Tornadoes, climate change, and real scientific literacy


The air grows thick. Dark clouds churn like a pot of boiling water overhead. The colors of reality become oversaturated–greens too green, yellow a sickly gold. This is what tornado weather looks like, and the United States has been hit with a lot of it lately.

Make no mistake, the past two months don't just seem to be particularly twister-laden. This isn't one of those situations where an increased awareness of what's happening outside our own home states has made an average number of tornadoes appear more spectacular. In just five months, the United States has experienced more tornadoes than we often get in an entire year. And far, far more people have died. 2011 is already the deadliest year for tornadoes since 1953. As of May 23rd, 498 people have been killed. That's a big jump from normal. I was born in 1981. In my entire lifetime, annual tornado deaths in the United States have only broken the 100-person mark three other years–1984, 1998, and 2008. Clearly, there is something different about this year. The question is, "What?"

The number of tornadoes is simple fact. It's relatively easy to measure. Definitely easy to report. Easy to process and memorize. The reasons behind the numbers, however, are decidedly more confusing.

When weather-related disasters happen, the first thing most people want to know is whether the disaster was caused by global climate change. And there is more riding on the answer than just another statistic to remember. These tornadoes have been painful. The destruction they've caused is visceral. If this is what climate change looks like, then maybe Americans will be forced to look at decisions about climate-related energy policies in a new light. On the other hand, if the tornadoes of 2011 aren't caused by climate change, then maybe climate change isn't such a big deal. What can we do to us that nature isn't already doing?

The trouble with looking at disasters this way is that tornadoes do not fit neatly into little, politically polarized ticky boxes. Science, in general, seldom works like that. In a May 23rd editorial for the Washington Post, environmentalist Bill McKibben took Americans to task for refusing to make a connection between environmental disasters–including the 2011 tornadoes–and climate change. His basic message: All these disasters must be connected and only willful ignorance allows us to ignore that.

I have a slightly different perspective. What we have here is not a failure to communicate and accept the obvious effects of climate change. Instead, it's a failure to communicate and accept a critical point of how science works, without which scientific literacy is reduced to mere talking points. This is about nuance and uncertainty, and if the American public doesn't get those things, then we'll never get climate change.

When scientists study climate they aren't really studying just one thing. Climate is a complex system, involving multiple natural subsystems and many variables–both "natural" and man-made–that can alter the way those systems work. This is such a complicated subject that we really only developed the computer processing power necessary to start making any sense of it in the mid 1970s. What scientists have learned since then is vitally important stuff. The Earth, as a whole, is warming as humans pump more and more greenhouse gases into the atmosphere. And those rising global temperatures, and rising carbon dioxide concentrations, will affect our lives in a variety of strange, and often surprising, ways. This is the science that should be influencing the way we plan for the future. But it's not. Not really. And I think the reason why has a lot to do with how science is taught to the vast majority of Americans, the people whose science education really ends along with the end of high school.

In this country, we teach kids that science is a collection of hard facts. We teach them that scientists come up with a hypothesis–an idea that might explain some aspect of how the world works. Scientists then test their hypotheses and find out whether it's correct or not. If it's correct, then it becomes something that children must memorize. That story is true. But it's also vastly oversimplified. It gives people the impression that every scientific question can be answered with "yes" or "no." And if it can't, then the real answer is probably "no."

That perspective might work okay when you're sitting in a high school science lab, studying the digestive system of a fetal pig. But it doesn't work as well in the real world. And it leaves people completely unprepared to understand something like climate change, and how we assess the risks associated with it.

That's because all risk–and especially the risks associated with complex systems like climate–come with uncertainty. To a person whose knowledge of science comes from that simplified story we tell school kids, "uncertainty" sounds like saying you're wrong without having to say that you're wrong. But that's not the case. Instead, "uncertainty" is about complexity and randomness, it's about probability, and it's about how you attribute the cause of one effect that is really likely to have multiple causes.

Case in point: Tornadoes and climate change. If you want a simple, talking-point answer on whether the tornadoes of 2011 were caused by climate change, the best you're going to get is: Probably not, or at least, not entirely. But there's a lot of uncertainty behind that statement, and you can't really use it to project your future risk.

When scientists evaluate the connection between tornados and climate change, there are really two big questions they're asking: First, are the 2011 tornadoes part of a trend? Has tornado activity changed along with rising global temperatures so far? Second, scientists ask whether the factors that create tornadoes have been affected by climate change, and whether those factors are likely to be affected in the future. This is where all the uncertainty comes in.

For one thing, our data on tornado trends is imperfect. At first glance, you might think the number of tornadoes has increased since the 1990s. But most of that is actually the result of better technology improving our ability to spot smaller, weaker tornadoes, and to notice tornadoes in places where few people live, according to the National Oceanographic and Atmospheric Administration. Bad data means that we can't reliably say whether tornado counts are increasing.

So, instead, NOAA looks at the variables. Tornadoes are somewhat random things. We can find factors that are associated with tornado formation–things like moisture content in the atmosphere, quick changes in wind speed and direction, and what meteorologists call thermodynamic instability, basically mixing between layers of warm air and cooler air. But just because those factors exist doesn't necessarily mean a tornado will appear.

There are many things that can affect these factors that make tornadoes more likely. Scientists have found that climate change is something that can affect tornado conditions. But when NOAA looked at data for the past 30 years' worth of Aprils in the Mississippi Valley, they didn't see evidence of any trends that would mean tornado weather is already becoming more frequent.

Because of that, NOAA says it would be problematic to claim the recent spate of tornadoes in the Southeast were caused by climate change. But that's not the same as saying tornadoes can't be caused by climate change. It's not the same thing as saying that climate change isn't a contributing factor. Or that tornadoes won't be caused by climate change in the future. It's not even the same as saying that, years from now, with better data and technology, we won't look back and see a trend happening that isn't obvious today. NOAA's assessment is based on indirect evidence focused on one area of one country. The big question–Are tornadoes caused by climate change?–is made up of lots of little questions. And we don't know all the answers to the little questions yet. This is still good science. We still have enough information to say something about how the world works. But that statement comes with a lot of caveats.

It's not really just a "yes" or "no" answer. It doesn't follow party lines. And it doesn't tell us what we should expect in the future.

This is scientific uncertainty–where the things we know and the things we don't know collide, and we are left to figure out how to use what we have to make decisions anyway. If we want people to understand science, we can't just give them facts to memorize. Scientific literacy isn't about being able to win a game of quiz bowl. It's about understanding how science works, and how science can be used to guide human decision-making. It's about knowing that we don't have all the answers. But it's also about knowing that "we don't have all the answers" isn't the same thing as "we don't know anything." If we pump people full of facts, but don't teach them about uncertainty, then we can't be surprised when they dismiss anything that isn't 100% certain.

The future of human life depends on how we respond to the risks of climate change. How we respond to those risks depends on how well the general public understands the messy world of real science.

Cross-posted as a guest blog for the World Science Festival.

Images:Tornado Weather, a Creative Commons Attribution (2.0) image from mccun934's photostream. And Image: Parker Tornado, a Creative Commons Attribution No-Derivative-Works (2.0) image from roper's photostream.