University Place

cc >> Welcome, everyone, to Wednesday Nite at the Lab. I'm Tom Zinnen. I work here at UW Madison at the Biotech Center here. I also work for UW Extension Cooperative Extension, and on behalf of those folks and our other co-organizers, Wisconsin Public Television, the Wisconsin Alumni Association, and the UW Madison Science Alliance, thanks for coming to Wednesday Nite at the Lab. We do this every Wednesday night, 50 times a year. Tonight, I'm delighted to have Professor Jonathan Martin here to talk to us about tornadoes and derechos. One spins like this; the other spins like this. That's how much I know so far, and they both inspire admiration and fear.

LAUGHTER

Professor Martin joined the faculty at UW Madison in 1994 after completing his PhD in atmospheric sciences at the University of Washington. He is currently nearing the end of a nine-year term as chair of our Atmospheric and Oceanic Sciences department. Professor Martin has received numerous accolades for his teaching, including the Underkofler Excellence in Teaching Award and a fellowship in UW's Teaching Academy. He was chosen for the prestigious UW Vilas Distinguished Service Professorship for distinguished scholarship and excellence in teaching and service. The Princeton Review recently ranked Professor Martin among the top 300 best professors in the nation, and I know quite a few people who think it ought to be top 100. Professor Martin's research expertise is in mid-latitude weather systems. Over his career, he has contributed to numerous scientific writings and he authored the leading textbook on atmospheric dynamics. You also know him as one of the two Weather Guys that appear on Wisconsin Public Television, excuse me, Wisconsin Public Radio, along with Steve Ackerman, his colleague. And I think the folks in Wisconsin are a lot more science savvy when it comes to weather because of those two gentlemen than anybody, possibly with the exception of Increase Lapham.

LAUGHTER

>> Yeah, there you go. >> Did I get that right? >> I think so. And he's younger... >> Oh, Increase has been gone for a while! So please join me in welcoming Jon Martin to Wednesday Nite at the Lab.

APPLAUSE

>> Thanks very much, Tom, for that generous introduction. In case you're kind of puzzled between myself and Steve, I'm the one who speaks English on the radio. >> Whoa...!

LAUGHTER

>> That might be the way you can distinguish us as I go forward. Actually, Steve and I had the opportunity to be on the radio this Monday, and it's always exciting. We've been doing it for about 15 years, and it started, actually, in the summer. >> How many years? >> 15, since 1998. >> Years? >> Years.

LAUGHTER

Oh, that's another thing, if you don't quite understand my accent, don't worry, I won't hold yours against you.

LAUGHTER

But you can ask me to clarify something. I'll do my best to do that as we go forward. So we're on the radio on Monday, and we're getting all kinds of questions about the heavy rain we had over the weekend and various things like this, and it made me think that this really is a weather conscious part of the country. I've lived in a couple of other parts of the country. Growing up in New England, you're kind of conscious of other types of severe weather. It's not so much thunderstorms that brings tornadoes and derechos. Although, that does happen there as you'll see. But it's more snowstorms in the wintertime and the threat of hurricanes in the late summer and early autumn. And then in the Pacific Northwest, it's really the threat of boredom that you worry against.

LAUGHTER

When it comes to the weather in the Puget Sound, the weather is about as boring as there is any place on Earth.

LAUGHTER

However, if you've never been there in the summer, that's the best kept secret in the world. That's why people live in Seattle and Portland is because summer is so beautiful. So anyway, it's a pleasure to be here tonight, and I want to tell you a little bit about why there's really no place like home. In this part of the globe, and I mean the whole globe, there's no place like it for the kind of severe weather that can develop and the threat that that kind of weather brings. So here's a beautiful picture of a tornado. I've never seen one. And it's not because I've conspicuously avoided it, although I haven't really aggressively sought to find one, either. But I would like to see one sometime, although sort of the same way I'd like to see a

dinosaur

from a safe distance.

