In June and July, parts of western Japan were hit by torrential rains, with some regions experiencing record-breaking totals of more than 1,000 millimeters (39 inches) of precipitation in only a few days. The heavy rains led to widespread floods and mudslides throughout the country as rivers broke through embankments and floodwaters reached more than 16 feet above normal levels in some areas, forcing millions to evacuate. At least 225 people were killed in the disaster, making it Japan’s deadliest flood in 36 years.
Catastrophe modeler AIR Worldwide estimated insured losses from the disaster to be between 284 billion Japanese yen ($2.6 billion) and 423 billion Japanese yen ($4 billion). These numbers only account for residential, commercial, industrial and automobile property losses, however, and do not factor in business interruption, losses to land and infrastructure, construction and erection all-risk, marine hull and marine cargo insurance losses, and demand surge costs—the true toll is likely much higher.
In the wake of the catastrophe, experts have sought to gain a better understanding of what went wrong and determine how such widespread damage can be prevented in the future.
The Need for Greater Awareness
Takashi Ohkuma, head of Niigata City Lagoon Environment Research Institute, has been critical about the lack of preventative measures taken by government authorities to prevent flood damage. Many overlook the fact that Japan is built in a disaster-prone area, he noted. “The residential areas around Hiroshima were created by floods,” he said. “Fifty years ago, this area was farmland, and through time it was developed as a residential area. Yet, the people who developed the areas hit by the disaster lack such awareness—both decision-makers as well as ordinary citizens. And so they approved the area for construction while forgetting the reality that it’s a risky, disaster-prone area.”
He believes it is crucial to increase awareness of this inherent level of risk. “I think it is important for people who live there to understand what caused the disaster in the first place,” Ohkuma explained. “First of all, if there is no major levee breach on the river, this big disaster wouldn’t have happened.” In the past 30 or so years, his institute has warned regional authorities responsible for the rivers to reinforce the dikes. “At that time, the methods and technology to do this were lacking,” he said. “But now we have a lot of reinforcement methods and technology. Despite the availability of such methods, the government didn’t implement the reinforcement for the dikes.”
Yukiko Kada, former president of the Japanese Association of Environmental Sociology and former governor of Shiga prefecture, agreed that more preventative measures should have been taken. “As governor of Shiga prefecture, we made a flood prevention policy,” she said. “From my own experience, I believe that, with proper watershed flood protection policies, areas could have been saved.” She cited the example of the city of Kurashiki, one of the cities hardest hit by the rainfall and flooding. “If you look at the Kurashiki hazard map, which shows the dangerous zones with higher risks for floods, it is remarkable that the areas covered by the map are almost identical to the actual floods,” she said. “The two completely overlap.”
Despite the existence of an accurate hazard map, lack of communication proved to be a major problem. According to Kada, residents were not even aware of the map’s existence. In addition, river maintenance and management were not done properly. “Management inside the river must be done by the national authorities and the prefectural government, and the areas where people were living were not managed by anyone when it came to controlling the water level,” she said. “Lack of communication made it difficult for people to know what was going on.”Japan used to have what Kada called a “disaster” or “resilience” culture—there were strict criteria set by governments that had to be considered before constructing houses and villages in certain areas. “For example, if you wanted to build a house with a high flood risk, you needed to elevate it,” she said.
But ever since areas like Kurashiki that were once primarily used for farmland began to see more industrial development in the 1970s, these strategies seemed to be forgotten as more construction occurred in flood-prone areas. “Because of the industrialization, this wisdom was lost,” she said. “Ideas, concepts and technologies changed.”
One of the new strategies for water management was the building of dams, but ultimately Kada believes this may have given people a false sense of security. “People thought, as long as we have a dam, we should be all right,” she said.
Despite assurances from the government to the contrary, experts warn that dams do not make communities completely safe from flooding. “Some dams are effective, but in some instances, dams were built 50 years ago,” Kada said. “Dams fill with sediment over the decades. It will be a negative asset for our future generations.” Additionally, even if you construct a dam, once rainfall exceeds the dam’s capacity, the dam will not be effective. The people living downstream feel safer when the dam is built, but this feeling is false, she said.
According to Kada, the most important thing is to raise awareness among the people living in disaster-prone areas that they are at risk. But she believes such risks are often downplayed for political reasons. “Politicians are the ones selling the land, most notably members of the powerful Liberal Democratic Party,” she said. “They want to make money by selling land without letting people know the risk of building on that land. It’s quite sad though, as people died.” In the Mabi district in Kurashiki, for example, 51 people were killed in the flood. Of those deaths, 42 drowned on the first floor, 36 of whom were over 65 years old.
The sobering fact is that there are many more villages in Japan that are built in disaster-prone areas, but Kada believes steps can be taken to help minimize the scale of such tragedies. “Before, we made an ordinance, announced information, educated people, and tried town planning—for example, let’s not build hospitals where the flood risk is high. If you want to build the hospital, you must elevate it,” she said. “To protect people’s lives, you must prepare for that.”
Evaluating Prevention Measures
Looking back at the flooding, what could have been done to prevent so much damage and loss of life? Andrew DeWit, professor at the School of Economic Policy Studies at Rikkyo University in Tokyo, said weather detection radar was able to provide some advanced warning of the rain, but technology can only do so much. “The radar didn’t fail,” he said. “Even if you have very highly advanced radar that can tell you 30 minutes in advance how much rain is going to fall, you can’t get the people out of the way. They must listen to you.”
