And that’s true not just in the U.S. but in other countries, too. The kind of inner city flooding we saw in Houston is very common in China. In June I was in Nanjing and there was a huge rainstorm; about one-third the annual precipitation fell in three days. They typically have a rainy season in early June to late June, spread out over three weeks. But in this case they got all that rain in three days. So if you look at the mean annual rainfall it doesn’t look that unusual, but the intensity of it was extreme.
Half of the city of Nanjing flooded, and it wasn’t because of water being introduced into the city via the river. It’s just the fact that you had so much water coming into the city, and the city generally isn’t prepared to cope with such a large accumulation of water in such a short time… The water doesn’t stay long — maybe a couple of hours, or as in Houston, maybe a few hours. But that’s enough to cause major damage.
In the case of Houston, some blamed the city’s lax zoning, particularly in parts of the city that were already flood-prone.
: When we do some zoning classification, even for cropland, we create the zoning based on the mean state of the climate. But, again, we don’t take into account the extremes. We can adapt to mean change reasonably well, but it’s the extreme that is problematic. One example I often cite is Berlin. On most days the climate is pretty cool, so very few people have air conditioning. But when you have a heat wave that exceeds 100 degrees and persists for a week it can be hugely stressful. But that’s not the case in, say, Tucson, Arizona because that’s the mean state and everyone is ready for it. But when you have such a deviation from the mean state then society gets stressed out.
That’s the danger of extremes. A lot of city planners don’t have the mindset of preparing for extremes. I’m not a city planner, but when you think about how plans are made a lot of times they are done with a relatively short time horizon: 5 or 10 years down the line. When Brad [Gentry] and I did the class, “Cities in Hot Water
,” we talked about how cities create hazard mitigation programs built around pretty short time horizons — like five or 10 years. But when a 100-year flood comes, it’s far beyond that horizon. And when it comes you’re not ready for it.
That class gave students a real opportunity to apply these ideas and methods in a really tangible way. Were there any outcomes from that course that would be wise for other cities — and not just New Haven — to heed?
: One outcome that jumps to mind is that we just don’t have good enough data to support concrete recommendations in so many cases. There are some areas of the city, for instance, that are considered flood hot spots because of their proximity to, say, the Quinnipiac River or because and they’re close to the coastline. When you have a combination of high tide and storm events in those areas you have flooding. But there are many other locations that are also prone to flooding, but we don’t have the detailed spatial data to help us pinpoint exactly where they are. That’s something we think is going to be very helpful; neighborhood by neighborhood planning for flood prevention. We need that very detailed work.
Sometimes, for instance, you have a street going down below an overpass, and you need to understand where that is in relation the water outlet. So you can say, ‘This is going to be a chronic flooding issue when precipitation goes beyond a certain limit.’ The problem is that it’s very difficult to plot… We have elevation models for the natural environment. But when you fly some instrument to measure elevation in the urban environment it doesn’t always know if it’s measuring a street or a bridge or the top of a roof. So if you mistake a roof for the ground surface obviously your prediction can be completely wrong.
What are some of the smaller-scale strategies cities can take to make them less vulnerable?
: We tend to model places where ‘good practices’ are used and we try to learn from them. The Netherlands, for instance. But the lessons there might not be applicable because the overwhelming issue there is sea level. So they build sea walls and cities in such a way that identifies vulnerable spots and then try to reinforce those spots. In places like New Haven we have a seawater problem, but we always have inland flooding. In places like that, having public awareness of flood prone areas would be helpful.
People have a hard
time relating to extremes — I know I do. So when there is an extreme event you have to say ‘OK, let’s learn from this.’ You cannot envision what the future climate will look like, but we certainly have vivid memories of our last hurricane. In our case, Sandy… that’s the kind of future flooding we’ll be seeing on the coastal areas. Using those extremes to help people increase their awareness of what future climate will look like is very helpful.
In the U.S. we’ve had so many such events, such as Sandy or Katrina [which devastated New Orleans], that do capture the public’s attention if only for a short while. Are there examples where extreme weather events have influenced government to explore vulnerabilities and engage the public?
: I have experience in China, and I know they tend to use those experiences to learn how to prevent future flooding. And it’s not political over there. It’s more of a practical solution. They learn from those extremes, and people use the extremes to help prepare for the future. In 1998 there was extreme flooding on the Yangtze River that caused devastating damage along the lower reaches of the river. A lot of cities were flooded which resulted in high mortality rates. So the government decided to introduce an ambitious plan to restore forests in the upper reaches of the river basin and in other degraded lands, and a reduction in the amount of timber harvesting. They also built a series of dams on the river, which of course is controversial. This past summer in terms of rain was quite strong, almost as strong as that flooding year, but there was not major flooding. So it’s quite useful to learn from these major events.
It would be nice if here in the U.S. we could respond this way to extreme weathers, and not to simply politicize the extremes. Too many politicize the extremes. People continue to say, ‘this isn’t climate change,’ so they don’t take action. But whether it’s climate change or it isn’t we can still learn from extremes.
Of course a larger solution would be to cut down on carbon dioxide emissions but that becomes an international effort and brings up questions about who has the right to pollute, who has the responsibility to curb those emissions. But again if you motivate local actions important solutions can happen at that level. We’re good at creating global scenarios. We know, for instance, that if we do business as usual the temperature will likely be 8 degrees warmer 100 years from now, but if we curb it to 2 degrees then we’ll have another scenario… But most citizens have a hard time relating to those scenarios. So we need break down those scenarios into something that is relatable to them.
You said it’s difficult to conclude that climate change causes more hurricanes. But as someone who studies the connection between terrestrial landscapes, human activities and the atmosphere, how confident are you that what we’re doing is contributing to the extremes that we’re seeing?
: I think that there’s a general consensus on that. You get more extremes when you put more greenhouse gases into the atmosphere. The linkage to the mean state is relatively easy to understand: If you put more CO2 into the atmosphere it traps energy, and therefore the climate becomes warmer. And when the temperature goes up the atmosphere can hold more moisture, and therefore you’d expect to get more precipitation. So in terms of the mean state, it’s easy to explain.
So it’s really an issue of perception. When it comes to extremes you need statistical models and you need some level of training and expertise to understand the arguments. But in terms of trends, it’s in the data. If you look at distribution, the mean’s shifting. But we’re also seeing that the tail ends are spreading out from the means, or more of a
larger standard deviation from the mean. What does that mean? We’re seeing an increase in extreme events.