LAUGHTER

dinosaur

The word tornado is really an amalgam, as far as we know, of two different Spanish words. One of them tronada, which is the Spanish word for thunderstorm; and the other is turno, the Spanish for to turn or to twist. And here's a picture of a tornado from Nebraska, and the official definition of the tornado from the glossary of meteorology is a rotating column of air pendant from a cumulonimbus cloud, that is a thunderstorm cloud, and nearly always observable as a funnel cloud or a tubal, which is another kind of noun for funnel shaped structure. So they are undoubtedly of great interest. They are awesome displays of nature. They absolutely are one of the most awesome things that one could ever hope to see in the natural world, I think. The first North American observation of what was at the time called a whirlwind occurred over the Albemarle Sound in North Carolina on the 23rd of June, 1586. This was the famous lost colony of Roanoke. Sir Walter Raleigh's experiment in the new world. And this was their first summer there, I think. This is also a water spout right over the Albemarle Sound from the current day. So this might have been what they actually saw. Something quite like it. June 23rd, 1586. So for almost 450 years, at least European descendent people in this country have been observing these things and probably hadn't seen them before they came here. The first time the word tornado was actually used to describe a storm that conforms with the way we use it today was in a storm in Killingly, Connecticut, on the 23rd of August, 1786. And it's really interesting when one begins to read some of the accounts of storms in the colonial period, this is of course just the federalist era, but just before in the colonial period, the words hurricane and whirlwind and a variety of other types of words are thrown at you in descriptions of these kinds of events when you're probably pretty sure that it's a tornado or a derecho. And so hurricane was the word of choice for strong winds. And it was used with abandon. And not until about the end of the 18th century did people start to fixate on the word tornado for these vertically rotating columns of air. It was generally used across the board by about 1840, but you can even see in reports from the 1790s and the early part of the federalist era that the word tornado was making its way into the language by then. Certainly well established by 1840s. Interestingly, I just heard this the other day, I was just in Washington last week, the British of course took, I think it was 1814, I might have the wrong year, August 25, 1812, let's say, the British had just invaded Washington, DC. It was brutally hot and humid in the middle Atlantic in the week leading up to that. So that cost them lives on their side of the battle because the British soldiers were not used to this kind of weather. And then after they set the city afire, including the White House, they were visited by a tornado that night, and it killed several more British soldiers. So Mexico has Montezuma's revenge; the United States has a different kind apparently.

LAUGHTER

dinosaur

All of these type, both species of severe weather, tornadoes and derechos, are born out of thunderstorms. It's worth our time to talk about the two species in terms of thunderstorms. There are a benign species of thunderstorm, the kind that might rain heavily on you in the afternoon, but really poses no other substantial threat. That is, no wind threat, no hail threat usually, and nothing more substantial than just heavy rain, which, of course, can be a major threat in and of itself. Those kinds of thunderstorms are known as air mass thunderstorms. And this is a schematic of what they look like. They just kind of look like, before they start to rain that is, they look like a pile of mashed potatoes. Fair weather cumulus cloud, and you can get this on any sort of warm, muggy, summer day. Usually after about noon time because one of the mechanisms to drive the development of the cloud is the daytime heating of the ground underneath. And it can make a local hotspot in the atmosphere that will promote the growth of clouds. And the way that clouds grow is that warm, moist air from the lowest levels in the atmosphere is forced to rise by the updrafts, perhaps, of one of these local warm spots. And these red arrows, they're supposed to be red, they're really red on my screen and they don't look so red here, but anyway, they'll be distinguishable, I hope, from the other ones. These red arrows signify the ascent of that moist air. As air rises from one elevation to a higher elevation, the pressure goes down as air rises. If you climb the top of this building, the pressure on the roof is lower than the pressure at the base of the building. So pressure always decreases with height, and as an air parcel rises into an environment of lower pressure it expands. It doesn't have a choice, but it expands into that lower pressure environment, and that expansion costs it energy in the form, most predominately, of temperature. So the air temperature cools as the air rises and expands into its lower pressure environment. The cooling is sometimes sufficient to transform what's invisible water vapor, no more visible than nitrogen or oxygen gas, into observable tiny water droplets. And then those water droplets, through a variety of processes, can grow and turn into precipitation size particles and fall out of the bottom of the cloud. In this depiction, we're just showing the initial cloud growing stage where only very small, tiny liquid water droplets called cloud liquid water droplets are growing in these updrafts. It takes a million cloud droplets to form a single average sized raindrop. So the next time you get caught in a rainstorm unaware, while you're cursing you might want to remind yourself that a million cloud droplets have somehow gotten their act together to produce any one of those single raindrops that's falling on your head. So if that's not a miracle and doesn't stop your clock, really what will?

