italian, the mechanics are German, the police are British, and the whole place is run by the Swiss. Hell is where the cooks are British, the lovers are Swiss, the mechanics are French, the police are German, and the whole place is run by the Italians.” Not to be outdone, one of the Chinese adaptations moves it many notches up on the politically incorrect scale, stating, “Paradise is having Chinese food, a Japanese wife, and an American house, whereas hell is having a Chinese wife, a Japanese house, and American food.” Leaving aside the sexist aspects here— where some horrible stereotypes are at play—and focusing on the infrastructure aspects only, since this is the topic as hand, the story suggests that Japanese housing is less than desirable. It may be true that a Japanese house, at roughly one thousand square feet,65 is not spacious when compared to American houses, which average slightly more than two thousand square feet66 (and more than twenty-six hundred for new homes67), but it is actually still twice more spacious than those in China68—which goes to show the Chinese jokers that, if anything, stereotypes are for fools.
Granted, the Japanese have not helped their reputation for cramped space, with real estate statistics reporting that 25 percent of the Tokyo population has 120 square feet of living space per person (down to as little as 50 square feet in some cases).69 In fact, 70 percent of Tokyo’s population is living in less space that the bare minimum recommended by Japan’s own government for citizens to have a “healthy and culturally fulfilling life,” pegged at 250 square feet for a single person, including bathroom and kitchen.70 When space is at such a premium and every inch of space counts, it is surprising that Japanese music lovers have continued buying their music on compact discs and have been slower to embrace music streaming than those in other countries, but it shows that the Japanese psyche and practices are— well—uniquely Japanese.71 Beyond the discomfort of cramped space and high urban density, those lucky enough to own a detached home face the problem of high earthquake vulnerability if their house is in an older residential neighborhood. During the rebuilding period following World War II, up to 1981 when construction standards were substantially updated, the traditional Japanese house was built of timber posts and beams without much internal partitions that could provide bracing against the lateral forces imparted by earthquakes. Furthermore, to overcome the uplifting forces due to typhoon winds, roofs were typically built of heavy clay tiles. While the weight of the tiles works well to withstand wind forces, adding mass is the worst possible thing to do when it comes to earthquake forces. When the ground moves, it imparts an acceleration to the building. As Newtonian physics demonstrated in the seventeenth century, a mass that accelerates creates an inertia force. The bigger the mass, the bigger the force. In essence, this is why an eighteen-wheeler needs the force of a 600hp engine to haul a trailer full of merchandise while a 123hp Ford Fiesta is sufficient to lug around a small family with grocery bags in the trunk. In this case, the heavier the roof, the larger the force that the house must resist when its heavy roof accelerates sideways because of the horizontal ground motions. As a result, over 60 percent of the wooden structures in Kobe were heavily damaged or collapsed during the 1995 earthquake.72 To add to the injury—in another pernicious “double whammy” effect—almost nothing burns better than a wooden house. In the days following the earthquake, broken gas pipes, toppled kerosene heaters, downed power lines, and damaged electric appliances and wiring combined to ignite 148 fires that spread over 163 acres (equivalent to the surface of 100 US city blocks), consuming 6,900 buildings and killing five hundred people.73 Note that these were genuine post-earthquake fires.
Incidentally—as an aside—in many countries, homeowners do not know that standard insurance policies, by default, do not cover earthquake damage unless a special endorsement is added to the policy, at an extra premium, but that losses from fire following an earthquake are generally covered.74 When that is the case, it would not seem to be in the best interest of insurance companies to publicize this loophole, as it could motivate the unscrupulous owners of homes hopelessly damaged by earthquakes, and uninsured for that damage, to set them on fire solely to collect the insurance money.
Yep, indeed, disasters are looming.
ON THE DISASTER TRAIL
The “Can’t Escape It” Moment Approximately 80 percent of Japan is covered by mountains and is therefore uninhabitable. The remaining 20 percent consists of crowded plains where industries and infrastructure are so densely packed together that it is also uninhabitable.
Although things are not nearly as dire as suggested by this case of bad Japanese humor, it remains that the Japanese have developed a distinct societal code of conduct that has made it possible for them to achieve some level of “harmony” while living in such close quarters—as far as harmony implies conformity to the group. Among the various social rituals needed to achieve this peaceful integration, getting down to some serious drinking with colleagues after work is a prized one. In an environment where many will perspire soaking wet at the mere thought of having to say “no” to someone, getting drunk allows everything to be said—and to be forgotten the day after.
Not for the drinking, but rather for the attraction of a serious culture shock, I had accepted an invitation to spend six months in Kyoto with my family, starting in mid-January 1995. Six months before departure, unable to purchase plane tickets to land in Japan on January 16—because the flight was sold out—we had to settle for the next available flight, arriving on the 18th.
The evening before our departure, my mother called, wondering if Kobe was close to Kyoto (yes, it is), because it was reported on the news that a big earthquake had just happened in Kobe, killing fifty people. More than twenty million people live in the Kansai area of Kobe, Osaka, and Kyoto, so fifty casualties (0.00025 percent of the total population) did not give rise to concerns that evening.
Besides, the flight had not been canceled, so how bad could it be?
Amazingly, the brand-new Kansai airport was open and we landed safely on the 18th, but by then, the news reported that the death toll had exceeded five thousand. All concerns were legitimate now.
Some will call it luck, some disappointment—it is all a matter of point of view—but we missed the Kobe earthquake by twenty-four hours. We got the key to the rented house, opened a bank account, did a first grocery, and off I went to Kobe. I was fortunate enough to join the Architectural Institute of Japan mega-team tasked to perform earthquake damage reconnaissance through the entire Kobe urban area, as the only “Gaijin” on the team—the jury is still out as to whether Gaijin is a pejorative, neutral, or positive way for the Japanese to refer to a foreigner.
This was a three-day expedition by 110 engineers driven by the desire to survey thousands of buildings throughout Kobe, from sunrise to sunset, to document the extent of the damage produced by the earthquake. As all hotels were evidently closed, the organizers had arranged for the team to sleep at a Buddhist monastery, as long as our entire team would be kept out of view of the monks—presumably so as not to disturb them in their meditations. The monastery was located at the top of a hill in Kobe, where no earthquake damage had occurred.
After grueling days of earthquake reconnaissance spent walking in dust—and sometimes, rubble—back to the monastery in the evening, some members of the reconnaissance team pulled out playing cards and bottles to partake in social rituals not quite in harmony with those of their hosts. As the monks were heard chanting their prayers in separate distant rooms—moving along on a path undisturbed by the earthquake—some of the engineers merrily violated the “Fifth Precept” of Buddhism (“do not take intoxicants”).
Then, the ground started to move, and all that playing and drinking (and distant chanting) stopped. Time stopped.
A time pregnant with the uncertainty of the future, that provided a glimpse into a universal human nature. An instant during which the desire to forget the weight, burden, and gravity of the situation, was robbed by a sudden, unwelcomed return to reality. That instant when everybody wondered if the tremor would grow or stop—or rather, whether the building would collapse, as all were engineers.
This is the moment when it becomes crystal-clear that there will be a point—anytime, anywhere—when it will be too late. A time when one must face the consequences of previous decisions made by others in the past: the “can’t escape it” moment.

The Water Magnet
THE ATTRACTION Like bugs that are hypnotically attracted to light,1 humans are inherently attracted to water—it is in our DNA.
2 It is probably not that surprising given that life on earth first appeared in water; given that we spend the first nine months of our life floating in mom’s amniotic fluid (which is mainly water enriched with electrolytes, proteins, and a bunch of other life-nourishing stuff); given that 50 to 80 percent of our body weight is water (depending on age); and given that we will not survive long without water. Not to forget all the joys that can be had with water, such as boating, fishing, surfing, snorkeling, swimming, and skinny dipping, to name a few. As any Florida real estate agent will attest, it seems like everybody wants a pool and/or a water view—be it of the ocean, the Intracoastal Waterway, a bay, a lake, a pond, or a bird pool. If it is possible, when precariously leaning over the railing at the very end of a balcony, standing on toes and twisting neck, to see a tiny bit of the sea through the tree leaves on a day when the wind conditions are just right and the horizon is clear, then it is a sure bet that this condo will be advertised as having an “ocean view.” Humans naturally gravitate toward water, period. Nearly 80 percent of the world’s population lives within sixty miles of an ocean, lake, or river—the closer the better, as property prices unequivocally demonstrate.3 That attraction is certainly fine, and it can be healthy. However, the danger of that vivid attraction to water arises when it hijacks all other considerations—when reality is distorted to fit a perception that the waterfront is a safe, peaceful, and idyllic location.
Unfortunately, that distorted reality exists when it comes to hurricanes.
THE DNA OF A HURRICANE Those who fundamentally believe that humans rode dinosaurs like horses a few thousand years ago, or that wet blanket wraps can cure tuberculosis as effectively as they can cure concrete (technically speaking, concrete does need to cure at a desired moisture level to gain strength), or that it was a bad idea for the gills-less residents of Atlantis to upset their gods, are likely to find this section offensive and can easily skip it without any impact on their longevity on this earth.
It was long believed that hurricanes were the product of a god with some anger management issues. Hurricanes and typhoons are essentially both cyclonic disturbances of the same nature. Cyclones over the Atlantic Ocean are called hurricanes,4 because they come from the Mayan god Huracan,5 whose single foot is actually a snake head blasting wind, water, and fire (lightning) from its open jaws.6 Cyclones over the Pacific Ocean are called typhoons, because—well, it is less clear. In Hindi, Arabic, and Persian, the word “tufan” means “big cyclonic storm”; in Chinese, “tai fung” means “a great wind”; and there is a multiheaded fire-breathing dragon called “Typhoon” in Greek mythology.
7 While Typhoon’s cute offspring included Cerberus, the three-headed mean dog that guarded the Underworld, and the Sphinx,8 who killed all those who could not resolve its riddles, it is unknown if the reputation of that legendary Typhoon somehow traveled along the silk road to India and China,9 but it is plausible.10 A tropical cyclone, be it a hurricane or a typhoon—or simply a cyclone, as folks in the southern hemisphere call them—occurs when a small weather disturbance “sucks-up” enough energy from warm tropical oceans to grow into a rotating, organized system of clouds and thunderstorms that produces maximum sustained winds of more than seventyfour miles per hour. At less than that, it is deemed a tropical storm. At less than thirty-nine miles per hour, it is a tropical depression.11 Neither of those is a great day at the beach— except for extreme parachute surfers who think that the joys of huge waves and high winds are worth defying the hazards of thunderstorms.
The threshold for hurricane-force wind was set at seventy-five miles per hour by Herb Saffir, the structural engineer who developed in 1969 what was to become the Saf f ir-Simpson hurricane scale, because he felt, from his experience with building codes, that there was no way to quantitatively predict structural damage from winds greater than seventy-five miles per hour.
12 Saffir provided various descriptions of damage as wind speed increased, and the meteorologist Robert Simpson, director of the National Oceanic and Atmospheric Administration (NOAA) hurricane center, supplemented Saffir’s scale by adding wind speeds threshold for each category of damage on the scale, as well as corresponding expected storm surge heights.
A storm surge can be defined as follows. As the high winds from a hurricane push water toward the shore, it can create a substantial temporary rise in the water level above normal tide levels. This can be one of the most dangerous parts of a hurricane. The storm surge produced by Hurricane Katrina in 2005 pushed the sea ten to twenty feet above tide level along the southeastern Louisiana coast, and twenty-five to twenty-eight feet above tide level along parts of the Mississippi coast.13 That effectively brings the ocean into town. In other words, a small storm surge will bring water into the streets, a moderate one will bring waves crashing on building walls, a large one will swallow the neighborhood. This is the difference between “surprise, the beach is gone,” “surprise, the beach is in the living room,” and “surprise, the neighborhood is gone.” Hurricane wind speed obviously is the source of the storm surge problem, but the height of the surge is also affected by the speed and angle of approach of the storm, the slope of the ocean floor near the shore, and the topography of the shoreline and shape of coastal features— all features unrelated to wind speed.14 As a result of a better scientific understanding of the above phenomena and evidence from past hurricanes, in 2010, the predicted storm surge levels were stripped from the Saffir-Simpson scale and it returned to be only a wind-speed scale.
The Saf f ir-Simpson scale arbitrarily sets wind speed thresholds of 74, 96, 111, 130, and 157 mph for the hurricane categories 1 to 5, beyond which various kind of wind damage are expected—but not guaranteed.15 In that scale, the explicit descriptions of damage corresponding to hurricane categories were formulated in the 1960s and 1970s, and a lot of those were more “educated guesses” than hardcore engineering and science.16 Furthermore, as building codes evolved and construction quality in coastal regions improved over time, the damage descriptions became less relevant, but the wind speed threshold limits remained. Incidentally, the specific numbers used to define the threshold levels of the scale—for example, using 74 mph instead of 75 mph to define Category 1—might suggest great accuracy, but it is not the case. Round numbers could have been used at the transition points, but there is no possible conversion of wind speeds in knots, mph, and km/h that could satisfactorily give round numbers in all three measurement units.17 Incidentally, there is no wind speed ceiling in that scale.
The Category 5 label is assigned to all hurricanes having sustained winds of at least 157 mph, but given that Levels 1 to 5 are “spaced” about 20 mph from each other, and that nearly twenty hurricanes/typhoons to date have recorded sustained winds in excess of 175 mph, and a couple even reaching 195 mph, some have suggested the need to add Categories 6 and 7 to the scale.
So, to sum it up, hurricanes are a well-known natural phenomenon. A hurricane is simply a rotating low-pressure system that builds-up strength over warm ocean waters.
The physics of how it forms and intensifies is relatively well understood.18 In the Atlantic Ocean, this long-distance traveler often shoots off from the coast of Africa, meandering along westward with the trade winds19 until it either: (1) dies for not finding the warm ocean waters it needs to feed upon; (2) veers off north as it bumps into high pressure systems; or (3) hits shore and moves inland, which is technically defined with scientific certainty as the point where “all hell breaks loose.” At that point, three problems arise that can affect all those humans previously attracted by the beauty of the ocean at that very location of possible hurricane landfall: (1) winds speeds up to nearly 200 mph, (2) ocean levels surging up to tens of feet above normal, and (3) literally feet of accumulated rain.
Wind speed and storm surge are intertwined like the two twisted coils of a DNA strand. They both happen at the same time, and both can kill—so both are reviewed individually below. Floods due to rainfall will be the topic of the next chapter.
HUFF AND PUFF AND BLOW YOUR HOUSE The cannon was facing an eight-inch-thick masonry block wall. The distance was small. Those in the room did not think much of it, as they had fired that cannon time and time again. Even though they knew firsthand the destruction it could wreak—execution style—it remains that firing a cannon is a job. Like any other job. It probably can be fun, but a deadly tool is still only a tool after a while. Yet, the blur of routine cannot be allowed to distract. Safety first.
A cannon is powerful, and errors can be costly.
The cannon was loaded carefully, as always. The target was in the crosshair of the viewfinder. The trigger was depressed. The fifteen-pound projectile flew out at more than 100 mph, penetrated the wall, and went all the way through, as if the wall provided no defense.20 Bullseye!
Another wind impact test successfully completed. Propelled by the large missile pneumatic cannon at the research facility, the twelve-foot-long southern pine two-by-four that flew through the room was not stopped by the wall.21 Hopefully, when that happens during a future hurricane, because debris are flying all over the place in 100–170 mph winds, there will not be people on the other side of that wall —people convinced that hurricanes are not a big deal and determined to “ride the storm,” adamant that the news media always make things sound a lot worse than they really are.
Granted, the news media are guilty of doing just that— which does not help. If an entire town survives a storm unscathed, except for one poorly constructed roof that is blown off, all the networks will park their cameras and news anchors in front of the roofless home and fill the airwaves with it. They will show all the rain that fell in, interview neighbors for lack of experts, show the devastated owners— a bonus if they can catch them sobbing—and inflate the story beyond measure, describing the place like a war-torn disaster zone. Some will call this great journalism, but that is a matter of opinion. The problem is that, over the long run, after calling wolf before each puny windstorm that rips a few shingles here and there, credibility erodes and there is the risk that the serious threats will be ignored. Those who rode the previous puny windstorms will be empowered to stay put for the real deal. Unknowingly, they will have to face the projectiles of the missile cannon—except that Mother Nature operates her own missile cannon recklessly, with lots of projectiles flying in all directions, because Mother Nature is a loose cannon.
The energy of a projectile increases with the square of its velocity.