And that is a key problem: So many died in western Japan because so few people listened to evacuation orders. In fact, DeWit found out that only 0.3% of residents evacuated. In some cases, local governments can be blamed, as certain towns were late in warning residents to evacuate. In Mabi, the local government had to delay a project to install advanced rainfall measuring radar due to local opposition and lack of funding. Perhaps such a system would have provided the necessary warning to convince more people to evacuate.
DeWit believes additional funding is necessary to adequately protect against the risk. “You can see that in Hiroshima prefecture, where the authorities put through a new supplementary budget with an increase of 11%, so they’re spending a lot of money on risk management,” he said. “The question is what they will do with that money. How effectively will they use it, are they going to use a lot in local training programs, organize people to practice disaster evacuation, or encourage people to move away from disaster-prone areas?”
One issue is that some parts of Japan are simply too dangerous to live in, yet people still do. Yet authorities cannot force people to move to safer areas because, as DeWit pointed out, there is a strong negative sentiment about the central government in much of Japan. “If they come in and say you have to move, that becomes a political issue,” he said. “Most people really don’t get how rapidly climate change is accelerating, how dangerous current circumstances are. They rapidly forget about that. You need to convince them about this first, before more drastic measures such as forcing people to move.”
He believes authorities need to make a strong case for moving to urban areas. “Japanese planners link urban planning with transport planning and other policy topics,” he said. “They prefer to move to a compact city arrangement, where you shift public assets and residences to concentrated areas to reduce transport costs and make areas more walkable. But that takes time. You must find ways to incentivize people to move downtown. You need to find a way to incentivize those elderly people, who are dependent on cars and live in suburbs and convince them to move away from vulnerable areas.”
One way to do this could be through a carrot-and-stick approach. The carrot would be to make social services such as libraries and hospitals readily available in safer areas. “The stick is that you change the insurance regime,” DeWit said. “If your house gets destroyed, you don’t get a payment any longer in certain areas, or payment will be reduced, or you could increase insurance payments for people moving away.”
DeWit agreed with Kada that local authorities have a responsibility to create a better system to inform residents about threats before, during and after a disaster. “In Mabi town, where a levee failed, the flood hazard map and the actual flood were pretty much the same,” he said. “You gave information to people saying, ‘here’s what the flood hazard is.’ They should have known this in advance.” But many people do not bother to read flood hazard information when it comes in the mail, so they do not know what to do when a disaster strikes.
Improving Disaster Resilience
A number of other natural disasters have struck Japan in recent months, including a summer heatwave in Tokyo, the strongest typhoon to hit the country in a quarter of a century in late August, and a magnitude 6.6 earthquake in Hokkaido in September. In the wake of these disasters, increased focus has been placed on improving Japan’s national resilience strategy.
Government authorities have said that they want to expand the program to include greater emphasis on both soft resilience measures like hazard mapping and hard resilience actions such as raising levees on river banks and installing advanced radar systems with more rapid computing power. “This will bring in a whole lot of information and crunch the numbers even faster, so you can adjust your sewerage, dams and infrastructure networks to reduce flooding risks as much as possible,” DeWit said. In addition, green resilience measures to help regenerate forests will also help reduce the likelihood of landslides.
Nevertheless, the cost of these risk prevention measures continues to be a significant factor. “It is obvious we have to adapt, but the question is how to adapt most effectively and at the lowest costs,” he said. “Can you build enormous dikes against the North Sea, or is there a limit and you accept some degree of flooding? You can’t do it with the entire coastline so there will be some degree of flooding in a catastrophe. Risk managers will have to decide to write off certain areas to save other areas.” He pointed out that a similar debate about risk mitigation measures happens with respect to wildfires, especially in Australia and the United States, where many experts advocate using controlled burns in an effort to mitigate the possibly greater risk of a catastrophic blaze.
For disaster managers, convincing the public of the need to spend money on resilience is paramount. “You have to spend money,” DeWit said. “The United Nations Office for Disaster Risk Reduction and the World Bank argue that investing one dollar in disaster prevention saves you six dollars. That’s a figure you see routinely quoted. You are also saving people’s lives. It is imperative to invest in adaptation. All kinds of studies point that out. Budgets aren’t limitless, however, so what should you prioritize? How do you get a comprehensive public debate going so that people understand? Public authorities only have a finite amount of money to spend on public goods, so how do you provide those goods most effectively?”
DeWit suggests that one strategy is to bolster critical infrastructure and monitoring systems and, at the same time, move people out of harm’s way, so governments do not spend money and resources on areas that can never truly be made resilient. It is a hard thing to do, but it does not have to be accomplished all at once. “What happens is you ratchet up your ability to do that disaster by disaster,” he said. “You can see that in the evolution of Japanese earthquake measures. After the Hanshin earthquake in Kobe in 1995, for example, they learned about the risk of having heavy tiled roofs and wood-framed housing. But the lesson was expensive—6,000 people had to die before they reached this conclusion.”
Underlying all of these discussions, however, is also the prospect that climate change is increasing the frequency and severity of natural disasters regardless of what measures are taken, and the public is slowly coming to this realization. “There’s a very strong understanding that this is climate change,” DeWit said. “I see that every morning in Japanese news with how frequently they mention climate change. You are starting to see mainstream media linking it to climate change. Attribution studies have doubled.”
As a result, improving disaster preparedness has never been a more urgent matter, both in Japan and around the world. Complacency may no longer be an option. “Climate scientists—who have been arguing for 40 years now about the impact of climate change on weather circumstances—warned us this was going to happen and it is getting worse,” he said. “In fact, their worst-case scenarios were underestimates—a lot of these phenomena are happening several decades faster than most people expected.”