PODCASTS   ABOUT PODCASTS   ABOUT RSS

Research in NOAA

Interview with Harold Brooks
November 8, 2007

BARRY REICHENBAUGH: This is Barry Reichenbaugh. I'm with the NOAA Research Communications Office here in Norman, Oklahoma. And I'm with Harold Brooks. Harold, could you give us your title?

HAROLD BROOKS: Sure. I'm Harold Brooks. I'm a research meteorologist with NOAA's National Severe Storms Laboratory in Norman, Oklahoma.

BARRY REICHENBAUGH: Harold, your area of expertise is tornadoes. Can you go into a little bit about when and where tornadoes occur?

HAROLD BROOKS: Tornadoes have been observed in every state of the United States and in all the continents of the world except for Antarctica. And they have occurred any time throughout the year in the U.S. Now, most tornadoes occur in the late afternoon, in the springtime, and primarily in the region between the Rocky Mountains and the Appalachian Mountains.

BARRY REICHENBAUGH: What are the risks in different parts of the country?

HAROLD BROOKS: If we try to think about why tornadoes occur, we need to step back and think about, Well, what are the ingredients that we need to have in the atmosphere to make a tornado? And to start with that, we actually think about what kind of ingredients we need for a thunderstorm, because tornadoes come from thunderstorms.

If we think about the ingredients we need for a thunderstorm, we need basically three things. We want to have warm, moist air at the low levels near the ground. We want to have relatively cool, dry air aloft, and then some mechanism to lift that warm, moist air up. And as that moist air goes up, it cools off less rapidly than the air around it, then it actually becomes warmer than the air, less dense, and so it rises just like a hot air balloon.

So we need those three things to have a thunderstorm at all. And to get the kind of storm that will make a tornado, we need to have the winds increase in strength from the ground as you go up into the atmosphere and typically change direction. Usually out of the south at low levels and out of the west aloft.

And so when we think about how things work in the United States, it's pretty easy to see why the central part of the country gets more tornadoes than any other place. If we have air at low levels coming out of the south, that air is coming from off the Gulf of Mexico, bringing in lots of the warm, moist air. If we have air aloft coming from out of the west, it's coming from over the Rocky Mountains. And the best way you can get dry air aloft in the atmosphere is to lift it up over a wide, high range of mountains. And the Rocky Mountains are the perfect kind of place to do that.

And so when we bring the ingredients together of the warm, moist air at low levels and the dry air aloft, in the middle part of the U.S. we typically have the air out of the south at low levels, out of the west at high levels, and that brings everything together to make the potential for tornadic thunderstorms.

And, in fact, the central part of the United States is probably the perfect laboratory on earth to produce those kinds of conditions more often than any other place. If we look at the other places that get tornadoes frequently, there's at least one of the ingredients that it's really hard to bring together. And so it just happens a lot more often in the central part of the U.S.

Now in the wintertime, the atmosphere, in effect, has to work a little bit harder. It has to bring the ingredients together with larger-scale conditions happening, and that tends to happen more often in the southeastern part of the United States -- say, Alabama, Georgia, Florida -- in wintertime. The southeastern part of the United States has a problem with tornadoes in a different way than the central part of the U.S. To get the ingredients together, the atmosphere has to, in effect, work a lot harder. It has to bring large-scale conditions together to bring the ingredients into the same place at one time.

As a result the seasonal cycle, where we have the very strong peak in the springtime and maybe in the early summer in the central part of the U.S., we don't see that in the southeast. The seasonal cycle is a lot less distinct, and it actually has a lot of tornadoes happening in what we call the "cool season," sort of November to March.

And they also tend to get more tornadoes occurring after dark in the southeastern part of the United States, because the atmosphere doesn't really care about the diurnal cycle in that part of the country as much as it does in the central part of the country. And so while in the central part of the U.S., our threat tends to be late spring and typically late afternoon, early evening, in the southeastern part of the U.S., there's not so much of a distinct timing mechanism with the threat.

BARRY REICHENBAUGH: Okay. You've talked some about the regions that tend to get tornadoes. Can you talk a little bit about Tornado Alley and why we use that term?

HAROLD BROOKS: Okay. Well, Tornado Alley is one of those terms that if you ask 20 people to study the subject, you'll probably get 20 different definitions.

We tend to think of it as perhaps including the region that gets the most tornadoes, but also perhaps we get them kind of regularly.  You can, kind of, count on the fact that every year is going to be a pretty good year. If we just think about the part of the country that gets the most tornadoes, we'd probably draw something that looks kind of like an L or a C shape maybe. That starts with the bottom left part of the letter around Dallas, Texas, over to Midland, Texas, and then maybe up through the High Plains, up into Nebraska, maybe South Dakota.