22 In simple terms, the impact of a projectile at 80 mph can produce four time more damage than one at 40 mph; one hitting at 160 mph produces sixteen times more damage. To make it worse, the number of things that can get ripped up—and thus become projectiles—also increase with the square of the velocity. This is because the pressure created on an obstacle by blowing wind quadruples when the wind speed doubles.23 If construction is of good quality, it may be that no parts of a building will suffer damage in an 80 mph wind, but increase that wind to 160 mph and many things start to fly out. Roof shingles, parts of home siding and gutters, pool enclosure frames, outdoor furniture, traffic signs, tree branches, and all kinds of other debris. The kind of stuff that, at some point, is bound to kill one of those daredevil weather-anchors that feel compelled to broadcast their report while trying to stand up to hurricane wind and rain, for dramatic effect—and the kind of casualty that will endup on YouTube for posterity.
In Florida, the latest building code requires impactresistant windows, or shutters in all new homes built within one mile of the shore in regions where buildings must be designed to resist more than 100 mph winds.24 Impactresistant windows are so rated if they have been subjected to the missile cannon test and have successfully “survived”—which means that during the test, the glass panes can fracture, but the window should still remain in place and prevent water from penetrating. Windows can be certified for various speeds of the projectile, typically up to 55 mph.25 Typically, only a quarter of the impact force of a 100 mph projectile, and one sixteenth that of a 200 mph projectile, but during a storm, not all projectiles ram windows head-on like the fifteen-pound board in the missile cannon test.
In some coastal areas across the United States, when building a new home, buyers are typically offered the option to install hurricane-resistant windows, albeit for a premium.
Some may find it mindboggling that such a requirement is not mandatory in all coastal regions, but then again, in the land of the free and home of the brave, some towns do not even have a building code to start.26 Such holdover towns typically are incentivized to change their mind after being devastated by a hurricane, or when insurance companies refuse to issue coverage to buildings not built in compliance with a specified building code.
In fact, as far as incentive is concerned, the insurance industry itself learned the hard way the fact that wind alone can produce massive losses during a hurricane—that is, even without any help from storm surge and floods. Every business exists to make a profit, and the insurance industry is no exception. Except that, unlike most businesses, the insurance industry collects money from its customers without providing any immediate good or service in return.
Although an unflattering analogy, the insurance broker is a sort of bookmaker taking gambling money on long odds, as an intermediary placing bets and paying out the winnings on behalf of the gambler—except that in this case, the bookmaker takes the money upfront to avoid having to hire shady characters to collect from insolvent customers.
Obviously, the long-odds bet in this case has nothing to do with predicting which horses will finish first, second, and third in a race, but the principle is the same. The insurance company collects the money from all, calculates the odds that the “winning combination” will occur, and works out the math such that the payout to the “winner” is less than the sum of all money collected. The only difference is that the insurance industry is dealing with unlucky outcomes—in other words, the “winning combination” in this case is the occurrence of an unfortunate event, be it a fire, a theft, a health problem, or any other type of infrequent circumstance that derails one’s normal pursuit of happiness.
In principle, everybody understands the insurance principle—except, in some rare instances, the insurance industry itself. The whole principle of pooling everybody’s money is to help out the unlucky few who run into trouble due to unforeseen circumstances, so the concept breaks down if everybody needs to collect at the same time because everybody ran into trouble due to the same unfortunate circumstances.
In 1992, Hurricane Andrew made landfall at the Biscayne National Park, about twenty miles straight south of downtown Miami. As the first Category 5 hurricane to hit the United States since 1960, it packed sustained winds of more than 140 mph,27 and the corresponding punch delivered $26.5 billion in damage28—most of it from wind alone.29 The wind forces wiped out sixty-three thousand homes; severely damaged more than a hundred thousand; left one hundred seventy-five thousand people homeless; and damaged eighty-two thousand businesses, schools, and hospitals, as well as thousands of traf
f ic signals, and thirty-three hundred
miles of power lines. As the state was busy trucking away the twenty million cubic yards of debris, it could only be forever grateful that it had been so lucky. Immensely lucky.
Had the storm hit shore forty miles further north, it would have crossed the most heavily populated part of Florida, and would have left an even more significant trail of destruction.
Less lucky was the insurance industry. It was as if everybody at the horse races had bet on Secretariat in first, Sea Biscuit in second, and American Pharaoh in third, and that trifecta of thoroughbred lined up perfectly as it crossed the finish line. If such a racetrack catastrophe happened, the bookies would run away, unable to pay all their clients.
That is pretty much what many insurance companies wanted to do following Andrew. The hurricane trigged an insurance crisis in the state of Florida, mostly because the property and casualty insurance companies in the state had not realistically accounted for the possibility of a damaging hurricane hitting the state when calculating the insurance premiums. They collectively had “misevaluated” the risk— effectively living in denial of risk—to remain competitive on the market place. As a result, after Hurricane Andrew, they were left with $16 billion in insured losses, something they were unprepared to absorb. They were immensely lucky that the storm did not hit smack dab in the middle of Dade, Broward, and Palm Beach counties, where there was $370 billion in insured property—on top of much uninsured property—but it did not matter. Many insurance companies went belly-up.30 Others wanted to withdraw from the state.
Giving them further incentive to flee, builders and developers claimed that construction practices were unlikely to change because building hurricane-resistant houses would be too costly, positioning the housing market beyond the reach of too many people.31 To prevent a massive exodus of underwriters, the state legislated rules to allow limited and progressive disinvestment over successive years, such as not to leave state residents without coverage. In parallel, the state created an association that required participation of all private insurers to serve as an insurer of last resort for those abandoned by the insurance market. In spite of this, over the past decades, after each hurricane, some of the largest property insurance firms pulled out of Florida’s insurance market,32 not renewing the policies of millions of homeowners.33 For those that remain, standard policies have a deductible for wind damage equal to 10 percent of the insured property values, meaning tens of thousands of dollars—although smaller deductibles of 2 percent or 5 percent are available as options, for higher premiums.34 Practically, hurricane-resistant homes can be built—and are being built all the time. An investigation conducted following Hurricane Andrew revealed that a lot of the damage occurred, first, because the South Florida Building Code might have called for buildings to be designed to resist 120 mph winds, but its provisions effectively did not match that claim, and; second, because buildings were either poorly designed or poorly constructed, and because building inspection was inadequate.35 In other words, having great plans and design is only half the solution; implementation is key. For example, in many of the roofs that were ripped up by Hurricane Andrew, it was found that the nails driven by shoddy workers had missed the roof trusses altogether—the type of errors that could have been caught by building inspectors.36 The best building plans are useless in the hands of a negligent contractor that cuts corners with the same disregard for safety as Napoli drivers who treat traffic lights like cute Christmas decorations—severe warning: it is very dangerous to go through a green light in Naples.
Constructions of the same vintage as those damaged by Hurricane Andrew still exist across all of Florida, including from Miami to Palm Beach. If that part of the coast narrowly escaped the blunt force of Hurricane Andrew, it is because proximity to the hurricane center makes a huge difference when it comes to damage. The spiral of clouds that constitutes a hurricane, as seen in satellite photos, can be gigantic and drop rain over multiple states at the same time, but wind speed is the highest near the center of the storm, along the eye-wall. The eye of the hurricane is the very middle of the cyclone. Somebody standing in the eye of the hurricane effectively is in a column, twenty to forty miles in diameter,37 of relatively calm air under a clear blue sky. In fact, a great way to avoid hurricane winds would be to stay in its eye, walking/running/driving/flying with it all the way as it travels over the continent—a stunt impossible to accomplish, as one would have to go through half of the hurricane to reach its center in the first place. Right around the eye is the eye wall, which is where the strongest winds are developed. Then, moving farther away from the wall, the wind speeds progressively decrease. For Hurricane Andrew, peak sustained winds exceeded 140 mph along the wall at landfall, but were down to 110 mph in Miami and 75 mph in Fort Lauderdale,38 respectively twenty and forty miles away. That also corresponds to nearly two-and fourtimes smaller wind forces acting on buildings. All of the old construction that still exists from Miami to Palm Beach got lucky in 1992, but it will eventually hit a wall someday—that of the eye of a future hurricane—and get somewhat less lucky then. So will those who believe themselves invincible, claiming they “survived the hurricane” because the clouds above their abode rained for a few hours in 1992, unaware that distance matters. Just like everybody can survive a magnitude 8 earthquake if standing far, far, far from the epicenter, everybody can survive a Category 5 hurricane if standing far, far, far from its eye—which is not a guarantee of survival in future events striking closer.
SURFING IN THE LIVING ROOM On October 9, 2018, Hurricane Michael approached the coast of Florida, dead set on hitting the panhandle as a Category 4 hurricane.39 Mandatory evacuation orders were issued that very morning for oceanfront counties lying within the projected path of the hurricane. Florida’s governor had already declared a state of emergency days before, making it clear that this “monstrous storm”40 was heading toward the coast, warning that it was going to be “the most destructive storm to hit the Florida panhandle in decades” and “life-threatening and extremely dangerous.” He added: “You cannot hide from this storm. You can rebuild your home, you cannot rebuild your life.”41 Wind speeds of 155 miles per hour and twelve-foot storm surges were projected.42 The only way to make it scarier would have been to forecast that the twenty-five-story-tall monsters of “Pacific Rim” would emerge from the sea at the same time to trample everything else that had not been already destroyed by wind and water alone.
Yet, in spite of these unambiguous statements, hundreds of people decided to “ride the storm” for one reason or another. For some, financial hardship made it impossible to temporarily relocate, and all they could afford to do is to huddle down and hope for the best—very sad indeed. But for everybody else who could afford a tank of gas and drive away from the coast, their decision seems less sensible. For some, it was an unconscientious bravado spirit that made them feel invulnerable to the elements, summarized by “no big deal, we’ve been in hurricanes before,” possibly combined with the appeal of financial gains hoping to sell storm videos to the media—the video selfie of a fool drowning in a storm surge may indeed be worth something.
For others, it might have been the sheer attachment to material possessions that made it emotionally impossible to abandon everything, or the misplaced hope that the storm would hit somewhere else and that nothing could possibly be as bad as projected (weather forecasts are always wrong after all, aren’t they?). When some said after the hurricane, “I don’t know why we stayed,” they essentially confessed to having made an uneducated bet that they would survive.
For some, the bet paid off—there are, after all, always five winners for every loser in Russian Roulette—but others were not so lucky.
Not many of those who decide to “ride the storm” have usually bothered assessing the “seaworthiness” of their home—not unlike those who boarded the Titanic—as their home will literally attempt to play the part of a ship during a storm surge. Seawater weighs a little bit less than a ton per cubic yard—1,728 pounds to be exact—which makes it dif f icult for the flat wall of a house to stop a crashing wave.
A better strategy is elevating the house, as if on top of a wharf, to allow waves to travel below it. An alternative approach, applicable when the ground level story is enclosed, is to purposely allow water to flow inside the house, such as to equalize the pressure applied to the walls by the water inside and outside the house—albeit an inconvenient consequence if the first level is anything other than garage space.
When Hurricane Michael made landfall in Florida’s panhandle, it chose as its target the coastal community of Mexico Beach, a small seaside town predominantly consisting of 1960s and 1970s bungalows, intertwined with some relatively more recent constructions. Water pushed to shore by the cyclonic winds surged to roughly twenty feet above mean sea level.43 In this day and age where everybody has an HD camera in their pocket, with plenty of people dead set on “riding the storm” against better judgment, the entire world (wide web) could see footage of homes collapsing or floating away like boats, and bungalows turned into submarines with water up to (or above) their roofline, among floating debris, crashing waves, and pouring rain.44 Receding waters revealed roads and lands filled with sand and piled-up debris of wood frame homes, dead bodies that got trapped in the floating debris, concrete slabs where homes used to be, homes standing without a roof and/or missing a few walls, dead vegetation, and pretty much nothing to celebrate.
Likewise, drone videos of “ground zero” taken following the hurricane showed the widespread devastation across town. The fifty-year-old bungalows had been engulfed—their seaworthiness was nil, and they had been for the most part destroyed. However, among the mix of debris from houses, dead trees, boats, and more, the keen eye not distracted by the scenes of destruction but looking for good news can find in these videos some buildings standing up, with little or no visible damage. Data retrieved from real estate websites confirm that these surviving homes were the most recently constructed buildings—in other words, those built in compliance with the latest code requirements that call for elevating the home and designing it for 120 mph wind forces.45 These were few and far between because, after Hurricane Andrew, while stricter design requirements were enacted in the South Florida Building Code, the legislature did not impose those requirements along the entire state coastline, and certainly not in Florida’s panhandle, to minimize the impact on construction costs. This changed in 2007 after Category 3 Hurricane Ivan hit the panhandle in 2004. From that point on, the same building code was applied to all of Florida,46 but too late for Mexico Beach, as the code update benefited only those that built new homes thereafter, and only 20 percent of the town’s population lived in such new homes at the time of Hurricane Michael.
Very little attention was paid to those homes that more successfully “rode the storm” with minimal damage—as the media loves death and destruction more than successes.
However, one oceanfront home that survived particularly well—called the Sand Palace—became an overnight celebrity because it conspicuously stood out pristine in a wasteland. Granted that being a survivor in a neighborhood where nearly all surrounding houses have been obliterated and the beach dunes have been washed away is a success of mitigated benefit, it showed that it is possible to design buildings to survive a Category 5 hurricane head-on. This— and maybe the fact that the house had a sexy name to boot and its own Facebook page—caught the media’s eye.
It had been designed and built above and beyond the code requirements and splendidly survived the battering by wind and water. The attention to details it received in its construction was lauded by the popular media47 as well as trade magazines.48 The popular media was wowed by the fact that the owner had purposely required the house to be designed to resist 250 mph winds, had elevated it on top of piles driven 40 feet deep, and had used sacrificial breakaway walls at ground level that could tear-away without damaging the building—although it did rip away the stairs to the elevated house, such that it temporarily had to be accessed by a ladder. Some newspaper articles speculated that the cost to build this house had to be double that of a regular home—but one must never forget that journalists are ready to print any wild guess number they hear as long as it comes from the mouth of an architect they can cite, because citing a source is easier than fact-checking when under a tight deadline.
Beyond newspapers, trade magazines were more factual.
Companies that provided products that went into its construction focused on the benefits these products provided, while other professionals discussed the features that helped it survive, such as having a first floor fifteen feet above ground, small roof overhangs, limited number of windows, and reinforced concrete construction49—seveninch-thick insulated concrete forms and reinforcing bars, to be exact.50 The fact that the owners worked with a structural engineer to design the house to survive “the big one” is significant, as most homes are instead built following the minimum requirements of the building code, using empirical rules deemed to meet intent, without a tailor-made engineering design. Straight from the mouth of the horse, the structural engineer who designed the house indicated that the total cost premium for that house was on the order of 15 to 20 percent more on a per-square-foot basis51—a far cry from the more sensational “double the cost” previously reported.
Now, while much attention was paid to that building, and much emphasis was placed on the fact that its wall strength might have been greater than required by codes, it remains that the windows and doors were only certified to meet the code-specified 140 mph winds (as qualified by the projectile-resistant criteria), and that none of those broke during the storm.52 Keeping the wind outside of the building is winning half the battle, because once windows are broken, the roof is pushed up by the wind pressure entering the building, and pulled up by the wind forces outside uplifting it like the wings of an airplane. That is twice the pressure, and generally too much to ask, turning the roof into a kite. Indeed, small details make a big difference.
So why does all this matter? On one hand, it defines the hazard as it is, and on the other hand it defines the hazard as it is perceived—either magnified or demagnified. Among those residents of Mexico Beach who survived Hurricane Michael, some were determined to move out of the area and never come back, probably to never ever live again near an ocean shore, or even a lake or a river. They could very well end up, unknowingly, in areas prone to earthquakes, tornados, or forest fires, but that is another story—or another chapter. Yet, some will return and rebuild in Mexico Beach, and when they do, new construction will have to contend with significantly more stringent building requirements. For a start, the design wind speed will be 140 mph, as defined by the Florida Building Code.
Paradoxically, at the same time, across the state on the Atlantic shore, in Palm Coast, on the streets where the first floor of some bungalows turned into indoor salt-water pools because of the storm surge created by Hurricane Matthew in 2016, construction also resumed a few years later. In the aftermath of that hurricane, some of these new homes were built with elevated first floors, to prevent flooding in future storm surges, but surprisingly, some were bungalows at grade level and likely to be flooded in future hurricanes.
Given that Hurricane Matthew was a “near miss” that did not hit shore, resulting in a storm surge much smaller than it could have been, maybe it will take a hurricane the size of Michael to make a difference there and change the perception of what is a hurricane and what is needed to survive one.