LAUGHTER

dinosaur

It's pretty remarkable. It's pretty remarkable. So, once this starts happening, it takes a long time. Oh, I should say, the whole life cycle of an air mass thunderstorm is maybe an hour or two. So when I say a long time, I mean a large fraction of the whole life cycle. So a long time before this cloud can actually start to produce some of those large enough particles to fall out. But once it does, it starts to rain. Maybe there'll be a little ice mixed in, usually not. That rain is going to induce a downdraft. I think you can tell the difference in the color. So the blue arrows are air that's sinking. And notice that a couple of those arrows in the schematic are actually drawn from air outside the cloud. That air is unsaturated. And when it gets mixed inside the cloud where most of the air is saturated, it's going to hasten the evaporation of some of those liquid water droplets, whether they're precipitation sized particles or just cloud droplets. And that evaporation cools the air and makes it more dense than the air surrounding it and it sinks. So, if you've ever stood at the base of a waterfall in the summertime, hopefully you've been able to do that, if you haven't, you should try to do this, and I don't mean right underneath it so you're taking a shower in it. And, in fact, I don't mean so close that you feel water droplets coming off of the waterfall fountain hitting the stream. I mean just close enough that you can feel the evaporative cooling that's going on. That's what's going on in these clouds to help produce this downdraft. And there's still a little bit of updraft, but notice that half of the cloud is now overcome by downdrafts. And, in fact, if you keep on going, this is maybe all in one lunch period. If you're lucky, you can see this whole thing happen. It's true. You can actually see this whole thing happen during your lunch hour, if you're lucky. By the time you have to go back to work, the entire cloud is overwhelmed by downdrafts. So the downdrafts in this benign air mass thunderstorm strangle the updrafts, and that is really the end of the story. That's why they only last about an hour. These do not spawn any significant severe weather threats. Not only because their lifetime is short but also because there's no way for updrafts and downdrafts to cooperate in a way that might help to produce some of these more significant and severe weather threats. They actually work against each other in this environment. In the atmosphere in the middle latitudes where we all live, there are very large scale processes and structures that conspire to help to produce some of the lifting mechanisms that can force the air to start to rise and thereby produce clouds and precipitation also that can help to provide certain other elements of the environment necessary within the development of a thunderstorm environment to help make it a longer lived storm and one that has the possibility of spawning much more significant weather. One of those environmental factors is among the most basic observational characteristics of our spherical Earth. We're 93 million miles away from the sun. We're tilted at an angle to the sun right now. In fact, we're just a couple of days past the maximum tilt of the northern hemisphere towards the sun, longest days of the year are the result of that. Since the Earth is spherical, the high latitudes, near the North Pole, don't absorb as much radiation as the low latitudes. And so over the course of the year, especially, it's almost always colder as you head northward in the northern hemisphere. In fact, it always is colder the further north you go at any level in the atmosphere and always warmer as you go southward. So that there is what we call a pole to equator temperature gradient, and I've represented that schematically here. The dashed lines are lines of constant temperature even though I haven't labeled them. And the air is cold to the north, the so-called tropopause, which is the boundary between the troposphere, the lowest part of the atmosphere in which we all live, and the next highest level, the stratosphere, is lower over a cold column of air near the pole. And the warm air to the south has a correspondingly much higher tropopause, and the slope of that tropopause is what forces there to be a jet stream. And so the jet stream is always connected to the regions of, the strongest jet stream is connected to the regions of the strongest horizontal temperature contrast. And that's indicated by those dashed lines. Because that horizontal temperature contrast manifests itself as a very sharp slope to the tropopause. So there's a lot of ways to look at that. But these jets are local regions of intense horizontal wind speed, and they're not only horizontal circulations associated with them, they also have vertical circulations associated with them. We don't have enough time to really talk about all those details, but some of these people in the room tonight already know it. We can spend a whole semester talking about those things. And, believe me, it's fun to do that because you learn something new every time you start talking about it. But these jet structures are very, very common in the mid-latitude atmosphere. And they embed themselves within a series of waves in the mid-latitude atmosphere. This is a schematic of the atmosphere maybe three to five miles above the surface. I've put the north, south, east, and west so you know it's a horizontal map. And the atmosphere in the mid-latitudes all through the year, but especially in the cool season, spring, fall and the middle of the