Part of the reason that professor André Sorensen, an urban geographer at the University of Toronto, chose Japanese city planning in the early 1990s as his academic niche is that the topic had barely been explored at the time, at least in English. “Japan was the second largest economy in the world, and there was almost nothing written about it,” says Sorensen, whose Ph.D. focused on Tokyo’s problematic sprawl and whose books have included 2004’s The Making of Urban Japan.

Sorensen’s research and subsequent teaching assignments also had him living in Japan from 1994 to 2002, which means he was in the country to witness the devastation caused by the 7.2-magnitude earthquake that struck the city of Kobe in 1995.

But if his area of study once seemed fairly obscure, it has suddenly taken on new relevance. That’s because of the double blow of a 9.0-magnitude earthquake and 30-foot tsunami that leveled the Pacific coast around the city of Sendai on March 11, killing more than 10,000 people and leaving hundreds of thousands homeless.

During the Industrial Revolution, Japan was slow to build modern infrastructure. Its cities didn’t really use sewers until the 1950s; instead, they carted away human waste to use as fertilizer. But, Sorensen says, construction methods in the years since have been fairly state-of-the-art, so much so that there may not be many new building techniques left to experiment with, to keep the country safe in the future.

Instead, Japan might be better off relocating cities to higher ground. During a recent interview, Sorensen shared his thoughts on that approach and what else the expected five-year, $180 billion rebuilding effort should entail.

CH: What lessons can we take away from how the buildings and infrastructure fared?

AS: Well, you have to divide this into two pieces. First, regarding the quake, modern buildings did pretty well, so I think the earthquake vindicated modern building techniques. This was the biggest earthquake they ever had, so it was a good test of the standards.

The tsunami was a different problem. You can’t protect the coastline from a tsunami with walls that are five meters high [16 feet]. And there will never be the will to build a 15-meter wall across the whole coast.

If you start with that premise, that you can’t protect the whole coastline, then it makes sense to not build settlements in low-lying lands near the coast. But this is really difficult because of property rights.

Many of those settlements will be rebuilt where there they were located. But in the long run, more should be placed on higher ground. By 2050, Japan will see a population loss of 30 million; by the end of century, its population will be half what it is now because of its low birth rate and lack of immigration. So you need a new kind of settlement system.

Is abandoning cities a practical solution?

In every disaster in human history, people pretty much have rebuilt what was there before, with the exception of Tokyo after its 1923 earthquake, when the city came up with a whole new urban design, and a new road system. But then it was bombed during World War Two, and, in rebuilding, they went back to the 1923 model.

Abandoning cities has never been a solution in the past. But it should be part of the conversation. If these settlements were already losing population, there is already a sense of abandonment, especially in small villages. People are already saying, "Forget it, there aren’t enough people here to keep it going." People need to be rational about which settlements you want to maintain in the long run.

Has anything changed since the Kobe quake in terms of approach to temporary housing that’s applicable in Sendai today?

The government was slow and clumsy in the relief effort. It didn’t mobilize the army fast enough. It didn’t realize how extensive the damage was because of bureaucracies. It also was criticized for not taking advantage of volunteer relief efforts. The government wasn’t very good at helping people help themselves or allowing not-for-profit groups to do their work.

After Kobe, the government changed the laws to let nonprofit organizations look for victims. There’s been a rethinking of the role of nonprofits in civil society.

One of the bigger problems with Kobe was that they moved people out of the destroyed area and assigned them kind of randomly to new shelters. Now in Sendai, they are trying to keep neighborhoods together. They realize that stress and loneliness from living in shelters means more people will die from stress and heart attacks than from the initial earthquake. You will cause way more harm by moving millions of people apart rather than keeping them together.

What were some other responses to Kobe that might be applied in general?

There were quite a bit of “land readjustment projects” in Kobe, where they created large public parks in order to have evacuation areas in the event of future earthquakes or fires. But, they will have to get land from somewhere. In Tokyo, roads are narrow in areas with wooden buildings, so they are planning to have wider roads. In the long run, you need to survey land again and buy more land to create larger roads and parks. It’s a long process. But the Japanese are sort of good at constantly improving on original ideas.

What about that third piece in this: radiation leaks from damaged nuclear power plants. How might Japan rebuild to guard against that?

The lack of foresight on behalf of TEPCO [Tokyo Electric Power Company, which runs the plants], as well as the Japanese government was really terrible. All the reactors [about 50 plants nationwide] are on the ocean, yet they didn’t take sufficient measures to protect the reactors against the maximum possible tsunami. It wouldn’t be hard to have a sea wall that would protect them, some kind of big concrete barrier. And why did the generators [for cooling systems] have to be on sea level? They could have been on higher ground.

Are any other parts of Japan at risk from similar disasters?

Tokyo. There are 124 square kilometers [48 square miles]—about 20 percent of the city—below the high-tide level. And five percent is below low-tide level. About 2.5 million people live in this entire area. Those places are very much at risk of high-tide events. While they have dikes and locked gates at ports, and pumps, if electricity for the pumps goes out, you’re in trouble. And 2.5 million people…that’s way more than were affected in the tsunami.