The bottom part of the L would extend perhaps into Mississippi or Alabama, and the top part goes over towards Ohio. Now, if we want to include information about the fact that tornadoes happen almost every year and regularly, then we probably lose the eastern part of that extension, and maybe we only go as far east as Tulsa or the Oklahoma/Arkansas border, where we get a lot of tornadoes in that region that goes from Midland to Dallas and north for several hundred miles.

We get a lot of tornadoes in there, and we also have them almost every year. Most places will have a tornado come pretty close to them.

BARRY REICHENBAUGH: Let's get a little bit into how we communicate the risk, and maybe if you could talk a little bit about this Enhanced Fujita Scale.

HAROLD BROOKS: Good. Well, one of the real challenges we have in studying tornadoes is that we'd like to be able to describe how strong a tornado is. We know that not all tornadoes are the same, but that we don't really get wind measurements very often. It's a very different problem than you have when you look at hurricanes. We fly aircraft into the hurricanes. The hurricanes are over a fairly large area, and so there's a lot of surface stations that collect data.

The chances of a tornado hitting an anemometer are really small, and while we have some mobile radars that can go out and collect information inside the tornado, we still don't see very many tornadoes that way. Maybe 20 or 30 a year out of the 1,200 that happen every year in the United States, so we don't have a very big sample.

So in order to estimate how strong the winds are in a tornado, we have to look at what damage the tornado causes. And to do that there have been some scales that have been set up to try to look at that, that basically start by saying, Well, how much damage did the tornado itself do?

And the oldest and longest serving of those was developed by Ted Fujita at the University of Chicago in the early 1970s. And for years the National Weather Service rated all tornadoes on the Fujita Scale, which was based upon what kind of damage would be caused to a well-built frame home. Well, that causes some problems, because if you're an engineer, you look at the house and you go, Well, that house, maybe, wasn't really well built; or, This other house was really well built.  And so maybe the differences we see in the winds are because, actually, the houses were built differently. If the house wasn't attached to the ground very well, then it might fall apart easily.

So in 2007 the National Weather Service, after a several-year project to try to improve the Fujita Scale, brought out what was called the Enhanced Fujita Scale, which instead of basing the damage off of just one kind of thing, the well-built frame home, now it uses a large number of damage indicators and a large number of degrees of damage.  Was the house barely hit or was it completely destroyed? Were the shingles taken off? Was it a gasoline station awning that was removed?

And so now damage estimators can go out and have a much broader range of techniques to try to apply to it. Now, it still needs, probably, some improvement, because it was not an engineering rated study; in other words, it wasn't the kind of thing that engineers sat down and wrote equations out and said, Here's the amount of force it takes to blow this object down. And so we still have some issues about exactly how to do it. And so hopefully it'll be improved as it goes along.  But it does allow estimators a much broader range of information to use when they try to estimate the wind speeds.

BARRY REICHENBAUGH: Let's touch on some general information about some tornadoes, what we know and what we don't know.

HAROLD BROOKS: Sure. Okay. Well, the biggest thing we know about tornadoes is the kinds of storms that'll make the strongest tornadoes. We call them supercell thunderstorms. And in fact in a supercell, the updraft of the storm, the part of the air that's going up, is in fact rotating throughout the entire depth of the storm, from just above the ground maybe up to 50,000 feet above the ground.

And that's one of the things that makes them an attractive target for radars, is because with Doppler radar, we can see the motions inside that storm. So they've been studied for a long time. And although we don't have precise numbers, it certainly looks like that almost all of the strong tornadoes, the ones that will actually cause damage to a house, come out of supercell thunderstorms. And so they become a particularly important kind of a problem.

And we understand the large-scale conditions that lead to the development of those kinds of storms. Something we don't understand quite so well is what makes a supercell produce a tornado, versus a supercell that doesn't make a tornado. And we have a pretty good idea of the kinds of conditions that lead to that. We need to look at things that are happening very near the ground, both in the environment of the storm and within the storm itself. And that makes them harder to study, because now instead of being able to look over this large depth of the storm and a large depth of the atmosphere, we have to have really precise kinds of measurements down there in the very lowest part of the atmosphere.

The second thing we don't really know is exactly what happens in the last few minutes before the storm makes a tornado. The time between when some things start to look like they're potentially tornadic until the storm actually makes a tornado can vary from a few minutes up to half an hour or more within a storm. And so that's a real problematic aspect.

We also don't know why tornadoes die. In many ways it's easier to understand how a tornado starts than how it actually ends.  And it would be a big help for warning forecasters if we actually start to predict, Well, this tornado's about to end. Then you wouldn't have to put a warning out for, say, the next county, because you knew that the tornado wasn't going to be happening.