winter, has lots of waves in it at latitudes like the latitudes of southern Wisconsin. Those waves moves a little bit further north in the summertime, but they're always there. And within these waves you might have a local region of very high wind speeds, a little jet streak that's sitting on the upstream side of what we call a trough. So whenever these lines bow downwards toward the equator and then go back up, this is a trough axis. Over here is a ridge axis where they go up and around in a clockwise fashion. And this jet is depicted as occupying the northwesterly flow upstream of a trough axis. And it turns out that regions of low pressure, where the L is, develop downstream of the upper trough axis, and regions of high pressure develop downstream of the upper ridge axis. These are well established rules, and there's very good physical reasons why they are true. We'll get back to that much later in the talk. The severe thunderstorm, of course, is what brought you all here today. I want to talk a little more about that. One of the things that differentiates the severe storm environment, so that is the environment within which the severe storm is going to form, is different from the environment within which that benign air mass thunderstorm is going to form is there on the right side of the diagram. The winds at different levels in the atmosphere have different speeds. Sometimes they have different directions in the severe storm environment. And that difference, as you change in the wind speed or direction with height, is known as vertical, because you're going in the vertical direction, wind shear. Wind shear is a description of winds changing direction or speed. But if I put the adjective vertical in front of it, I'm talking about how they're changing in the vertical direction. So the environmental winds in the severe thunderstorm environment have substantial vertical shear. The jet stream that I showed you just a minute ago sits atop a column of air that has got very strong vertical shear in it. In fact, so strong that at the top of that column, the winds are stronger than anywhere else. That's why you have a jet stream. So the jet and the vertical wind shear are kind of tied together. So imagine what would happen if you superimposed this vertical wind shear on an environment in which there might be convergence of very warm, very moist air in the lowest part of the atmosphere. Well, it might force that updraft to go up at an angle because it will go roaring in at the bottom and it will kind of, from the east in this diagram, and then it won't really be roaring away to the west at higher altitude. So it kind of spreads out and slows down as it goes up. The result is that the updraft is tilted instead of straight up and down. As this updraft air is lifted to higher elevation, it's cooling by expansion and it's going to start to produce precipitation sized particles on the backside of that massive cloud. And that's where the precipitation will form in a severe thunderstorm. On the backside of the cloud. The down shear side. Or I guess actually the up shear side. And, again, I've got some of the air that's in the downdraft coming in from outside the cloud. It's dry air. It mixes in and makes the air heavier than it would be by evaporation. I might get some hailstones in a severe storm environment. The thing I have missed describing that you might remember from the description of the air mass thunderstorm is that there's no competition between the updraft and the downdraft. If the updraft is tilted, then the downdraft air stays separate from the updraft air for a long time, which allows these storms to last six, 12, 18, 24 hours. And they rain heavily. Imagine having a severe thunderstorm, a super cell sitting right on top of you for 12 hours. Watch out. You're going to have 15 or 20 inches of rain, and there's no place for it to go in most parts of this country except into your house or something like that, which you don't want. So, the lack of destructive interference between the updraft and the downdraft produced by virtue of the vertical shear of the environmental winds is the major characteristic that differs between the air mass and the severe storm. And so all this ice and liquid water that's falling out the bottom is subject to evaporation if it's liquid, sublimation if it's ice, and you're going to cool the air in the bottom and you'll produce a gust front. And you've all felt one of these if you've had the guts to stand out in front of one of these thunderstorms as it comes near you, and it's exciting to do that. You can feel that big gust of wind and the temperature drops 10 degrees before it starts to rain. That's the leading edge of the gust front, and that air has been cooled by evaporation and sublimation of water and ice coming out of that storm. And it's also one of the features within the severe thunderstorm environment, broader scale than just one storm, that can actually be a spot where you get secondary and tertiary development of new thunderstorms because it's a pretty sharp boundary and it provides some of the convergence that can force the air to rise in the storm. Leonardo da Vinci had a notion that gust fronts ought to exist. He had pictures of them. Not only in his head, he had a lot of things in his head, but he drew some of these. So they must have happened somewhere in central Italy occasionally. So that's the picture of the severe thunderstorm environment. Look at this. This is the annual average severe thunderstorm watches per day, I mean per year, sorry, it's number of days per year. I was going to say per day. I said the wrong word. Number of watches per year from 1993 to 2012. So it's about a 20-year period. And it's clear there's something quite special about Oklahoma and Kansas. They have upwards of 20 severe thunderstorm watches per year in the last 20 years. The maximum is 23 in Osage County, Oklahoma. 