In the meantime, try telling a homeowner in the Florida Keys sitting on a house deck so low that their feet dip in the ocean at high tide that this very house could someday be thirty feet under water. That homeowner will likely reply, with conviction, that waters around the Keys are too shallow for hurricanes and storm surges to happen.
ON THE DISASTER TRAIL
Hello Sandy, We Were Waiting for You A few years ago, I visited the New York State Office of Emergency Management command center in Albany, New York. It is effectively an underground nuclear bunker, with self-contained water supply, electricity generation, all the needed services, and enough food to keep an emergency response team alive for months. Since nuclear conflicts have been few and far between, the command center has been used for other emergency management purposes and for coordination of response to less radioactive disasters.
Rooms full of computers, overhead screens, direct phone lines to key government agencies and private sector stakeholders, and all kinds of other high-tech gadgets filled the command center. This was all impressive, but one thing that caught my eye during the visit was a poster showing expected inundation zones along the coast of Long Island, and New York City, in future Category 1, 2, 3, 4, 5 hurricanes. I eventually found a copy of that map and posted it on my office wall.
When Hurricane Sandy made landfall on October 29, 2012, near Atlantic City, New Jersey, with winds of 80 mph53 and storm surge inflicting $70 billion worth of damage along the coast of the Northeastern United States,54 it was sad to read and hear it called “the storm nobody expected,”55 “the Storm of the Century that no one believed would really happen,”56 and other similar names. There, on my wall, was the map that said it would happen, and when Hurricane Sandy happened, it flooded pretty much the exact same areas shown on the map.
Surprise is always a relative concept.

Flood
WATER RISING Water is the most important resource on earth—vital to human life. Yet, it is not advisable to say, “You can never have enough of a good thing” to people who have six feet of it in their living room.
Every little kid discovers early on that water flows down —and how fun it is to jump in puddles. Likewise, the same kids quickly realize that it is futile to push water uphill. The equation is therefore simple: rain, water downhill, and people uphill. Move up to stay dry. End of story.
If only it were that simple.
Worldwide losses due to floods are estimated at $40 billion per year1—including $8 billion per year in the United States alone. Most of that is predominantly due to the overflow of rivers due to heavy rains, melting of snow, or both at the same time—and not so much from earthquakeinduced tsunamis or storm surges along the coast during hurricanes, although those evidently add up to the bill every now and then. The size of this annual loss due to floods underscores how dif
f icult it is for people to stay away from the edge of water. It is not happening naturally, and will not happen easily. Being close to water is an ingrained attraction—and, often, a necessity. Any real estate agent can tell you that a property that has a bit of a shoreline is worth more than the landlocked one across the street—a lot more.
There is nothing mysterious about the “mechanics” of where flood-water comes from, so the mythological stories are typically not about what creates the rain, but rather about the wickedness of temperamental gods who control the on-off switch for floods or droughts—sometimes on a whim, sometimes out of anger. Indeed, if one is a god, and water is aplenty, what better way to punish people than to flood them. Water was the first weapon of massive destruction in the history of the world—or at least, in the mythology of the world. Zeus—the big honcho of the Greek pantheon—unleashed a massive deluge to destroy all of humanity, but Deucalion and his wife got a hint of his plan ahead of time and managed to float through it all, survive, and rebuild humanity one person at the time.2 In another region further southeast, Noah and his family did the same, but brought an entire zoo along for the ride.3 A bit more harshly, nobody survived when an Aztec deity flooded the world with fifty-two years of tears, but thankfully a couple of humans were resurrected from their bones to restart civilization,4 although in an alternate version of the myth, the couple survived the flood without needing to be resurrected5—after all, it was a while ago, so some details understandably get blurry. Likewise, Hindu gods, Mesopotamian gods, Chinese gods, African gods, and many others6 also flooded their part of the world, showing that gods are usually a well-connected fashionable crowd when it comes to unleashing disasters. Obviously, there were some more timid gods, apparently not part of the in-crowd, who only flooded to oblivion an island, like Poseidon who sank Atlantis—but apparently not for good, as it seems to have resurfaced in the twentieth century as a five-star vacation resort in the Bahamas.
FLOODPLAINS By definition, the areas that periodically get flooded when rivers get out of their bed are typically defined to be floodplains. Or, as the joke goes, an area is called a floodplain because it is plain obvious that it floods. The area of a floodplain is not of a fixed size because how much land gets flooded pretty much depends on the size of the flood.
For Noah’s proverbial deluge, the entire planet was apparently the floodplain. For more practical purposes, planners often refer to 100-year and 500-year flood zones, which means that lands within those zones statistically have a 1 percent and a 0.2 percent chance of being flooded each year.
For agricultural purposes, floodplains are desirable because the receding floodwaters leave silt and clay deposits that help make the soil fertile. Ancient agrarian civilizations adapted to the natural flood cycles and thrived because of it.7 The Nile River in Egypt provided such benefits until 1970 when the Aswan High Dam upriver eliminated the annual flood cycles, but created many other problems—such as forcing farmers to use chemical fertilizers instead, which in turn polluted the river, and longterm soil erosion in absence of the floods that used to bring new silt and clay deposits to replenish the shores.8 Modern city dwellers who love (and pay a premium for) their river view are less thrilled by floods. Rare are the riparian residents and businesses who benefit from ten feet of muddy water flowing through the ground floor. When residents of a floodplain wish to take action to keep their land dry, they typically build levees. As a result, all the water that would normally fill a valley many miles wide is channeled between the walls of the levees, rushing through that artificial corridor at greater speed to flood those downstream that are without levees. The cycle of levee building then repeats itself downstream, until the entire river is surrounded by levees. For example, there are currently over 3,500 miles of levees on the Mississippi River and Tributaries system9 (for comparison, the Great Wall of China is 3,889 miles long,10—although China prefers referencing its own State Administration of Cultural Heritage’s study which reports it to be 13,170 miles long,11 or 53 percent of the earth’s circumference of 24,901 miles).12 Unfortunately, as water levels rise higher and higher, the weakest links are the lowest levees. During the Great Mississippi Flood of 1927, 145 levees were breached along the river, flooding more than twenty-seven thousand square miles (“widening” the river to eighty miles in some locations), leaving five hundred people dead and seven hundred thousand homeless, producing $1 billion in damages (flooding the same area nowadays would produce $1 trillion in damage),13 and—most importantly—inspiring the song “When the Levee Breaks,” composed by Kansas Joe McCoy the same year, and popularized decades later by Led Zeppelin.14 Yet, contrary to the song’s lyrics, the good folks that survived returned to their land once the water receded and the mud dried up.
Many of the levees across the United States at that time were privately owned and originally built to protect farmland from flooding,15 so maybe it was not surprising that they failed, but even state-of-the-art, top-notch, best-of-the-best, government-built levees can fail when flood levels exceed the levels considered in their design. As unambiguously stated by FEMA: “While levees can help reduce the risk of flooding, it is important to remember that they do not eliminate the risk. Levees can and do deteriorate over time and must be maintained to retain their effectiveness. When levees fail, or are overtopped, the results can be catastrophic. In fact, the flood damage can be greater than if the levee had not been built.”16 In short, all levees can (and do) fail.
It is estimated that there are one hundred thousand miles of levees in the United States, located in 22 percent of the nation’s 3,147 counties (the exact number is unknown).
Approximately 43 percent of the US population live in these counties, although, evidently, not everybody in each of those counties is at the same elevation and thus exposed to the same risk.17 The US Army Corps of Engineers is responsible for roughly 10 percent of the nation’s levees, protecting ten million people. Beyond that, maintenance or improvement of levees is the responsibility of their respective public or private owners, which can be federal, state, or local entities.
Breached levees allow water to return to the floodplains in a less than desirable manner. When crops planted behind the levees become inundated, the cost is borne by taxpayers, because the US Department of Agriculture provides a Federal Crop Insurance that subsidizes farmers in the occurrence of such losses.18 Beyond damage to crops, when residential areas are flooded, the losses are substantially more significant than in rural areas. These losses are covered by the National Flood Insurance Program, a federal government program that carried over $20 billion in debt in 2018, in spite of Congress having previously canceled $16 billion of debt in 201719—therefore, a cost partly borne by taxpayers.
When levees are raised higher to reduce the risk of flooding, that often is done by the Army Corps of Engineers, thus also a cost borne by taxpayers. It is estimated that it would cost $100 billion to repair and rehabilitate US levees to an acceptable level—expensive, but cheaper than doing nothing and waiting for the floods.20 Therefore, at its core, the flood protection problem revolves around the issue of what is the desired protection level. In other words, for how rare of a flood should the defenses hold? Should taxpayers’ dollars flow before or after the waters flow?
CAN’T STOP THE RAIN Millions of citizens live in a 100-year floodplain. The boundaries of such a plain are determined for the average size of a flood having a 100-year return period. This implies that there is only a 1 percent chance per year of being underwater. At first glance, it does not look so bad, but this is a statistical delusion. Beware of the real estate agent who confidently tells potential buyers that one hundred years is a long time away. Not only can a 100-year flood happen tomorrow, it can happen multiple years in a row21—like getting tails when flipping a coin is not a guarantee that the next outcome will be head,22 or like getting one divorce is not a guarantee that the next marriage will be divorce free.
Those puzzled by the concept that a 100-year flood can happen multiple times within any 100-year period are the same people who are easily fooled by statistics—which is pretty much almost everybody, as will be shown later. In truth, compared to the return period used for many other hazards, a 100-year return period does not provide a very significant protection. This may have been considered acceptable by whoever drafted the flood maps years ago because being wet is generally not a deadly condition— except for drowning—but the financial losses can be significant. Again, like a divorce.
For example, in 1973, the waters of the Mississippi River reached the 100-year flood level; only twenty years later, in 1993, an even larger event occurred, with more than one thousand levees failing or overtopped,23 flooding fifty thousand homes and seventy-five communities over sixteen thousand square miles (other estimates report thirty thousand square miles),24 drowning forty-eight people, and producing more than $15 billion in damage25 over a fourmonth period.26 Then, the great Mississippi River Flood of 2011 that followed, again roughly twenty years later, is also considered to be one of the most damaging in history, with tens of thousands of floodplain residents evacuated over seven states.27 After that, some real estate agents are probably telling potential buyers that since the 100-year flood has happened three times in the past forty years, this has taken care of three centuries of bad weather and that there is now nothing to worry about for the next 260 years (equal to 300 years minus 40).
How to cope with recurring floods of such magnitude?
Turning off the tap is not an option, since nobody can stop the rain (besides, doing so would deprive musicians of a steady source of income, since rain has inspired so many Billboard hits). Staying out of floodplains is a logical option, but it can be problematic. For example, prohibiting settlements in the Mississippi River floodplain would be equivalent to creating a “no-man’s land” of forty-seven thousand square miles, which is not appealing or practical.28 Yet not all rivers flood across so many miles like the Mississippi, and even then, people find it difficult to move away from their home.
Take the Chaudière River, for instance. The first dwellers to settle along its banks in 1736 were quick to discover its temperamental seasonal water levels—particularly when its frozen surface melted in the spring and ice floes, hitting obstacles in their downstream travel, created ice dams that blocked the flow of the river and made water upstream rise up fast and high. Those setting up the mail route between Boston and Québec commented in 1773 on how badly the primitive road was flooded, and two years later, flooding in October 1775 slowed down the troops of General Benedict Arnold’s problem-plagued expedition on its way from Boston to besiege Québec City.
29 Yet, it appears that nobody thought it worthwhile to redraw property lines to account for this annual event. Instead, in 1778, the locals came up with a brilliant solution to control floods: building a church dedicated to Saint-Anne to implore her to enlist divine protection against floods. A larger one was built in 1828, followed by a third one—in stone this time—in 1890.30 To no avail, as all three were built in the flood-plain and were repeatedly flooded—which suggests that coaxing saints into interfering with the higher powers that control weather is a futile endeavor. A dam completed in 1968 in the hope of improving things did not make a difference either. In some curious mix of resignation and pragmatism, at the first sign of thaw each spring, local residents empty their cellars and make sure that canoes and boats are ready to provide travel across town when the water level will rise.31 Flood after flood, the local provincial government provided financial compensation to those hospitable folks that welcomed the Chaudière River into their homes. Then, after the 2019 flood, recognizing that it took more than two years to process the 6,171 claims filed after the 2017 flood, the government eliminated the requirement that invoices be produced to justify expenses; instead, to speed things up and lighten the bureaucratic burden, it issued a list of standard amounts it would refund for specific items: $1,275 for a French window, $800 for a washer, $30 for an ironing board, and so on.32 Tired of providing financial relief to repeat victims year after year, in 2019, the government also decided to set limits to lifetime indemnification, and offered a $200,000 “buy-out” to residents willing to relocate outside the floodplain. While the government might not have thought through how this plan would work for an entire city, like Beauceville that regularly sees all of its shops, businesses, and institutions visited by flood waters, some owners of residential properties were takers.
33 That proposed Québec policy was essentially modeled on one that has been implemented in parts of the United States for decades. In fact, returning to the Mississippi for a moment, a 10thAnniversary Anthology of Stories of Hardship and Triumph published by FEMA34 proudly lists a couple of dozen examples of happy community planners and property owners who have benefited from federal and state programs that provide generous financial packages for people to move out of flood-plains. However, in spite of all the uplifting stories, this has been a slow and expensive process with limited success. The Harris County Flood Control District, in the Houston area, is believed to have done more such buyouts than any other county nationwide—a coastal flooding area, but a flooding area nonetheless for sake of illustrating the point. Over three decades, from 1985 to 2017, Harris County has spent $342 million to purchase roughly 3,100 properties. This is laudable, but beyond that, the county still had 3,330 more to purchase on its priority buyout list alone—which, in itself, is a relatively small percentage of the 69,000 properties in Harris County that were flooded during Hurricane Harvey.
35 The county’s acquisition program budget only allows purchasing about a hundred modest homes per year, so even for homes that previously got flooded up to the roof line, a buyout can still be decades away. Like the little kid in class raising a hand and shouting “me, me, me” but ignored by the teacher, those willing to relocate by taking advantage of this program are simply “not eligible” if they are not on the county’s priority list—but even those who are, at the current funding level, might have to wait centuries before being bought-out. Therefore, some of these owners instead elect to sell at a loss, often to private investors who perform quick cosmetic fixes and rent the property to low-income residents, shifting the exposure to hazards to those less fortunate.36 At the same time, floodplain or not, some folks simply will not sell. Again, as real estate agents will tell you, people everywhere love their waterfront properties. In fact, part of today’s problem stems from the creation of the National Flood Insurance Program in 1968. With the best of intentions, this program’s goal was to “reduce the impact of flooding on private and public structures . . . by providing affordable insurance to property owners, renters and businesses and by encouraging communities to adopt and enforce floodplain management regulations.”37 The real estate market was quick to latch on to the fact that this made it possible to build new residential developments in floodplains where no insurance company would have been willing to provide flood coverage before, simply because the federal government was now willing to pick up the tab in exchange for a low/subsidized annual premium. True to the law of unintended consequences, the availability of this flood insurance—from an underwriter that can never go bankrupt, to boot—therefore translated into a population increase of 5 percent in zones where the flood risk is high.
More than that, it was found to provide an incentive for people to rebuild at the exact same place after a flood—as long as they were willing to have fishes as roommates every now and then.38 Likewise, beyond the federal government, even the incentives of local governments are often counter to moving people out of the floodplain, because that often also means moving them out of the county, which translates into a loss of tax income to the county. To overcome this problem, some counties use federal dollars to buy out and demolish homes damaged by floods, and resell the land for the construction of homes on stilts or walls elevated above projected future floods levels39—which often means more expensive homes, and thus larger taxes collected. Isn’t it brilliant?
Definitely, not all Little Pigs are born the same.
On the positive side, flooding is a disaster that happens one drop at a time, usually leaving plenty of time to evacuate. Flash floods are the exception. These can be loosely defined as floods that happen so quickly that they catch people off-guard. In arid regions, a dry riverbed on a sunny day can be filled within seconds by a strong current full of debris launched by a downpour that occurred miles away. This is deadly to hikers who cannot escape a canyon when the flash flood arrives, as well as to drivers with the unrealistic expectation that an SUV can cross eighteen inches of raging waters.
LIVING IN A YELLOW SUBMARINE If one remembers the “water flows down” concept, it would appear odd to build with the intent to live below sea level— unless building a submarine. Yet, 26 percent of the Netherlands is reclaimed land below sea-level, and 21 percent of the 17.5 million Netherlanders live in such negative elevation zones,40 the lowest point being at minus twenty-two feet. Only 50 percent of the country has an elevation above three feet. It is not uncommon when driving along a road on top of a dike, to see water five feet below on the right, and cows in pasture fifteen feet below on the left. It is up for debate whether it takes a certain temerity, an optimistic faith in the future, a solid dose of risk denial, strong relaxing drugs (legal in the Netherlands), or all of the above, to lounge in one’s home sweet home while looking upward to watch boats floating on the other side of the dike.