And I think the final thing that we don't really understand is why tornadoes are as intense as they are. It's very difficult to know, watching a storm, Will this storm make a very strong tornado or will it make a weak tornado? And in fact even knowing what other tornadoes a storm has produced, if it's producing more than one, you don't always know how strong that next tornado is going to be. And if we could actually learn that process that determines the intensity of the tornado, then we might be able to tailor the warning statements to be able to say, Well, this storm is particularly dangerous. And you really, really need to be doing very special things to protect yourself. Or, This is going to be a relatively weak tornado, and if you're inside a well-built home, the wind's going to blow pretty hard and you might get a window broken, but there's not a real big danger to your life.

BARRY REICHENBAUGH: Sounds like there's still room for some more research.

HAROLD BROOKS: Oh, there's still lots of room for research.  We don't theoretically understand everything that goes on in the tornado particularly well, and we know that we have to do a lot better job with the observational things. I think the biggest thing that we've learned in the last 15 years is that the kinds of questions we ask now are much more detailed than the questions we asked 15 years ago. And if you go back even further, if you go back, say, 40 years ago, we weren't even sure what part of the thunderstorm tornadoes came out of. Now we're asking really narrow questions.

BARRY REICHENBAUGH: Tell me how you got into this field.  Where did you get the inspiration?

HAROLD BROOKS: Well, I kind of stumbled into meteorology in an offhand way. I was always interested in science and math, and those were always, I thought, the easiest subjects possible through school.  And when I was in college I was a Physics and Math major. And as I was looking around for what I was going to do, I wasn't real thrilled about what I saw as the job prospects and exciting topics in physics; they didn't match up with what I thought was the fun part of physics.

And in my junior year in college, my undergraduate department chair said, ‘You really ought to do something over the summer. This summer after your junior year, before you go to graduate school, try something out.’ And he suggested that I go to the Summer Institutes on planet and climate at the NASA Goddard Institute for Space Studies in New York. A previous graduate of our college had gone there, had become a world's expert on black holes, and he said, ‘You ought to go there. It really helped Don a lot.’

And so I decided, ‘Well, that sounds like a good idea.’  And I went to New York and spent the summer there and actually worked on climate modeling. I put in the boundary conditions for modeling the last glacial maximum, the last ice age, into a climate model and ran those. And then Columbia invited me to come back for graduate school, and so I went back and we had discussed whether I was going to do it in physics or in geology. And they said the geology department looked like it was a better fit, and so I actually looked at how volcanic aerosols are moved in the atmosphere.

And so what happens is if a volcano erupts, what's the strongest signal it has on cooling the earth if you -- you know, where's the best place to put it? And after a few years of that I wasn't completely comfortable with what I was doing right then. I'd grown up in the Midwest and thunderstorms were always a big part of the summer fun.

And I had the opportunity to come to the University of Illinois with Bob Wilhelmson, who is probably the leading person in the world on the numerical modeling of thunderstorms. And that sounded like a fun thing to do. And Bob and I were probably about as good of a match of advisor and advisee as you could hope to have. We got along great.  Did some work on modeling of why some storms make tornadoes and others don't make tornadoes.

And then I actually came to Norman simply because of Chuck Doswell, who was a scientist here at the Severe Storms Lab. At a conference I had been interacting some with forecasters, and there aren't a whole lot of researchers that actually like to spend time with forecasters, because forecasters and researchers think really differently. And Chuck said, ‘Well, before you go off and get your academic job, why don't you come to Norman, come to the Lab for a year on a post-doc and come do some things.’ And the post-doc was a two-year opportunity, and so I came down for the two years and was fortunate enough to be offered a federal position when I got done with that, because I decided that interacting with forecasters and doing those kinds of questions was actually a whole lot more fun than what I would have in the academic world. And so I've now been at the Severe Storms Lab for over 15 years as a research meteorologist.  I think it's been just a ton of fun.

I talked with friends who seem to have a lot more job stress than I do. I have to deal with things like a few little bureaucratic/management things that I have to do that take about 5 percent of my time that aren't much fun. But the rest of the time I get to work on whatever I want to and work on things that I think are important and interesting.

BARRY REICHENBAUGH: What do you say to a person who's considering a career in science?

HAROLD BROOKS: Well, I think one of the things about science that's a lot of fun is you get to solve problems. You constantly get to think about things that you want to think about and to try to find answers for things.

And for any job it's important to actually enjoy what you do, because otherwise you're pretty miserable. And I think one of the really fun things about science is the notion of discovery and that at some point, every scientist gets the excitement of realizing that they know something that no one else on the planet knows, and then they get to tell people. And that's just a whole lot of fun.

When you realize something's true that no one's ever known before, and then you get to stand up and tell people about it or write about it. And people go, Wow, you learned something important. And that's just a whole lot of fun. And it happens fairly regularly.