23, so divide 365 by 23. It's just less than 14. So every two weeks you have a severe thunderstorm watch in Osage County, Oklahoma. That's unbelievable. That's just unbelievable. So what we're going to investigate a little bit today is what makes that part of the world so special. I haven't said anything about instability yet. I've talked to you about the dynamics that help drive the development of these storms, and the environments within which they form is characterized by a certain dynamical structure. We will eventually start talking about some of the thermodynamics, the stability of the atmosphere, and it's an unusually easy spot right there in the Central/Southern Plains of the United States to produce poorly stratified atmosphere and really drive the development of extreme weather. No place like it in the world like that, especially from the stability point of view, and I'll point out why when we get there. >> What keeps them out of the far west? >> The mountains. The Rocky Mountains. They are so massive, they have weather systems coming across, generally from west to east, again conforming with colder pole, warmer tropics so there's a westerly vertical shear. The jet stream has usually got a westerly component to it, and the mountains are so massive that what they do is they effectively trap all the low level moisture from the Gulf of Mexico right up against them. So everything happens where the air is moist in the lowest part of the atmosphere, and not much happens anywhere else. Good question. So what's another effect of vertical wind shear? Well, let me just, again, some schematics. This is a vertical cross-section that I'm trying to extend into the page. That's what the dashed lines indicate. So I'm showing you just the bottom and top part of the wind structure that I showed you just a minute ago that characterizes the severe thunderstorm environment, the vertical wind shear. And let's just put it on the back side too so that it extends infinitely into the page just as a means of imagining that. I hope you get the point of that dimensional diagram. Then, if you take anything, like a pencil, don't take your cup, although it would work, it's full, anything that you have, just imagine extending that pencil all the way along the line that goes into the page. And then having those winds at different elevations act on other sides of that. One side of the pencil, the bottom side, will be spinning to the left fast; the upper point won't be pushed as fast to the left. So that whole thing, a long, linear, maybe like a strand of spaghetti aligned along that axis into the board is going to present itself as sort of a tube. And it's an infinitely long tube of air. And if you let that tube of air get spun around by those winds, it's going to have a certain rotation to it, right? It's going to rotate like that. That's why if you have a pencil you should do this. Imagine the wind going away from you that way faster and slower that way in the same direction. It's going to spin like that. Okay? This is all before there's any updraft. There's not thunderstorm. This is just the background environment that's provided by the presence of the vertical wind shear. Now let's put an updraft that will help to start develop a thunderstorm into that. And there it is. And that's going to distort that tube. This tube is not made out of cement or wood or plastic. It's malleable. It's air. So you got a rotating tube of air that's definitely deformable by that localized updraft. It's going to start to bend that tube into two pieces, one of which spins in one certain direction, the other one spinning in the other direction And if that updraft persists, it's going to actually make ends of that tube, which used to be oriented perfectly parallel to the ground in a horizontal form, parts of that are going to be oriented vertically. And you're going to start to see that same rotation that had been in the horizontal reoriented into the vertical. And suddenly, you look up, you're underneath this massive super cell thunderstorm, and it's rotating. And you can see the rotation on that thing. All right? And that's scary. That's really scary. What happens next is still a bit of a mystery. How does this thunderstorm scale rotation, that's all I'm talking about here, how does that vertical rotation get a focus down into the local funnel of the tornado? We don't really know the answer to that yet. We have some pretty good ideas, but we don't really have the full answer to that problem yet. So if there's anybody out there who's thinking I like tornadoes, I'd even like to maybe devote my life to studying them, there's plenty to study. You won't run out of it. It won't be like you get to be 35 and all the answers are there, and you'll say what do I do now? That won't happen. You don't have to worry about that. We don't know how that happens. In fact, even with the sophisticated numerical models of the day, and I'll mention something more about these later on, the way we can model atmospheric phenomenon in a significant way in the current era, only my colleague Greg Tripoli has been able to create a tornado funnel in a very high resolution expensive numerical forecast of a severe thunderstorm, and he admits he had to cheat. Don't tell him I said that.