For over two thousand years, Netherlanders have been building dunes, dikes, and seawalls.41 While some countries waged war against other nations to expand their territories —and the Netherlanders naturally got caught in a few of those too—the country mostly went at war against the sea.
And, as in all wars, they lost some major battles. Tens of thousands died when dunes, dikes, and other sea defenses breached.42 More than one hundred thousand in one event in 1530; roughly fourteen thousand on Christmas day in 1717; more than twenty-five hundred in 1953. Harsh lessons as they might be, and as much as one would not be surprised if people decided to move out to higher grounds after such disasters, instead, each time, the breaches were repaired and the land was reclaimed anew—in some cases, the dikes were moved even farther seaward.43 Actually, in spite of the current forecasts of sea-level rises, nobody is packing up and running away. On the contrary, Netherlands is fortifying its sea defenses and its population is still increasing and projected to continue doing so, even in the parts of the country that might someday become octopus’s gardens (in sing-along joy, in the shade).
Yet the Netherlanders are not alone in defying a destiny predicated on the laws of gravity. The classic US example that comes to mind is New Orleans. When the French established their Nouvelle-Orléans outpost there in 1718, they built it on a couple of slivers of land about ten feet above sea level, surrounded by swamps. These marches were not attractive to urban residents, but, given that an inch and half of rainwater can fall on any given day there (the daily rainfall record being thirteen inches), they served as a convenient outflow destination to all the drainage canals dug throughout the city to help dry the soil after each storm.44 The advent of steam power helped launch in 1895 an era of massive civil works to build a new drainage system, first to collect and pump out water faster, and then to expand the city by drying up the marshes. By 1926, fifty square miles of marshland had been reclaimed, and within decades, developers turned swamps into attractive suburbs of houses built on concrete slabs at ground level.45 However, true to the law of unintended consequences, pumping the water out after each rainfall and building levees to protect against seasonal flooding from the river itself, allows the sedimentary soil of the Mississippi delta (on which the city is built) to dry out and compact itself over time. Therefore, thanks to human activity intended to keep water out of former swamps that were originally more or less at sea level, New Orleans progressively started to sink.
By 1960, 321,000 people lived in areas that had sunk by up to seven feet below sea level.46 Nowadays, measures have been taken to slow down this subsidence, but the lowest point in the city remains at eight feet below sea level.47 The fact that large parts of the city could be flooded during a hurricane was a predictable outcome well known prior to Hurricane Katrina. In 2004, FEMA started to hold a series of emergency response simulations with state and federal of f icials using a Category 3 hurricane scenario that assumed that half a million New Orleans residents would be under ten feet of water, but several of those meetings were canceled in early 2005 because FEMA could not find in its budget the $15,000 needed to cover the travel expenses of the people it invited.48 Sure enough, when Hurricane Katrina made landfall in Louisiana as a Category 3 hurricane a few months later, in August 2005, some of the levees and floodwalls were breached and 80 percent of New Orleans and many neighboring parishes were flooded. Then, the FEMA “floodgates” opened and the agency provided a lot more than $15,000 in post-disaster assistance (public accounting experts can explain how governments can be broke and rich at the same time). The hurricane left behind $161 billion in damages and nearly two thousand deaths.49 Beyond the immediate post-Katrina response, Congress approved $14 billion in funding for the US Army Corps of Engineers to strengthen the levees and walls that surrounded New Orleans, such as to provide protection against a 100-year flood—in other words, leaving a 1 percent chance each year that all that work will not be suf f icient to prevent flooding.50 Furthermore, while this massive project was completed in 2018, barely a year later, the Army Corps indicated that, as a result of subsidence and sea-level rise, it projected that the level of protection provided by this $14 billion investment would actually fall below the 100-year flood level by 2023.51 Technically, once the risk to a property becomes greater than 1 percent per year, it may become impossible for it to be covered by the National Flood Insurance program, but it is fair to bet that there are enough motivated lawyers in the United States to make sure that such a technicality will not “hold water,” because it is the government that built these levees in the first place.
Maybe coincidental to the above situation, but most possibly thanks to brilliant statistical and mathematical models that would be fun to hear explained by technocrats, the “base flood elevation level” defined by FEMA as “the regulatory requirement for the elevation or flood-proofing of structures” in some New Orleans neighborhoods was set to be below sea level. For example, in parts of St. Bernard Parish—the very same parish that was completely flooded after levee failures during Katrina, the very same parish that bathed for weeks in five to fifteen feet of stagnant waters contaminated by a large oil spill, the very same parish where only a handful of homes out of the 26,900 were still habitable after the event52—the new base flood elevation level is two feet below sea level. Base elevation in some other neighborhoods was set to be as low as six feet below sea level.53 Possibly encouraged by these numbers, ten years after Hurricane Katrina, confident that twenty-three miles of floodwalls had been reconstructed, much of the very same St. Bernard parish above had been rebuilt and its population was back to forty-four thousand people—or two-thirds of what it was before Katrina. Most importantly, people were not prevented from rebuilding in the low-lying areas of the parish.54 The same is happening in many of the city’s other neighborhoods.55 It is a free country after all.
DOUBLE WHAMMY The California gold rush started when flakes of gold were found in a streambed near Sacramento in August 1848. Of all those lured by the dream of plucking a fortune out of water, a third made the trip by sea. The population of nearby San Francisco boomed from one thousand to twentyfive thousand in 1849 alone. Hundreds of ships that arrived in San Francisco in those days were abandoned by passengers and crews rushing to join the other miners.
Some of these boats were turned into hotels or used for storage, but most were sunk in the bay together with landfill as material to expand the city boundaries by reclaiming land from the bay.
56 As a territory ceded to the United States by Mexico in 1848 as part of the treaty signed at the end of the Mexican-American War,57 California was a patchwork collection of mission towns with a population of 7,300. Two years later, thanks to the gold fever, the first census counted 92,597 people58—plenty more than the 60,000 threshold required at the time to request statehood status, which it received in 1850 in a fast-tracked process.59 By 1853, at the gold rush peak, 250,000 “forty-niners” (as they were called, in reference to the 1849 boom) were at work in the goldfields—hardly any one of them getting rich, even though roughly $2 billion in gold was extracted during the entire ten years that the rush lasted when the easily accessible deposits could be exploited.60 In contrast, a century later, many Forty-Niners became quite rich and struck gold by winning five Super Bowls—the football franchise alone was estimated to be worth more than $3.5 billion in 2018.61 With a population of 379,994 in 1860,62 and not much future in gold, many became farmers. Turns out the Sacramento Valley where the gold rush started was part of a massive delta of the Sacramento–San Joaquin River. As the delta was mostly a freshwater marsh, it was easy to remodel the landscape. By 1870, enough levees had been built to reclaim 780 square miles of wetland.63 These early levees were destroyed by the great flood of 1862 that inundated 6,000 square miles of the Sacramento and San Joaquin valley for up to six months, in water up to thirty feet deep at some locations.64 Undeterred, higher and stronger levees were built, periodically breached, and rebuilt taller and stronger—and so on. Part of the problem lay in the fact that the top layer of soil in the delta is a thick layer of peat, which is essentially partly decayed plant material. In many locations, this layer is thirty feet thick or more. When the process of reclaiming the marsh-land was started in the 1850s, the land surface was at sea level and thus periodically inundated by floods and tidal movements.
Hence, the levees did not need to be very tall. However, once reclaimed and farmed, the land started to “sink” slowly and progressively, year after year—a phenomenon known as subsidence and recognized to occur for a number of reasons when peat is cultivated.65 As a result, the land progressively sank to lower elevations, but since water remained at the same elevation—it being sea level—the levees built on the sinking ground had to be raised.
Nowadays, some of the farmland in the delta is more than twenty feet below sea level.66 Today, California’s Central Valley has roughly 2,400 miles of levees, with 600 miles within the Sacramento–San Joaquin River Delta.67 Beyond all the usual matters that must be addressed when assessing whether levees can be breached by floods having various return periods, as mentioned earlier, here is a double whammy: levees can also be vulnerable to earthquakes. In 2011, the State of California Department of Water Resources considered that approximately 291 miles of levees in the San Joaquin River valley alone (effectively 75 percent of them all) were a high seismic hazard,68 which means that during an earthquake their failure is likely and of great consequences. In 2015, the Department stated, “the likelihood of a major failure of multiple Delta levees in the next 25 years was greater than 50 percent.”69 A video produced to illustrate the result of a simulation performed in 2009, as part of the Delta Risk Management Strategy, shows the likely flooding that will happen when multiple levees collapse in the delta during an earthquake.70 Four years earlier, the director of the Department of Water Resources mentioned to the California Senate that this could happen for as small an earthquake as a magnitude 6 having its epicenter near the delta.71 Beyond the fact that farmland and thousands of homes will be flooded, and the fact that rail transportation would be disturbed and that natural gas and oil pipelines would be ruptured, the truly big problem revealed by the simulation is the penetration of saltwater into the valley because it will corrupt the drinking water for millions of Californians living in the San Francisco Bay Area and in Southern California.72 It is estimated that it would take over a year to repair the water control network, and several years for the salt leached into the watershed to clear out naturally before its water becomes suitable again for irrigation and drinking. Some communities could find alternative water sources to ride out the disaster while others may become ghost towns for years until water service resumes.73 All that was in a simulation that was considered a plausible scenario, not a worst-case scenario.
To prevent such a dire outcome, a $15 billion project, called “California Water Fix and Eco Restore,” was proposed.
It consisted of two forty-foot-diameter water tunnels intended to pipe water through the delta and forward it directly to where it is most desired, but this and multiple other previous similar concepts were defeated by voters suspicious of who will decide where it is most desired.74 When it comes to the politics of who gets the water, nothing is simple in the desert states.75 As always, the simplest approach is to wait for the failures to happen first, and then see. As the saying goes, “If it’s not broke, don’t fix it.”
ON THE DISASTER TRAIL
Non-Volcanic Craters The Fallingbrook neighborhood, in what used to be called Orleans, in Ontario, Canada (totally unrelated to New Orleans, USA), was established on a flat plateau covered by odd-shaped craters. The thick clay deposits underlying this plateau (and plentiful fossils) testify to the fact that the area was underwater ten thousand years ago when the melting glaciers created the Champlain Sea at the end of the last glacial period.76 However, nowadays, short of another ice age, Fallingbrook is unlikely to be flooded by the nearby Ottawa River since the plateau is 150 feet higher than the river in elevation.77 To top it off, the deep craters there have nothing to do with volcanos. They are massive retention ponds that have been created as part of the city’s storm water management plan to collect excess overflow in the case of rare and extreme rainfalls—the type of storms expected to happen (apparently) once in a thousand years. For example, the Apollo Crater (the one in Fallingbrook, not on the moon) is roughly twenty-feet deep78 and nearly six hundred feet in diameter.
79 Most years, the craters remain dry. They provide perfect slopes for little kids’ snow sliding in winter and serve other recreational use for people (and dogs) in the summer. In the nearly ten years I lived in Fallingbrook, it only happened once that a torrential rain filled enough of the craters to make it possible for excited kids to jump into the temporarily-created pools—not a particularly bright idea considering that storm water runoffs carry oil, dirt, chemicals, lawn fertilizers, bacteria, and other urban pollutants.80 All of that to say that I never expected to be flooded when I lived in Fallingbrook—until the day it happened. The sky was blue and the craters were perfectly dry but the sewer pipes at the end of the street became partially clogged in a way that had nothing to do with rainfall. All the houses on our street ended up with a few inches of water in their basement. Not a newsworthy disaster, but an annoying mess, nonetheless.
How somebody managed to clog that sewer pipe is beyond understanding—and something one prefers not to try to imagine—but, if anything, it shows that it is important to carry insurance, no matter what the real estate agent said about floods.

Tornado Alley
DO THE TWIST Joe and Jane had originally left their home state to attend college in Los Angeles, lured away by the promise of yearround sunshine and beaches. They enjoyed a decade of warm and sunny weather in La-La Land—even though they spent more time parked on freeways, where nobody dances, than on the beach—but when one of those crazy wildfires that spread over tens of thousands of acres was stopped only a few miles from their brand-new residential district, it sparked their decision to move out of California. Sitting in a hotel room, miles away, watching firefighters in the distance struggling through the night to save their neighborhood, they realized how lucky they were to have been able to evacuate to safety. A similar wildfire in Butte County the year before—known as the Camp Fire 1—had burned for seventeen days, spread over nearly 240 square miles, destroyed more than eighteen thousand structures, and killed eighty-five people who did not learn of the fire or receive evacuation orders soon enough to join the twentyseven thousand others who managed to barely escape through one of the only three flame-surrounded roads leading out of danger.
2 Joe and Jane could deal with the threat of earthquakes, because “big damaging earthquakes never happened,” but after seeing dozens of wildfires in the news every year and the inferno in their own backyard, they felt it was time to return home, to Oklahoma.
Where a bungalow was all they could afford in the crazy Southern California real estate market, now in a state where homes cost four times less per square foot,3 they were looking forward to building the mansion of their dreams.
Pre-approved mortgage in hand, when they sat with the homebuilder to make their final selection of upgrade options, they discussed how to best use the remaining $5,000 in their budget for extras. One option was to build a safe room.
Safe rooms are typically small windowless rooms located in the middle of a house and designed to withstand winds of up to 250 miles per hour, flying debris, and windborne objects.4 Providing about five square feet of space per person—say, a room four by five feet for four people—it looks more like a safe closet than a safe room, but it is to be occupied for only a few minutes during the passage of a tornado. It is designed to be anchored to the ground slab— or better yet, the basement slab if there is a basement—and to protect the occupants from injury in a sturdy enclosure while the tornado shreds the rest of the house to pieces.
The same effect can be achieved by an underground cellar,5 which is typically more common in the farmlands as they can also serve as true cellars as well. The drawback of cellars is that, depending on whether or not early warning is received, one may need to run outside during the storm to reach the shelter, exposed to lightning, hail, extreme winds and dangerous debris, unless the cellar is accessible from inside the house (for example, some cellars are located in an attached garage). Builders/contractors and design professionals—and in fact, pretty much anyone else—can get instructions on how to build a safe room using documents made freely available by FEMA.
6 Having grown up in Oklahoma—right in the middle of tornado alley—Joe and Jane were fully familiar with what these shelters are and do. Their city’s website recommends that a safe room or underground cellar be considered for every residence, but this is not required by the building code or by any local or state ordinance across the United States. In fact, in Moore, Oklahoma, where four devastating tornadoes struck between 1999 and 2013,7 only 10 percent of the city’s homes have them.8 A safe room typically costs between $2000 and $10,000, depending on size.9 Joe and Jane did not need more than a minute to reflect on that. They looked each other in the eyes and knew exactly what the other one thought. Their priorities had always been clear. Without hesitation, they decided to use the remaining $5,000 to buy granite kitchen counters, with the hope that Joe would eventually, someday, build an underground cellar in the backyard as a part-time project when the family finances would allow it. Besides, what are the odds?
THE DNA OF A TORNADO Those who fundamentally believe that the world as we know it is at risk of collapsing if items from two different food groups touch each other, that eating beans is tantamount to murder because souls travel through beans while awaiting to be reborn, and that blowing out candles on your birthday cake is a guarantee that your wildest wishes will come true are likely to find this section offensive and can easily skip it without any major indigestion.
Some have joked that given the fact that “tornado alley” is smack in the middle of the Bible Belt, God must have a “twisted” sense of humor. However, while it is true that states in that Belt see more tornadoes than anywhere else, with peaks of more than 10 tornadoes per ten thousand square miles per year,10 corresponding to 96, 52 and 126 tornadoes per year in Kansas, Oklahoma, and Texas, respectively,11 it turns out that tornadoes happen quite frequently elsewhere too—that is almost anywhere in the United States, Southern Canada, Europe, South Africa, New Zealand, Australia, and large parts of Asia.12 In fact, the Freising (Germany) tornado of 788 might be the earliest one documented in Europe’s history. The first documented one in the United States dates back to 1671, sighted in Rehoboth, Massachusetts, and making one casualty.
13 It is only later in their westward march to settle the country that European immigrants would meet what some Native American tribes of the prairies called the “Devil Wind.” Tornadoes are violently spinning air funnels that connect the ground and the clouds, making for one of the most visually striking meteorological phenomena. So impressive that storm chasers have started to make a business out of taking passengers with them14—a relatively dangerous venture given that the path of tornadoes is wildly unpredictable and that some of these tornado tourists have already been killed.15 On the positive side, thanks to the chasers, pretty much everybody has already seen dazzling pictures of tornadoes from up close.
The size of tornadoes is measured by the Enhanced Fujita Scale, which ranges from F0 to F5.16 Like the Modified Mercalli Intensity Scale for earthquakes, it is the damage produced by a tornado that earns it a specific value on the Enhanced Fujita Scale. Wind speed velocities are inferred from this damage. An F0 tornado will rip out some gutters and break a few tree branches, corresponding to winds 65– 85 miles per hour—nothing to get excited about since most cities experience days when such wind speeds are reached, without any tornado. An F5 will throw cars, trucks, and train cars a mile away, and will level well-built residential homes, with wind speeds estimated to be in excess of 200 miles per hour—essentially, a wind storm on steroids. In the world of truly horrible Hollywood movies—and in that world only—an F5 can also suck sharks out of the oceans, carry them hundreds of miles over land, and rain them all over the place in an airborne gore-fest of bloody shark attacks, called a Sharknado.
17 Fortunately for both humans and sharks, F5s are rare occurrences. In the United States, there was an eight-year span between the 1999 Oklahoma F5 and the 2007 Greensburg, Kansas, one. A similar span of more than six years started in 2013 after the Moore Oklahoma tornado.
18 Given that tornadoes typically form in a few minutes, making it difficult to deploy measuring instruments at the right time and at the right place, and because the high wind speeds tend to destroy the measuring instruments, the science that explains the genesis of tornadoes is still a field in evolution.19 The ingredients required to generate a tornado (as for instance, a “supercell” thunderstorm) are known,20 and tornado warnings are issued based on these precursor signs—or based on sightings of actual tornadoes if need be for lack of earlier warning.
However, what matters here is that, on average, a tornado produces damage along a path one or two miles long and over a width of approximately fifty yards.21 Hence, while a damaging earthquake can affect an entire region, the path of damage from a tornado amounts to a small line drawn on a map (again, on average). As far as record holders are concerned, the longest path ever observed for a pack of tornadoes from the same supercell is 293 miles, while the largest path width is 2.6 miles (but at significantly lower wind speeds).22 The most deadly one was the Tri-State Tornado of March 18, 1925, which killed 695 on a four-hour rampage that crossed Missouri, Illinois, and Indiana, leaving a signature of debris 219 miles long and three-quarters of a mile wide.23
TORNADO-MAGNETS Considering the fact that only a small percentage of homeowners have found it to be a wise investment to install a safe room or cellar where they can bite their nails, sheltered, while a tornado blows away the rest of the house, there might not be much of a market for completely tornado-resistant houses in the near future. In other words, going beyond small tornado-resistant closets, making an entire dwelling able to resist the 200+ mph winds of an F5 tornado, while possible, carries a bigger price tag. Many new homes along the Florida coast are designed to resist 140 mph hurricane winds, which is equivalent to the wind speeds generated by an F2 tornado, but stepping up the technology to resist an F5 will likely require a bit more entrepreneurial “oomph” down the line.
Then, there are mobile homes. The National Severe Storms Laboratory of NOAA, using data from 1985 to 1995, reported that people were ten to twenty times more likely to die in a mobile home than in other types of dwellings during tornadoes,24 a trend that apparently had remained the same by 2007.25 While mobile homes can be designed to resist speeds of up to 110 mph in hurricane areas, winds speeds of only 70 mph have been considered in their design in most parts of the country—including in Tornado Alley.
26 As a result, their damage in tornadoes has been so extensive that some people believe that trailer parks are actually tornado magnets—a severe confusion of cause and effect. It is because of the above statistics that observers get the impression that funnels “sniff around” until they find their favorite food—that is, trailer parks—before touching down.
However, tornadoes are “equal opportunity destroyers” and will attempt to equally flatten everything along their path— only the strongest ones will survive, and trailer homes have a limited membership in that club. Tornadoes will more readily flatten a weak mobile home than a sturdier residential construction—so, in that sense, yes, trailer parks are fast food for tornadoes, but not magnets (another urban legend debunked!). However, what is more likely is that trailer parks are news media nmagnets following the passage of a tornado. News anchors love the sense of drama they project when broadcasting their reports standing in front of massive debris, and what better place to find destruction after a tornado than in a trailer park? A good case of lazy reporting, and nothing more.
Most unfortunately, though, in some states, people who live in mobile homes are less likely to recover from such a disaster because they are the most socioeconomically disadvantaged populations—or, in plain English, the poorest folks in town.27 It is a sad reality for those who live in trailer parks out of necessity. It is a matter of debatable priorities when people who could afford to do otherwise freely choose to live in a mobile home in tornado alley for sake of being able to buy a fancier car. Double whammy: they are often the ones who believe myths such as “opening windows will help reduce the wind damage”28—which should be now recognized to be a bad idea by those who paid attention to an earlier chapter’s discussion on how internal and external wind pressures detrimentally add up when windows break.
As for those brilliant minds who suggested that tornadoes could be killed using nuclear bombs to defuse their energy,29 a sarcastic slow clap is the only possible response.
ON THE DISASTER TRAIL
The Famous 1996 Twisters No matter how often I repeated that going to see tornadoes would be too scary, my son was determined. All my warnings fell on deaf ears and it was clear that the more I objected, the stronger became his resolve. A tornado is an understandable attraction. “The adventure of a lifetime” as advertised by storm chasers selling tours to tornado tourists willing to pay thousands of dollars for the promise of driving close to one or more twisters—one of the rare natural hazards that will kill everybody in its path and yet can be approached to within a mile30 for amazing souvenir pictures, and the occasional near-death experience.31 As one who has spent his life working with extreme events, I certainly could understand my son’s attraction with twisters.
In the end, this was a fight in my power to win, but sometimes parents have to pick their battles. Sometimes, bad decisions can be good ones on the strength of the lessons learned. “OK, you want to see Twisters, let’s go see Twisters,” I told him. It did not take long to see one, a roaring F5, with cars and cows among the twirling debris, and sure enough, as I had predicted, my son was scared to death. He buried his face in my shoulder, waiting for the noise to abate. Same reaction, tornado after tornado as most ten-year-olds would be expected to have done. After that, he never argued when we refused to allow him to see a PG-13 movie with scary over-the-top special effects and thundering THX sound.32 “Twisters,” released in 1996, grossed $500 million worldwide and scared plenty of little kids—and some adults too.

Sitting on a Volcano
SCHOOL OF LIFE Like all kids in Japan most early mornings, Sakurajima’s first graders and high schoolers walk along narrow sidewalks, in small groups, dressed in a school uniform that generally includes a matched backpack. What differentiates Sakurajima’s school kids from those in most other prefectures is that as a standard part of the uniform, either worn on their head or snapped to and hanging from a backpack strap, is a hard hat. What is also different along the sidewalk is that most bus stops are actually little concrete bunkers designed to potentially deflect the flow of lava coming from upstream should the Sakurajima volcano decide to eject more than ash on a given morning—as the helmet alone might not be sufficient protection when that day comes.
Sakurajima is not only a volcano, it is a volcano-made island, built-up as a typical volcanic cone by new lava piling up on top of old lava from prior eruptions. It also has the distinction of being Japan’s most active volcano. The 4,500 people who live on the volcano are in a far suburb of the city of Kagoshima that is separated from the volcano by three miles of water. That is not much of a buffer for those across the bay, but it’s an even bigger problem for those who live around the base of the volcano.
The Kagoshima city website provides information on the various dangers created by a volcanic eruption:1
Cinders, which are rocks up to three feet in diameter ejected from the crater and that come crashing down as far as a few miles away from the crater; Pyroclastic flow, which is a burning hot mix of rocks, ashes, and gases that rushes down the face of the volcano at speeds ranging from 60 to 400 mph (depending on the volcano); Lava, which is a slower-moving magma (molten rock) that ignites everything it touches and buries it at the same time; Ashes ejected into the air (blown up to fifteen miles away in the case of Sakurajima’s major eruptions) and falling like gray snow deposits, accumulating multiple feet thick; Debris slides of earth and rocks, created when rain loosens ash buildups on the face of the volcano; Moderate earthquakes that come along for the ride; Rockslides and tsunamis, triggered by these earthquakes; and Toxic volcanic gases, such as sulfur dioxide and hydrogen sulfide, emitted by the volcano.
The list is likely educating only visitors and tourists, however, as kids in Sakurajima need not be reminded. They have grown up on a volcano that has been erupting almost nonstop since 1955, with thousands of small explosions each year, some severe activity from 2009 to 2016, and ash clouds often rising a few miles above the tip of the volcano.
To boot, the local landmarks are a lava observatory2 and the stone door of a Shinto shrine partly buried in ashes and pumice,3 a reminder of the 1911 eruption. And, not to forget, they carry a helmet to school.
In the event that early signs make it possible to warn the island’s residents of an impending eruption, Sakurajima’s detailed evacuation plan4 identifies twenty-one designated ferry departure points to which they should converge. They are also encouraged to do so at once, given that nineteen of those are within the delineated danger zone—approximately three miles in radius from the volcano’s peak—considered to be within reach of pyroclastic flow and blasts once the eruption starts, and the other three are within reach of volcanic rocks shooting out of the crater at 350 mph.5 Note that rocks twenty feet in diameter were projected up to two thousand feet from the crater during the 1935 Mount Asama eruption,6 but smaller ones can reach greater distances, raining down from thousands of feet in the air, which can be equally deadly.
Even if ducking into a concrete bunker is possible, evacuation is definitely the sensible thing to do because nothing stops lava, and trying to predict which way lava will flow on a volcano is a little bit like making wagers on what side of the candle the wax will drip. If some paths have been established and seen to be recurring from eruption to eruption, the odds are good it could flow there again in the future. An aerial view of Mont Vesuvius in Italy clearly reveals where lava last flowed during the 1944 eruption,7 as vegetation has not yet taken hold on top of the solidified lava—the path of that lava flow has been well documented in scientific publications.8 Defying the odds, right at the tip of that flow, where the lava solidified and stopped its forward progress, the same aerial view shows a recently constructed mansion with two large swimming pools, across the street from a gourmet restaurant.9 Yet, these are odds, not guarantees. In fact, over the long run, volcanoes are cones for a good reason: they are building themselves up because lava flows all around and solidifies, adding to the volume of the cone. The more symmetrical the cone, the more the lava has been uniformly distributed all around. On that basis, if constructing a house on the slope of a volcano, situating it on top of the latest lava flow might be strategic, hoping that the next lava flows will go elsewhere to continue building up a nice symmetrical cone. Unfortunately, recurring flows from multiple eruptions are sometimes needed before a specific valley is filled with lava and future flows are redirected elsewhere, which may take thousands of years to accomplish—geology and humans definitely work on different timeclocks.
To put it mildly, settling on the slopes of a volcano is a dicey proposition, particularly since there is no truly proven way to divert molten rock that has been ejected at 1,300 to 2,000°F10 and that will flow downhill until it cools and solidifies.
Piling up concrete blocks alone will not work, since concrete (approximately 150 pounds per cubic foot) is typically less dense than lava (approximately 200 pounds per cubic foot) and will therefore “float” on it. With enough advance planning, the concrete could be tied down to counter its buoyancy, but concrete’s resistance to high temperature is not infinite. The moisture inside the concrete will turn into steam when its boiling point is reached, and if the thermal shock of hot lava hitting concrete is sudden, the pressure created by that steam could make the concrete explode11—unlike in a building fire, in which the slower temperature rise gives time for the steam to escape through the concrete’s matrix of materials. In addition, the properties of the Portland cement inside the concrete start to degrade at 1,500°F.
12 Special coatings can be applied to concrete to resist a rise of temperature to 2,200°F for ten minutes, but being in direct contact with lava creates a significantly faster thermal shock. Levees of rock and ash proved more successful in containing lava flow from the Etna eruption in 1983: a first levee 30 feet high, 100 feet wide, and 1,200 feet long was overtaken by the lava, but a second one built parallel to it 300 feet farther away managed to contain the flow.
13 Given that lava is molten rock, one approach that can be envisioned to block the flow of lava in a specific direction is to cool it down such that the resulting rock creates a barrier to the oncoming lava behind it, thus diverting the flow elsewhere. This strategy has worked to protect the Icelandic port town of Vestmannaeyjar during the 1970 eruption of the Eldfell volcano, where water cannons shot 1.5 billion gallons (12.5 billion pounds) of cold seawater onto a slow moving lava front over a five-month period, eventually redirecting the lava flow to protect the port.14 Unfortunately, not everybody is so lucky to have the perfect combination of snail-paced lava and an infinite supply of water, so this has been a unique success story so far. Furthermore, it should not be forgotten that the tradeoff to this success is that when 12.5 billion pounds of water hits lava, it turns into 12.5 billion pounds of highly acidic steam—enough to make a jolly bunch of clouds that will eventually drop 1.5 billion gallons of acid rain somewhere.15 Another approached tried in the past has been literally to bomb the lava flow, military style, in strategic locations, in an attempt to redirect the flow by using explosion craters and by destroying the tubes of hardened lava through which molten lava flows faster.
16 However, as with many losing wars, much damage was done to no avail, as the lava simply filled the craters and resumed its course—shockingly, turning out to be another world problem that could not be solved by a good bombing campaign.
Of course, in all of the few successful containment cases reported in the literature, it is understood that success was achieved partly because lava speed was only a few feet per hour and—most importantly—the volcanic activity that lasted months eventually subsided, which effectively “turned off the tap” and lava stopped flowing.
Unfortunately, that is not always the case. In particular, lava can flow at speeds of up to forty miles per hour.
17 Furthermore, all the diverted lava has to go somewhere else; while some critical location along the flank of the volcano might be spared, it could be at the expense of somebody else’s proverbial backyard.
Nonetheless, these are the risks one must be willing to accept when taking up residence on the slope of a volcano.
Sure, the view might be second to none, there may be tons of opportunities to cash-up if the volcano itself is a tourist magnet, the soil can be incredibly fertile for farming, and land for residential construction can sometimes be cheaper to buy there. However, it should not be forgotten that after an eruption, one might end up with more land than they paid for—if not in plan, at least in volume, thanks to all that cooled-off volcanic rock piled-up on top of whatever real estate once existed.
Take Vesuvius, for example. Volcanologists agree that Mount Vesuvius is not only active, but also overdue for a major eruption. Only 8.5 miles as the crow flies from its caldera to downtown Naples, the Vesuvius is world famous for its AD 79 eruption that killed many of Pompeii’s inhabitants and buried the entire town under more than twenty feet of stones and ash.18 Located six miles from the caldera, Pompeii’s population had prime seats to watch the volcano blow up a plume visible from hundreds of miles away (as recreated in some awesome YouTube videos).19 Those who recognized the danger had enough time to flee— more than 90 percent of the population apparently escaped in time.20 A few thousand remained, maybe mesmerized by the awesome show, or maybe diehard optimists reluctant to leave for the same reason that evacuation orders are often ignored nowadays. For them, all was fine until all that had flown up started to rain down. First came ashes, accumulating on the ground like gray snow a few feet deep, then rocks, which sent terrified people running to seek shelter in their homes. To no avail. What followed was a lot more ashes, making it difficult to breathe, and—the killer— the pyroclastic flow of 500°F poisonous gases rushing in at a hundred miles per hour. In a day, Pompeii was erased from the face of the earth, lost and forgotten for centuries, until it was accidentally rediscovered and excavations of its ruins started in 1748.21 Voids in the ashes left after the organic matter had decomposed served as molds that, when filled with plaster, revealed full bodies in the exact posture they had been in the instant before their asphyxiation— gruesome reminders of death by volcanic attack.
Vesuvius continued to erupt for over a thousand years thereafter, with some other major eruptions, particularly in 203, 472, 512, 685, 787, 968, 991, 993, 999, 1007, 1037, 1068, 1078, and 1138. Then, after a period of quiescence of roughly five hundred years from 1139 to 1631, Vesuvius woke up and has regularly flexed its muscles with eruptions in 1631, 1660, 1682, 1694, 1698, 1707, 1737, 1760, 1767, 1779, 1794, 1822, 1834, 1839, 1850, 1855, 1861, 1868, 1872, 1906, 1926, 1929, and 1944—that last one actually caught on film.22 On average, one eruption every thirteen years. Since 1944, nothing major—which is not necessarily a good thing, since it is expected that the longer it takes before the next eruption, the more severe it will be.