LAUGHTER

dinosaur

But he knows he had to cheat. He had to use this vorticity confinement theory which is a little bit of cheating to make it work. Pretty unbelievable. Nature doesn't cheat.

LAUGHTER

dinosaur

Pretty unbelievable. And, let me tell you, Greg Tripoli is about the smartest person I've ever met. If he can't figure it out without cheating, nobody can. There's no place like home. So, this picture looks great on my screen, not so good here unfortunately, but I hope you can see the green is agricultural areas and the red dots are locations of tornadoes. They only got up to 1985. And this by the famous University of Chicago scientist Theodore Fujita, after whom the Fujita Tornado Intensity Scale is named. He spend most of his whole career studying tornadoes. So this goes up to 1985. What's so special about the Central Plains in North America? And I say North America because our Canadian relatives just to the east of the Canadian Rockies have their share of tornadoes. The Edmonton Tornado was a mile wide in July 1988. So they've seen them. They are quite familiar with them up in prairie Canada. And prairie United States unbelievable how much more frequent and intense tornadoes are in this part of the world than anywhere else. There really is no place like it, and there has to be a reason for that. And I'm going to tell you what I think it is. Here's a picture, a schematic again, of an upper level wave just like one of the waves I mentioned to you earlier. In the late part of the winter and early spring, those waves can make their way as far south as south of the Arizona New Mexico border with Mexico. So this picture is not far-fetched. It's quite a common occurrence in March or maybe early April. And there's two physiographic features of great importance for the development of this maximum of severe weather tornadoes in the Central/Southern Plains of the United States. One of them is the Mexican Plateau which is an elevated heat source, effectively. It's 10,000 feet high. The air is bone dry at that elevation, and it's very warm because it's right on top of a rock face. Then you've got the Gulf of Mexico. Some of the warmest water in the western hemisphere and almost all times of the year, but that's at low elevation. So what happens is when a wave like this, and you can imagine the wind is parallel to all those red lines, those are streamlines of the wind, so as the wind comes out over the plains in this fashion, first of all, just to the downstream side of the upper level low, which I've depicted in red, is where you're going to have your surface low, and I've put that in blue. I wish it would have turned out a little more in blue. And so that's one thing that's going to work. The weather systems tilt back to the west with height. So you get a surface low further to the east than the upper level low. But the upper level low is dragging air, kind of that dusky orange color, right off of the Mexican Plateau at the same elevation as the Mexican Plateau out over the Southern Plains of the United States, 10,000 or so feet, maybe 8,000 feet. It's going straight out over the plains. It doesn't head northeastward and sink, it just heads northeastward and stays a level that it was generated. So it moves right out over there. Meanwhile, the low level cyclone has its own counterclockwise circulation, and the flow associated with that in the lower part of the troposphere comes off the Gulf of Mexico. And that's warm and moist. So this place is the best place on Earth for putting together the ingredients for what's known as convective instability. The most potent form of atmospheric instability that our atmosphere has. You put warm, dry air on top of warm, moist air, and then lift that whole column of air and it explodes. It's like leaving your gas tank open in the machine shed in your backyard. You don't do that. That's kind of