To make matters worse, no other volcano on earth has as many people living within the danger zone around it.23 Including the city of Naples, the surrounding urban area is home to more than three million people,24 including six hundred thousand living within the “red zone” who would have to be evacuated. That task would take roughly a week to accomplish, relying on 500 buses and 220 trains25—and, evidently, a truly optimistic outlook on Italian punctuality.
Obviously, whether such a mass exodus can be successful is contingent upon scientists being able to predict reliably, and with reasonable advance warning, when a major eruption will occur—something that is still far from an exact science.26 Of those living in the red zone, 97 percent are fully aware that the area is within a zone of high volcanic risk, and 61 percent recognize that the presence of the volcano makes it a “hostile place to live” and that they could be displaced by future eruptions.27 Logical numbers, given that Pompeii is an eloquent reminder, smack dab within that red zone. Yet, in spite of an Italian government program giving 30,000 euros to all those willing to abandon their home and relocate elsewhere, not many have moved out— and some squatters have moved in.28 So, why do people live there? Because living in proximity of the volcano seems like a normal condition. Why leave?
Nothing has happened since 1944, so life goes on. Maybe divine protection will suffice. Besides, as someone said, living on a volcano makes it perfectly fine to smoke and drink with abandon.29 Yet Naples is not alone in hugging a volcano. Many large cities border the fifteen hundred known potentially active volcanoes on earth,30 including populated areas in Mexico, Chile, Bolivia, Ecuador, Peru, Spain, the Philippines, Indonesia, Papua New Guinea, New Zealand, the United States, and Japan.
For example, Mount Fuji’s peak is sixty-two miles from the Tokyo city center, but only thirty-five miles to the western edge of the Tokyo-Yokohama region. This most iconic Japanese landmark has become a symbol of Japan, captured in multiple work of arts over centuries, such as in Hiroshige’s famous series of woodblock prints Thirty-Six Views of Mount Fuji which nowadays adorn T-shirts, coffee mugs, and everything else that can be bought in the country’s countless tourist traps.
In the dreariness of the urban pile up that comprises the country’s capital, residents there take pride in being able to see the beautiful, snow-capped Mount Fuji on a perfectly clear winter day—which practically almost never happens.
Every year, when all snow has melted (mostly in July and August), three hundred thousand people hike up to the top along one of four well-traveled routes, starting from parking lots at different heights from the base. Most hikers climb at night so as to be at the summit in time for sunrise, which can make for massive pedestrian congestion reminiscent of a Japanese train station. As the Japanese proverb goes: “A wise person will climb Mt. Fuji once; a fool will climb Mt. Fuji twice.” The amount of garbage31 and feces32 left on the mountain by all trekkers makes some wonder if even a once-in-a-lifetime pilgrimage up Fuji is actually that wise.33 Nonetheless, climbing or not, the Fuji Five Lakes area around the base of Mount Fuji is packed with resorts, spas, golf courses, ski stations, and even a roller coaster amusement park, all of it attracting nine million visitors every year.
34 This is possible because the volcano has not erupted since 1707.35 It is considered active, but dormant.
As it turns out, Mount Fuji might be due for a good show soon. Every sleeping giant must wake up and stretch every now and then, and when that happens, things get interesting.
A number of volcanologists over past decades have warned that an eruption is overdue,36 including some from Mount Fuji’s very own Research Institute,37 who have reported that volcanic activity has been abnormally insignificant in Japan in the past century compared to what has been recorded in geological history. They underscore that Fuji, which used to erupt once every thirty years but has been dormant for the past three hundred, might be about to make some trouble, because an eruption after a long dormancy has the potential to be highly explosive.
Predictions of Fuji’s impending eruption have been leveraged by recent studies that have recorded an increased level in the volcano’s seismic activity; this was blamed on the 2011 magnitude 9 Tōhoku earthquake, which was believed to have increased the pressure building up inside the volcano38 up to a value estimated to be greater than what caused prior eruptions.39 Studies have investigated what would be the consequence of an eruption today. Focusing only on the amount of ash ejected by the volcano, considering over a thousand different scenarios of eruption scale, wind directions, and air pressure to calculate ash dispersion, it was found that the next eruption of Mount Fuji would blanket the entire Tokyo area with somewhere between three inches to three feet of ash. Any of those scenarios would bring the metropolitan area to a standstill. Roads and railway operations would shut down, water drains and ventilation units would be clogged, and exposed machinery would jam. Failures would further cascade from there.
Massive power outages would be expected from ash clogging the intake filters of gas turbines or due to short circuits from rain falling on ash accumulated on power lines.40 There seems to be no estimates of how long it would take to shovel away all that ash, or where it could be dumped. As for lava flows, if any, plans estimate that threequarters of a million people would have to be evacuated from the vicinity of the volcano—which would obviously have to happen before too much ash piled up in the roads.
When the volcano last erupted in 1707, it killed twenty thousand people,41 at a time when the entire population in the Kai and Sagami provinces around the volcano was around six hundred thousand.42 Today, the Tokyo-Yokohama region is a fifty-mile-wide spread of entangled urban development with a population of approximately forty million.43 Average population density is roughly seven thousand persons per square mile, peaking at forty thousand people per square mile in the downtown areas. It is both the world’s most populous metropolitan area and the largest metropolitan economy.
44 It is not known if studies have predicted the expected number of deaths this time around.
On the positive side, this study has apparently prompted Japanese official to start drawing up emergency response plans for such a situation.45 This sounds like a good idea.
With more than 110 active volcanos in Japan,46 there is plenty of expertise around to tap; some phones will likely ring near Sakurajima.
THE DNA OF A VOLCANO Those who believe that fruit-flavored candies and soft-drinks are legitimate portions of a healthy diet’s daily allowance of fruits, that fireworks on the 4th of July celebrate the United States of America gaining its independence from France, that Titanic is a great movie about a mythical boat that never existed, and that laser beams can be cut into threefoot-long lightsabers, might find this section offensive and can easily skip it without disturbing the space-time continuum.
Earthquake duration is counted in seconds. Hurricanes and tornados are only wind—lots of it, but only wind, nonetheless. The pyrotechnics of a volcano though, that is quite something else: the smoke, the rumblings, the eruption, the fire, the lava—and the damn thing grows, from eruption to eruption, asserting itself, and creating new land.
No other natural hazard packs as much “shock and awe.” No wonder that in many civilizations, it was the same god that controlled both volcanos and fire—like Pele, goddess of volcanoes and fire and creator of the Hawaiian Islands.47 After all, by definition, a volcano is the chimney of Vulcan’s forge, where the gods’ weapons of war were created.48 Until the development of plate tectonics, nobody understood by what mechanism all that lava from within the earth could explode to the surface. Why would a volcano sleep for centuries and then, out of the blue, violently erupt?
For a long time, volcanoes were thought to reach to the center of the earth. So, logically, in Journey to the Center of the Earth, Professor Otto Lidenbrock’s 1863 expedition to that destination started in the volcanic tubes of the sleepy Iceland giant, Snæfellsjökull. However, contrary to what is vividly described in Jules Verne’s novel,49 no respectable nineteenth-century volcanologist believed that dinosaurs and giant mushrooms would be encountered on the way to the earth’s core. Artistic license prevailed nevertheless, since nobody had ever gotten close enough to the center to contradict good old Jules.
At best, scientist have barely scratched the surface of the earth. The distance to the center of the earth is roughly 3,959 miles, give or take a few, depending on if one stands on top of Everest (5.5 miles above sea level) or at the bottom of Challenger Deep (2.3 miles under the sea). By comparison, the Kola Superdeep Borehole, which is nine inches in diameter and took twenty years to drill, reached down 7.5 miles and stands as the deepest manmade hole to date.50 For the record, a couple of recent oil wells in Qatar and on Sakalin Island (offshore Russia) are a few hundred feet longer, but do not go as deep in terms of true vertical depth,51 because oil drilling can be done at different angles.52 Looking forward, an ongoing billion-dollar scientific project, using a specially rigged boat to drill into the earth’s crust from the bottom of the sea, is expected to go a little bit deeper—which is literally a sophisticated way to throw money into a hole.53 While poking a few holes here and there over the next centuries will greatly enrich knowledge on the composition of the earth, in the meantime, plate tectonics can be used to explain the existence of volcanos. Given that lava is magma that reaches the surface and that magma is molten rock, what better place to melt rock than at the interface between two tectonic plates exerting great pressure when rubbing on each other. Subduction zones, which are where one tectonic plate is pushed under another,54 are perfect settings to create magma, particularly at greater depths, where temperatures are higher. That remains the best explanation so far, and a compellingly logical one. At least while awaiting a robotic descendant of Professor Lidenbrock that could dive into the lava and navigate down the volcano’s chimney, which probably will never be possible.
So far, one wild soul wrapped in special protective gear made it down into the crater of the active Marum volcano in the South Pacific, so close to the edge of the gurgling lava that micro-splashes of it melted bits of his GoPro cameras, but he did not dare jump all in.55 Beyond the fact that humans can travel to visit a volcano, it is also possible for a volcano to come and visit humans—sort of. In Paricutín, Mexico, on February 20, 1943, a farmer saw a half-meter-deep crack in the ground of his cornfield suddenly swell into a five-foot-tall mound blowing sulfur-smelling smoke, ash, and semi-molten rocks. Within a day, the mound grew into a 150-foot-tall mini-volcano.
Within a week, it was 300 feet.56 After a year of activity, the cornfield had become a 1,475-foot-tall cone that buried the village of Paricutín.57 If any corn was left, it was probably popcorn.
Similarly, a geyser suddenly popped out of a backyard garden in Rotorua, New Zealand. In due time, the geyser gazers who came for the show saw it turn into a bubbling mud crater that grew and engulfed the garage next to the bungalow.
58 That is not quite a volcano, but another one of the potential hazards that cannot be ignored when living in a geo-thermal area.
VOLCANIC DOUBLE WHAMMY Historians may debate whether it is because cunning Vikings wanted to trick enemies with an eye on those lands that they called the country covered by ice over 80 percent of its area Greenland and the other one nearby with only 11 percent of its surface frozen Iceland.59 In any case, a more fitting name for the latter would be Fire-and-Ice-land.
Beyond being famous for its geothermal energy and worldrenowned hot springs, which attract flocks of tourists, Iceland is also home to thirty active volcanos, and one hundred more believed to be inactive.60 The magic of mixing fire and ice is that it makes it possible for volcanoes to erupt from under a glacier, thereby melting a significant part of the glacier. Eruptions too small to break the surface of a glacier can still melt lots of ice from its base. That water seeps under the glacier and eventually bursts out somewhere as a surprise flood—many Icelandic villages have been wiped out that way.
61 Scale-up the eruption, scale-up the flood. In pure Icelandic tradition, these unpredictable glacial floods are given an unpronounceable name: jökulhlaups.
When eruptions punch through the glacier, Iceland gets the full double whammy of a full-blown volcanic eruption— with ash, lava, bombs, and so on—and a menacing jökulhlaups.
During the April 2010 eruption of the Eyjafjallajokull volcano (yep, another tongue twister), punching through the glacier and melting much ice around it, the biggest fear in far distances from the volcano was that major roads and bridges would be washed away; also, people in the upcoming path of the flood were evacuated.62 Destroyed roads and bridges were exactly the consequences during the September 1996 Vatnajökull eruption,63 which punched through a 1,500-foot-thick glacier and sent water downhill at a flow rate of 45,000 cubic meters per second,64 or twenty times that of Niagara Falls.65 Luckily, nobody lived near Vatnajökull. As for Eyjafjallajokull, whose massive jökulhlaup was caught on video,66 the engineers had time to dig trenches through the causeway of a threatened national highway to divert the upcoming floodwaters through those channels to protect a bridge—as it is cheaper to rebuild segments of a causeway road leading to a bridge than the bridge itself—and it worked.
THE END OF THE WORLD In a special class of their own are the supervolcanoes. They are a sort of “end-of-the-world” type of event, best explained by making some comparisons.
Much like the magnitude scale for earthquakes, the Volcanic Eruption Index (VEI) measures the severity of eruptions on a logarithmic scale ranging from 1 to 8, considering how much and to what height material is ejected, and how long the eruption lasts. At mid-range of the scale, a VEI 5 eruption projects between one and ten cubic kilometers of volcanic material,67 from ash particles that will eventually mix with clouds and come back down as acid rain to large molten rocks that crash to the earth like bombs. One cubic kilometer is 0.24 cubic miles, or 1.3 billion cubic yards. This means a one-inch-thick coat of debris over a surface of 15,000 square miles—or one foot thick over 1,250 square miles. The VEI 5 Mount St. Helens eruption in 1980 scattered trace amounts of ashes over eleven states and five Canadian provinces. By comparison, the 2010 Eyjafjallajökull eruptions that shut down the European airspace for five days was only a VEI 4. On a logarithmic scale, that is ten times smaller than a VEI 5. VEI 6 eruptions eject so much ash into the atmosphere that it typically affects world climate for a few years. For example, because of the 1883 VEI 6 Krakatoa eruption, summer temperatures in the Northern Hemisphere dropped by 2.2°F and it took three years for things to return to normal.68 The domain of supervolcanoes is in the VEI 7s and 8s range. There have been ten VEI 7 eruptions in the past ten thousand years, and four in the past two thousand years.
The more recent one is the 1815 Mount Tambora eruption in Indonesia, which disturbed the world climate for a few years —to the extent that it snowed in Albany, New York, on June 6. Crops were ruined in parts of North America and Europe, leading to famine in a few countries.69 And that’s only VEI 7.
VEI 8s are the truly dreadful mega-monsters. One thousand times more powerful than the Mount St. Helens eruption. The devastating impact of a supervolcano can be as significant as that of a fifty-mile wide asteroid colliding with earth—keeping in mind that the fifty-mile wide rock that hit the Yucatan peninsula 66 million years ago projected so much dust in the air that it has been blamed for the extinction of dinosaurs. In fact, two theories70 have been proposed to explain this massive extinction: asteroid collision or a massive burst in volcanic activity. Both phenomena have the ability to fill the skies with gigatons of particles that end up blocking the sunlight that plants need and cooling the planet, a double whammy that—much like with the Krakatoa eruption but on a grander scale—starves the planet through years of darkness.
The occurrence of supervolcanos is not unusual from the perspective of the earth’s geological time scale, but from a human perspective, the odds are good that one thousand generations could squeeze between two events. The last VEI 8 occurred 26,500 years ago, at the Taupo Volcano in New Zealand—obviously, there are no photos or paintings of that event. Known as the Oruanui eruption, it ejected 1,170 cubic kilometers of rock, ash, and lava, with deposits found more than five hundred miles away. The one before that, seventy-four thousand years ago, in Indonesia, ejected twice as much material, standing as second place record holder in that category.
71 The United States has many competitors in that category. Although not a medal holder, the Yellowstone caldron has exploded three times in the past 2.1 million years,72 with the last eruption 640,000 years ago. Statistically, if the Yellowstone eruptions were taken to occur like clockwork, as Old Faithful almost does, that would leave another seventy thousand years before the country is significantly remodeled by such geological forces. The gold medal apparently goes to the eruption of La Garita Caldera, which is believed to be the largest volcanic eruption ever, estimated to have been so large that it has prompted volcanologists to add a level to the VEI scale, just for it—and they assigned it a VEI of 9.2.73 Erupting twenty-seven million years ago in southwest Colorado, it ejected five thousand cubic kilometers of materials. If that amount of material could be evenly spread over a geographic area, it would be enough to cover the entire contiguous United States with two feet of debris—not to forget the impact on climate from all those ashes obscuring the sky.
It is hard to find a worse way for an entire civilization to die. Short of a zombie invasion, there is probably not a more painful way to reach the end of the world than a supervolcano eruption.
ON THE DISASTER TRAIL
Dust in the Wind Dormant volcanoes can be sleeping giants, picturesque and great for hiking, but for someone like me who thinks that visiting volcanoes is an absolutely essential thing to do, active volcanoes are more fun. Most of them are national parks or official tourist attractions, each with their own special character that makes them uniquely enjoyable.
The Vesuvius observation area allows a peek into a steaming cauldron that is relatively peaceful and, all things considered, immensely less dangerous than the drive to it from Naples—the road is safe, but the Italians drivers are “completamente ignorante delle regole del traf
f ico.” The
drive to Mt. Aso Nakadake Crater, in Kyushu, Japan, is more rewarding, as far as “volcano tasting” is concerned. The large parking lot located along the rim can be filled with school buses, but the kids run away from the crater observation area at the first whiff of sulfur. Multiple signs warn that those with asthma and other breathing problems might be at risk, and the observation platform is closed when unfavorable winds engulf it in the toxic sulfur dioxide gas emitted from the crater lake two hundred feet below.
Over the years, a number of people have died from volcanic gas inhalation, and many more have been severely injured.74 Every now and then, the volcano also erupts.
White Island in New Zealand provides an even more intimate encounter with the lethal chemistry of volcanos. A ninety-minute boat ride away from the coast, White Island is a desolate landscape in a geothermal frenzy. After having signed a waiver acknowledging that eruptions are unpredictable and that we assumed the risk, our tour guide provided us with a gas mask along with a stern warning that anyone stepping outside the marked trail risked punching through an unstable crust and landing in an acid bath. The trail meandered between bubbling mud pools, steam vents roaring more furiously than locomotive chimneys, hissing hot volcanic streams, and an acidic crater lake known to instantly dissolve anyone falling in (this fun fact was probably discovered the hard way). Every now and then, White Island erupts too, as it did on December 9, 2019, killing twenty-one tourists and injuring twenty-five more with severe burns.75 Yet, of all my volcano visits over the years, the “quaintest” one—if volcano visits can ever be described as such—was an overnight stay on Sakurajima Island. Given that Sakurajima is the most active volcano in Japan, there is no public road to its crater, but there are roads and villages all around its base. We had elected to stay overnight in a Minshuku, which is the Japanese equivalent of a bed and breakfast. It was a traditional Japanese home, the hosts were charming, and conversation was limited to clumsy sign language, but sleeping at the base of an active volcano was an eerie experience—no waiver of responsibility was necessary to sign in this case, as this was normal everyday life as far as our hosts were concerned. The “icing on the cake” of this volcanic experience came in the morning, as we left and discovered that our car had been covered by a sheet of ash during the night—courtesy of an active volcano. Many times before, I have had to “dust away” snow from my car in the morning, but this time, the “dusting away” was as dusty as it gets.


Technological Disasters
NO SAFE PLACE ON EARTH Homer Simpson works at the fictitious Springfield Nuclear Power Plant, a facility that has received hundreds of safety violations and has averted multiple total meltdowns by sheer luck alone.1 Homer is an absolute, unabashed, incompetent safety inspector who would not be woken by flashing red alerts and would not see anything wrong in using fuel rods as paperweights. However, while Homer is a cartoon character, there had to be a real Homer Simpsonov operating the Chernobyl nuclear power plant when one of its reactors exploded in 1986 during a safety test gone awfully wrong. In fact, there had to be an entire cohort of Simpsonovs along the entire chain of command, including the engineers who designed the flawed nuclear reactor,2 the poorly trained plant operators who had turned off the automatic shutdown mechanism the day before the explosion,3 the officials who waited a day before issuing an order to evacuate the fifty thousand people in the neighboring town of Prypiat twelve miles away, and all the top-level comrades who downplayed the event for as long as possible.4 All of them contributed to the resulting mess.
How else to explain why the USSR kept quiet about the incident, if not for fear of utter embarrassment?
While the communist apparatchiks acted as if nothing had happened on the day the plant blew up, and the stateowned media devoted a full twenty seconds during the evening news to mention that a minor incident had happened (giving it less attention than the weather forecast),5 the world discovered that something had gone terribly wrong behind the iron curtain when radioactive rain came down in Sweden two days later.
6 The steam from the explosion and the fire that burned for a week before being contained, spewing radioactive clouds into the atmosphere, not only made Chernobyl the largest nuclear disaster in history—until the Fukushima Daiichi nuclear disaster twentyfive years later7—but it also arguably played a significant role in bringing down the USSR six years later.
8 At least, it was perceived as such by Mikhail Gorbachev,9 general secretary of the Communist Party of the Soviet Union at the time of the disaster. To its people, raised on the propaganda that the USSR was the top world super-power, best at everything, with truckloads of Olympic medals to prove it, the widespread incompetency at managing a public health crisis exposed the fallibility of their government and rapidly eroded whatever confidence they might have had in its institutions.10 Three decades later, for a mere 3,300 Ukrainian Hryvnia (roughly US$130), tourists were able to visit Chernobyl by booking a tour with the state enterprise in charge of decommissioning the nuclear power plant or with other organizations that marketed Chernobyl tours as an “eye opening experience of post-apocalyptic world.”11 One needed to sign a waiver releasing the tour organizers from “any damages caused to the health, property of the visitor” from a list of possible causes, including “the effects of radiation and other harmful factors,”12 and then, off to the tour—not exactly Disney World, but there was apparently a market for it.
Whether or not the Chernobyl incident has had an impact on the nuclear power industry has been a subject of debate,13 but there is global agreement that the 2011 Fukushima nuclear incident did. In the aftermath of the Japanese disaster, the industry faced significant project delays and— in some cases—roadblocks. The difference is that, in 2011, it was more a case of nature exceeding design assumptions than problems created by a bunch of Homerima Simpsonakis asleep at the switch.
When the March 11, 2011, magnitude 9 Tohoku earthquake struck 110 miles away from the Fukushima Daiichi Nuclear Power Plant, the ground accelerations recorded by instruments at two of the six nuclear reactors on site exceeded the values considered in the design of the plants by about 20 percent. The plant was subjected to two minutes of strong ground motions,14 but this did not create an issue because, unlike buildings where damage is expected if the earthquake considered in their design occurs (as will be described later), nuclear power plants are designed to remain undamaged for the considered design forces, for obvious reasons, as damage to them can have severe consequences.
As soon as seismic activity was detected by Fukushima’s instruments, the reactors were shut down. Emergency generators powered by diesel fuel kicked in when power from the grid was lost due to damage to nearby transmission lines. So far, it all played out by the textbook, and most of the world would never have heard the word “Fukushima” if the impact of the earthquake had been limited to that. Unfortunately, the tsunami wave arrived, and all hell broke loose.
When the power plant was designed in the 1970s, close to the ocean, it was designed to be protected from possible future tsunamis. As for all things, it all boils down to what is considered “possible.” At the time, a ten-foot tsunami had been recorded from the 1960 Chile earthquake, so that was taken as the design basis in planning for future tsunamis.15 Therefore, the power plant was built thirty feet above sea level and the seawater pumps that circulate the water necessary to cool down the reactors—which remain hot for quite a while even after a shutdown—were located twelve feet above sea level.
Then—oops—the Tohoku earthquake generated a massive tsunami wave that was forty-five feet tall when it reached Fukushima. Flooded by fifteen feet of water (and more for the pumps), the plant lost its emergency diesel generators. Electrical switchgear and batteries located in the inundated basements also became useless. This delayed cooling and was the beginning of a long process in the attempt to control the reactor. It did not help that many access roads had also been taken out by the tsunami.
Without going into the mechanics of how a nuclear reactor works, suffice to say that out of this mess, as dedicated personnel worked to prevent a bigger catastrophe,16 three nuclear reactors suffered meltdowns, three hydrogen explosions occurred, and radioactive contamination was continuously released over a three-day period.17 Post-disaster investigations revealed that studies conducted in the years prior to the Tohoku earthquake had brought to the attention of the government and the power plant owners that tsunamis of up to fifty feet were possible and pointed to the need to implement protective measures in anticipation of such events,18 but those warnings were not acted upon, awaiting further expert review.
19 Some postearthquake critics denounced the collusion that existed between regulators and industry in Japan, stating that many of the regulators were former employees of the power companies.20 Given the consequences of a nuclear reactor failure, one would indeed expect due diligence when a red flag is raised. As a result, beyond all the damage to the power plant itself, there was also damage to the people’s trust in their decision makers. The faith that public safety would be upheld by all stakeholders as the upmost priority was shattered. The Big Bad Wolf had blown off the house that was supposed to be safe; in the aftermath, baffled, many wondered how they had ended up with a straw hut.
In some countries, Fukushima eroded whatever trust remained in nuclear energy. For example, following Fukushima, Germany shut down eight of its seventeen reactors and promised to close the rest by 2022.21 Siemens, the German multinational giant, terminated its nuclear engineering industry, announcing it would no longer build nuclear plants across the world. Similarly, South Korea, Switzerland, Taiwan, and others enacted laws to phase down nuclear power in their respective countries.22 Arguably, support for nuclear energy in some of these countries was already shaky, and Fukushima only gave the needed incentive to trigger or accelerate plans to move away from nuclear energy. Likewise, a few countries that intended to open the door and adopt nuclear energy decided to keep that door closed for a while longer.
23 Some countries—like India—have elected to still move ahead with the building of many nuclear power plants but have slowed their construction schedule to ensure greater safety (or for political reasons).24 Not surprisingly—it takes all sorts of people to make a world—in some countries, Fukushima did not make a difference, and plans for new nuclear plants are moving ahead undeterred, or faster than before, out of need or for opportunistic reasons.
Part of the shockwave of the Fukushima failure on the global nuclear industry is attributable to the fact that, since the 1980s at least, Japan had been recognized to be a producer of top-quality, engineered products, with a track record of reliability, performance, and world-class quality controls. Japan is not one of those emerging or corrupt economies where crappy products and lack of safety are the norm—and where, incidentally, nuclear power plants are also being constructed.
In Japan itself, some fancy footwork was in order. On one hand, large companies like Toshiba, Hitachi, and Mitsubishi, were in the profitable business of selling nuclear power plants worldwide; on the other hand, public trust had collapsed. Mega-conglomerates cozy with politicians may have more money, but the public has more votes. As such, following the Fukushima incident, the Japanese government ordered all of the country’s thirty-eight nuclear reactors to shut down within a year and required compliance with new safety standards so stringent that only nine reactors had resumed service by 2019.25 Elected officials also insisted that there would be no new construction of nuclear power plants in the country—fourteen new ones had been planned to open by 2030, but Fukushima put a halt to those plans.
Paradoxically, Japan’s hope to continue building reactors abroad was undermined by the construction freeze at home.
Who would buy nuclear power plants from a country that prevents them from being built at home—from the very country where the largest nuclear disaster of the millennium happened? This would be like buying steroids from suppliers in a country that has stopped winning Olympic medals. In 2017, when Westinghouse filed for Chapter 11, facing billion-dollar losses on power plant projects, its parent company—Toshiba—could not keep it afloat, as it was itself sinking.26 The downfall of German and Japanese nuclear plant builders following Fukushima opened up the world market to reactors manufactured by China and Russia. In 2018, twothirds of the nuclear reactor projects worldwide were being constructed by Rosatom, a Russian state-owned energy company27—presumably on the expectation that the new Russia is not as corrupt and contemptuous of safety as the USSR of the bygone Chernobyl era. As for China, on the basis that some Chinese decision makers aim to build five hundred nuclear power plants across the country by 2050 (which is possible if extrapolating the country’s $57 billion 2010–2020 investment with twenty nuclear plant projects under construction, together with the fact that it is planning to spend an additional $850 billion doing so until 2050),28 some scientists are already calculating the probabilities of a major Chinese nuclear disaster.
29 One of the lessons of the Fukushima disaster is that designing for probabilistically based scenarios is—well—just that, namely setting up an acceptable risk and living with the consequences if an extremely rare event exceeds the design basis. Regardless how rare the event—be it a one in a thousand, one in ten thousand, or one in a million years event—eventually technology will fail. All technologies will.
Instead of relying on probabilistic calculations to bury a risk in mathematics, a better solution is to plan fail-safe approaches—that is, defense mechanisms, embedded in the infrastructure itself, that assume things will fail but that will be able to control and minimize the consequences. As a case in point, various types of elevators have existed for over a millennium—chairs and platforms pulled up by rope for the most part.30 Yet rare were those who wanted to buy elevators until Mr. Otis pulled a stunt at the 1854 New York Exposition, cutting the rope of his elevator, sending it into freefall (gasps!) until it was stopped by an automatically released mechanism that gripped the elevator guide rails and brought it safely to a stop (applause).31 Fail-safe. Sales boomed.
EVEN THE SKY IS FALLING Nothing ever fails because of an error in the laws of science.
Bridges do not collapse because of a sudden glitch in the law of gravity, airplanes do not crash because molecules in the atmosphere misinterpret the rules of fluid dynamics, and nuclear power plants do not melt down because of an unexpected change in the universal laws of thermodynamics and nuclear fission.32 The laws of the universe rule everything, from subatomic particles to tenbillion-light-years-wide superclusters of galaxies. They have created the world as it is, and as chaotic as this world might seem at times, the physics of it all is constant. Plainly said, the universe does not make “mistakes.” Therefore, when technology does not perform as intended by humans, it can only be because of human factors.
One of the primary laws of physics is the principle of minimum energy, which states that everything in the universe has a natural tendency toward the state where the least amount of energy is needed to be in equilibrium. Any parent with a teenager slouching on a couch can immediately relate to the purest expression of this law of physics.
Therefore, constructing a road hundreds of feet up in the air to connect the two sides of a canyon requires carrying material up there and taking appropriate measures to keep it up there forever. The point of least energy for all that material is at the bottom of the canyon. Whether fighting gravity that wishes to bring it down in big chucks, or corrosion that wishes to bring it down one atom at the time, it is a constant battle—and the same goes for any created technology. A number of human factors conspire to make the battle arduous—factors that could lead to failure.
Pressure to reduce costs and deliver on time may lead to “cutting corners” in design, fabrication/construction, quality control, and safety checks. Incompetence, negligence, or outright corruption (or even sabotage) can lead to a product that will eventually show its deficiencies either in the long or short term. Incomplete or erroneous data, ambiguous specifications, inaccurate models, and inability to account properly for the complex nonlinear interaction of interdependent systems, combined with lack of experience, can lead to improper design decisions. Lack of sleep, emotional or health problems, internal politics, personal biases, and other similar factors that affect performance can also allow errors to sneak through the process. And then there are errors that happen simply because people sometimes make mistakes without any reason.
In short, to err is human. The only way to achieve a successful outcome is to put in place enough mechanisms and processes to counteract all of the human factors that might lead to failure. This does not mean getting rid of every engineer who has spousal problems, but rather enforcing the application of accepted practices; using conservative and fail-safe designs; having realistic budgets, resources, and timetables; questioning existing practices and perceptions; continuously updating the state of practice to account for new knowledge; and developing effective and practical plans for continued monitoring during and after construction. That is a lot of work—just saying it, is exhausting. Yet this is most important for infrastructure projects because, contrary to other industries where manufacturers can stress-test a product for hours—or even crash-test it on a wall—before releasing it to the market, it is quite difficult to “crash-test” a nuclear power plant.
On the positive side, infrastructure projects are engineered today with more care and due diligence than most manufactured products. This is most reassuring given how manufacturing has evolved in recent decades, particularly in the high-tech sector where quality control has sometimes been thrown out the window under pressure to beat the competition and/or make a quick billion. Some decision makers have apparently concluded that it is less expensive to replace a small percentage of products shipped defective than to invest in quality control to identify them in the plant and prevent their release. In less politically correct terms, this means that too many companies do not mind selling crap as long as there are suckers buying it. This mentality is a direct extension of the despicable software engineering philosophy that consists of releasing flawed products early to cash in as soon as possible and then fixing bugs on the fly as problems are encountered by users—essentially being inconsiderate to these aggravated users stuck with the problems. In essence, this is shifting the responsibility for a part of product testing from the manufacturers to users.
When a bug in poorly written software prevents a computer-controlled enemy to activate as intended on the thirtieth level of a video game,33 or freezes half the screen at the 256th level of Pacman,34 who cares? The software developer who wrote the “spaghetti code” that led to the error is insignificant in the grand scheme of things—just like the video game in itself. However, when buggy software impacts the lives of real people (who usually have only one life, contrary to pixelated characters in games), that is more consequential. If code writers wish to promote themselves as software engineers, then they must be held to the same standards of responsibility and ethics that are required in other engineering disciplines—something many argue is far from the case.35 Software bugs are widespread, and defective updates are released via the internet for everybody’s pleasure and enjoyment. Some even make the news. For example, many people purchased the Nest smart digital home thermostat that promised to learn from the habits of its users, by identifying their preferred temperature at various times of day and programing itself to save and reduce the energy bill accordingly. Noble goal.
However, as smart as the thermostat might be, in a less smart move, those who maintain its soft-ware released an update in December 2015 that uploaded itself to all these internet-connected devices and caused them to drain their batteries and stop functioning by mid-January 2016.36 With relatively mild temperatures at the Nest headquarters in sunny Palo Alto, California, those venting their frustration on social media were not impressed that Nest’s programmers did not seem to have considered the possible consequences of a thermostat failure in less balmy mid-winter climates when they released their update.37 They were equally unimpressed when Nest stated it was aware of the issue and expected to have a solution ready to “roll out in the coming weeks,” together with instructions on how to manually reboot the thermostat in the meantime as a temporary fix.
This compassionate message must have provided great solace to all, especially those from northern states who were traveling when the “glitch” happened and had nightmares of returning to a water-damaged home buried in ice due to burst pipes.38 At the other end of the fire and ice spectrum, it was even more news-worthy when it was discovered that the battery in the new Samsung Galaxy Note 7 phone/tablet tended to overheat, release smoke, catch fire and sometimes explode.39 Not exactly an insignificant bug. Devices were recalled and reissued with different batteries supplied by a Chinese company instead of Samsung itself, but—adding insult to injury—a second recall had to be issued as some of the replacement units suffered the same problem of failure and combustion.40 Given that Samsung’s mobile business chief, in response to the crisis, stated, “I am working to straighten out our quality control process,”41 one can only wonder what quality-control processes he was referring to that could have allowed a million devices to be produced and shipped without anyone catching this problem.
The scary thought is that this computer science mentality of quick-to-market and imperfect quality control could be spreading. When a retired quality manager who had spent decades working for Boeing filed a whistle-blower complaint with the Federal Aviation Administration alleging that the flagship 787 planes had been shipped with problems that could lead to electrical shorts and cause fires,42 and when all the 737 MAX planes were grounded after two crashes while Boeing was trying to fix—surprise!—a software problem,43 regulators and lawmakers launched investigations, issued subpoenas, held hearings, and so on. Yet, given that not all corporations have the engineering skills and wherewithal of Boeing, or its visibility, that leaves many citizens wondering what might be the norm in various industries nowadays? Are existing practices, regulations, and assorted corporate cultures adequate to ensure that the lure of quick profit does not take precedence over safety?
Note that software errors having tragic consequences is not a new phenomenon; modern history is replete with episodes where fatal bugs were found, often where one would least expect it. For example, in 1985, many patients were killed or seriously injured before somebody realized that this was caused by a flawed software upgrade to the Therac-25 radiation therapy device44 that made the machine bombard people with one hundred times the intended dose of radiation.45 During the investigations launched by the US Food and Drug Administration after a few cases of a lethal dose of radiation had been reported, multiple software design errors were found.46 In particular, it was discovered that the software had not been independently reviewed and that testing of the device had not been redone after a software upgrade.47 Saying, “it’s not a bug, it’s a feature” is not as funny as the marketing department may think. To make it worse, from teenagers of the generation born with smart phones USB-connected to their pacifiers to senior nerds raised on punched cards and stacks of folded printouts, too many keyboard wizards have a condescending view of those they regard as being “technologically challenged.” Such egotism, lack of care, and, in some cases, lack of life experience, sadly can make the computer geek “tech-not-logical,” or “tech-illogical,” resulting in products whose newest version is lacking some of the best features of its previous one, and device operations that defy logic. It takes time to do a good job. It takes time and dedication to make sure everything will work as intended— safely and reliably. That slows down the release to market and does not generate more profits. For example, how many engineers get big bonuses or promotions for doing all the due diligence to prevent disasters? Probably not that many —if any—because nobody wakes up in the morning grateful for all the disasters that did not happen in the previous twenty-four hours but that would have happened if not for the duty of care upheld by those who keep things safe: the true pillars of society. Instead, it is those who spring to action after a disaster to pick up the pieces and set things on some sort of a course to recovery that are called heroes and get all the medals and accolades—sometimes because they are genuinely good at fixing up big messes, and sometimes only because they are blabber mouths who are quick to jump in front of cameras to claim credit—even when those disasters might have been prevented in the first place if these same heroes had displayed foresight, adopted different priorities, and taken different actions prior to the events.
Fortunately, on the positive side, as stated earlier, comparing the state of the built infrastructure with that of manufacturing—or, God forbid, that of the high-tech industry—should be avoided, because these operate in different worlds when it comes to accountability.
Yet, on the negative side, things are still far from rosy when it comes to infrastructure. Technological disasters constitute a subset of all possible disasters that can be caused by human action or inaction (a category known as anthropogenic disasters). The truth of the matter is that there are millions of buildings, bridges, and other infrastructures worldwide that have been designed according to obsolete requirements, waiting for hazards to bring them down. MASTER BUILDERS AND ENGINEERS Theodore Cooper had been hired as a consultant because he was one of the foremost bridge engineers at the time, recognized for his pioneering work on the analysis and construction of steel structures, and particularly long-span bridges.48 Cooper had even twice won the prestigious Norman Medal of the American Society of Civil Engineers, awarded in recognition of the impact of a specific technical contribution on engineering practice—a reward received by only a select few, with repeat award recipients being even more rare.49 The bridge was to become the world’s record holder for longest span, but as Cooper was in his sixties, in poor health, and unable to travel away from his New York City office, he hired Norman McLure—a young Princeton graduate—to be his eyes on the construction site.
At one point, mid-construction, McLure traveled to meet Cooper in his office to review in person, and in more details, worrisome measurements that had been taken on some of the bridge members over the previous weeks (a member being a component or element of a structure, such as a beam, column, or truss diagonal). Problems had first been reported on June 15, 1907, when some members of the bridge could not be connected because the pre-drilled holes on adjacent members did not line up. Errors of up to a quarter inch were noted. At the time, these were assumed to be fabrication errors, presumably in the pre-cambering of the member that had been performed prior to shipping— yet, this was odd, considering that pre-cambering is something that was calculated and done on purpose to make members line up perfectly when subjected to loading.
However, subsequent inspections in early August revealed out-of-straightness of as much as three-quarters of an inch in some members, which caused the field engineers to wire a message to Cooper. This triggered much discussions as to whether these members had been banged up and bent prior to shipping or not. In spite of McLure’s conviction that the bend in the members occurred after they were installed in the bridge, Cooper argued that the members had to have been hit by other members before being installed. Over a two-week period, the lateral deformation increased from .75 inches to 2.25 inches. That is enormous for a member that is supposed to remain straight, so on August 27, the contractor stopped all work asking for this issue to be reviewed. McLure traveled to New York to meet with Cooper on the 29th.
While McLure and Cooper met, unknown to them, work had resumed on-site because the local chief engineer employed by the entity that had issued the constructions bonds—essentially, a bunch of businessmen and politicians incorporated for the sole purpose of this project—declared that the stoppage was bad for the morale of the construction workers. This decision was backed up by the top brass at the Phoenixville (Pennsylvania) headquarters of the construction company, who called the site to assure them that it was safe to proceed. Meanwhile in New York, McLure and Cooper had concluded that no more load should be added to the bridge until they had thoroughly reviewed all the facts. Cooper wired this information to the Phoenixville construction company and traveled there at once for a 5 p.m. meeting the same day. Unfortunately, in his haste, McLure forgot to wire the same information to the construction site.50 The meeting at the Phoenixville headquarters was brief and concluded with a decision to reflect on this overnight and resume discussions the next morning.
At 5:30 p.m., on August 29, 1907, the steel members having the large out-of-straightness suddenly buckled. The Québec Bridge collapsed, killing seventy-five of the eightysix workers on the structure when the nineteen thousand tons of steel plunged into the Saint Lawrence River. Why did it happen? Could it have been prevented? How to prevent it from ever happening again? These are some of the questions asked after any failure with fatalities, be it a bridge collapse or a plane crash. The truth of the matter is that pushing the boundaries of technology means entering uncharted territories. As stated multiple times throughout history—by French Revolution decrees, Churchill, Roosevelt and many more,51 and popularized as the Peter Parker principle by Spider-Man fans—with great power comes great responsibility, which means here that extensive testing on components and systems must be the norm when developing new technology. Yet testing implies that one knows under what conditions something is to be tested— that is, what phenomena are to be replicated. Knowledge might simply be lacking, or the scale of the system might be so large that only small components can be tested, from which behavior of the entire system must be inferred.
For example, transmission lines have been weaved across continents for decades, but in building those, nobody anticipated that solar flares could be a problem. It turns out that the wild burning gas ball that is the sun, every now and then, has sudden outbursts of energy equivalent to thousands of nuclear bombs exploding simultaneously. Each of these powerful explosions, called solar flares, happens to shoot a magnetic wave straight toward the Earth at a million miles an hour. An unusually intense solar flare occurred on March 13, 1989.52 When it hit home, it disabled the entire Hydro Québec power grid, putting six million people in the dark for nine hours. The entire northeastern region of the United States barely escaped its own blackout, thanks to the fact that the loss of power from up north (which feeds a lot of the US grid) happened at 2:44 a.m. when demand for electricity was low. When stringing electrical wires over thousands of miles, who would have thought of the sun as a hazard—other than construction workers getting sun burns.
Learning from experience, the power grid has been upgraded to prevent future recurrences of this problem,53 but this illustrates how learning by failure is an important aspect of technology, and how foreseeing every possible problem and testing every system ahead of time for every possible technological failure can be difficult—in this case, it would have “simply” required blowing up, somewhere in space, thousands of nuclear bombs at the same time.
Coming back to the Québec Bridge, the Royal Commission that investigated the collapse revealed a combination of shortcomings in knowledge, engineering judgment errors, and organizational failures. In the 1900s, steel bridges had been built for over thirty years, but it was still a relatively new technology—somewhat like rocket science in the 1960s. Problems were many. First, the existing knowledge on the buckling strength of members in compression was far from perfect, and the entire Québec Bridge strength relied on the strength of large compression members. Then, in May 1900, the bridge’s clear span length was increased from 1,600 feet to 1,800 feet. This relocated the piers outside of deep water and thus eliminated the need to design them to resist the strong current at that site and large ice forces on the piers what would have come with it—and at the same time, it made Theodore Cooper the chief engineer of what was to become the longest cantilever span bridge in the world. In principle, this should not have been a problem in itself, but relying on his engineering judgment and because of budget constraints, Cooper did not deem it necessary to revise the design to account for the extra weight added to the bridge due to its lengthening.
Instead, he increased the value of the maximum stress allowed on members, essentially shaving away the margin of safety. Finally, Cooper was a prominent consulting engineer, apparently “self-confident to the point of arrogance,”54 without a counterpart on the construction site who would have had the confidence to challenge and contradict him. Therefore, as many failures often do, this collapse happened due to a combination of unfortunate circumstances. It ended Cooper’s career, who retired from public life.55 Defeated politicians may sign multimillion-dollar book deals to brag about their pretended accomplishments and greatness, but defeated engineers simply vanish, quietly.
Cooper was not the first nor the last bridge builder to see his work destroyed by forces of nature, be it gravity, wind, ice jams, or other hazards. Most of the time, phenomena unknown to exist—or unknown to have an impact on bridge performance—were discovered or rediscovered because of failures.
Collapses of large bridges due to lack of engineering knowledge have been well documented throughout the nineteenth and early twentieth centuries. For instance:56
79 people died in 1845 because the chains of the Yarmouth suspension bridge in England fractured when too many people assembled on the bridge to view a clown floating down the river in a barrel—hard to think of a more ridiculous reason to die. Oddly enough, the bridge had been designed by an architect with little relevant prior experience, who did not bother to perform the tests necessary to ensure adequate strength of the construction materials, and who showed up on the construction site for the first time on the bridge’s opening day.
57 It can be speculated that this architect’s specialization in designing churches throughout his career58 possibly provided him with the kind of holy contacts needed to avoid jail after the bridge collapsed, in spite of harsh criticisms by the British Institution for Civil Engineers.
226 soldiers died in 1850 when troops marched in step across the Angers suspension bridge in France and created resonance—a dynamic amplification of vibrations due to repeated impulses at the same fundamental vibration frequency as the bridge. The Broughton Suspension Bridge in England had similarly failed in 1831,59 so soldiers on the Angers Bridge knew of the dangers of walking in lockstep across a bridge— something armies going as far back as the Romans also knew. But the Angers Bridge was crossed during a wind storm that made it oscillate, so even though the soldiers had been told to space out and break step, the severe swaying of the bridge made them involuntarily take steps in sync with the bridge motions to be able to keep their balance. In other words, because the soldiers merely attempted not to fall down, the bridge did.
Nowadays, soldiers are still often cautioned to break stride when crossing bridges; for many large modern bridges, this might not be necessary anymore, but old habits die hard.
Many died when a number of bridge spans in the United States collapsed due to train overloads in the 1850s, and in many other countries in the subsequent decades, including the Gasconade Bridge in Missouri in 1855,60 the Desjardins Canal Bridge in Ontario in 1857,61 and the Sauquoit Creek Bridge in New York State in 1858,62 to name a few. What makes the Gasconade Bridge uniquely interesting is that it occurred during its inaugural train run—what better way to impress six hundred dignitaries and guests on board than to have the train plunge into a valley along with bridge debris. In the purest tradition of US trains arriving late, this one simply never arrived to destination. Around forty passengers died—or, from the point of view of a diehard optimist, more than 90 percent of those on board survived. Who would have thought that train travel could be so hazardous?
Many more died due to lack of knowledge on the various conditions that could lead to fractures in metals, a shortcoming that led to collapses of the Ashtabula River Bridge in Ohio in 1876,63 the King Street Bridge in Melbourne, Australia, in 1962,64 the Mianus River Bridge in Connecticut in 2003,65 and many more—although some more recent failures can be attributed not to ignorance but rather to deferred maintenance practices.
Many bridges have been washed away by floods, or by flash floods such as the Dry Creek Bridge, in Colorado in 1907, for which the sudden surge in water level unfortunately happened at the same time as when a train was crossing, resulting in a hundred deaths.66 Ice jams also did the trick every now and then, such as for the Upper Niagara Falls Honeymoon arch bridge which failed in 1938 when the ice in the gorge pushed the arch off its abutments.67 The world-famous Tacoma Narrows bridge failure, in 1940, whose deck vibrated in a twisting motion when subjected to a steady 40 mph wind before breaking apart,68 brought to the attention of bridge engineers the critical need to consider aerodynamic forces in the design of slender suspended spans. It also brought to the attention of high school students the fact that science teachers love to show videos of the Tacoma Narrows bridge collapse to explain the importance of resonance in physics—although, strangely enough, it is not resonance that killed the bridge but aerodynamic fluttering, a topic way beyond the high school physics curriculum.69 The Duplessis Bridge in Québec collapsed in January 1951, not because of sabotage by communist terrorists (as alleged by the prime minister at the time), but because the specific kind of steel used in those spans had brittle properties at cold temperatures and it was minus 34 degrees Celsius when it collapsed at 3 a.m. in the middle of a harsh winter night—definitely crisp weather, even by Canadian standards.70 And so on—the list is long.71 And that is not even considering bridges that are blown to pieces in war zones, or those doomed from the start because they are built in corrupt countries where infrastructure projects are, in large part, a way to divert public funds to line the pockets of politicians and other crooks.
Nowadays, bridge failures still occur for a wide number of reasons (some crazier than one can imagine), but mostly due to boat collisions, construction errors, or extreme events. For example, in 1980, a freighter boat suddenly caught in a wind burst and torrential rain that reduced visibility to zero did not have time to drop anchor before hitting a pier supporting the main span of the Florida skyway, causing 1,200 feet of the bridge and thirty-five people to plunge 150 feet to their death.72 Failures due to extreme events more commonly occur to bridges designed at a time when the magnitude of the possible extreme demands was not well known, or when knowledge on how to design these bridges to enable them to survive such events did not exist. For example, a milelong segment73 of the double-decked elevated highway in Oakland known as the Cypress Freeway collapsed during the Loma Prieta earthquake—killing forty-two people in the process—because it was built in the 1950s, at a time when engineers did not know how to detail reinforced concrete structures to prevent their collapse during earthquakes.74 After the earthquake, the road was rebuilt at ground level rather than elevated, at the mere cost of $1.2 billion dollars and nine years of construction.75 Although the vulnerability of the viaduct had been brought to attention and measures could have been taken to strengthen the structure prior to the earthquake and thus avoid its collapse,76 waiting for the earthquake to wipe it out and reconstructing it anew afterward had the distinct advantage that the federal government footed 90 percent of the bill.77 Maybe money flows more easily after a disaster for the same reason that nobody is rewarded for preventing a disaster. The more visible the pain, the more opportunities for well-intentioned politicians to shine (apologies for the oxymoron), and the greater the largesse.
Going forward, with millions of smart phones and surveillance cameras on the ready, bridge failures have become YouTube events. Some relatively recent collapses caught on video include an overloaded new pedestrian bridge in China in 2013,78 a century-old stone bridge in Greece in 2018,79 and a modern 450-foot-long arch bridge in Taiwan in 2019.80 “Wait a minute!” would say the astute reader at this point. “Are there not supposed to be design codes that engineers are supposed to follow and that exist for the sole purpose of preventing problems—and most importantly, failures and collapses? After all, there are codes for plumbing, electricity, roofing, and just about anything that can be constructed.” The answer to that excellent observation is a categorical “Yes, . . . but.” This requires a few explanations that will come much later in this book.