what's happening here. And what provides the mechanism to lift the air? Well remember, downstream of the upper trough the air is generally lifted for reasons that we can't explain tonight. But that's what's going on. So it's a perfect storm of circumstances to produce this maximum in frequency, intensity, and of course across the board times of year when you get these events. So it's really special. I've often thought about this. I used to tell my kids when they were small, we'd read books about dinosaurs or something like that, and I'd show them a map of the US; do you know that in the central part of the United States there used to just be a shallow inland sea, and on the shores of that sea you had dinosaurs eating swamp grass or whatever it was they were eating. They must have seen an endless array of water spouts go flying across this inland sea because these circumstances weren't different. The Gulf of Mexico was still really warm, the Mexican Plateau was still probably pretty high, and an elevated dry, warm air heat source, and they were all put together by the progression of waves across the latitude of about 35 degrees. Must have been awesome. I could've had have both my things at once then. Here's tornado activity in the United States. And absolute clear maximum right there in Oklahoma and Kansas. There's a secondary maximum you might imagine in sort of northern Alabama/Mississippi. There actually is, they call it Dixie Alley. The one in the Central Plains is called Tornado Alley. Everybody's heard of that. Dixie Alley is the one in the southeast United States. And there's a pretty good reason why that's true, but you'll notice something interesting about Dixie Alley in just a minute. And here in Wisconsin, we are on the northeast fringe of the most intense and most frequent tornado occurrence. Although, believe me, moving out here I'm certainly more attentive to the tornado threat than I was in Boston and certainly more than I was in Seattle.

LAUGHTER

dinosaur

So I pay attention a little bit more now. Plus, I own some property. That's different from when I was 15.

LAUGHTER

dinosaur

All those things conspire to make me just a little more nervous, but with good reason. There's no joking about that. When do these tornadoes occur? Well it's really quite remarkable. The Tornado Alley in the Central Plains has a huge seasonal spike. So the X axis, or the bottom axis here, is just dates of the year. January 1st all the way to the end of the year. There's a big spike right about May 20th, 25th in Lubbock. And then not much happens before the 1st of March and not much happens after the 1st of August. So there's this gigantic spike for the Central Plains and Lubbock is kind of representative for a lot of stations in the Central Plains of the United States. If you go to Hattiesburg, Mississippi, however, it's very different. Even though the total number of tornadoes might be roughly the same, and it is roughly the same, there's no obvious single season spike. There is a big one around the end of the winter, earliest part of the spring. So we're talking about late March and early April. That's the peak tornado season in the southeast. And then the secondary peak is somewhere after Thanksgiving. Can you imagine that? And then there's a threat at almost all times of the year with the exception of maybe July, August, and September at which time the hurricane threat takes over.

LAUGHTER

dinosaur

So if you don't want you property damaged, you don't go down to Mississippi because there's just too many threats, but pretty amazing. And there's a little bit more variability, the different colored lines surrounding the black line represent different five-year periods. So there's a lot more variability in the occurrence time of these tornadoes in the south than there is in the Great Plains. So that's very interesting. What about Wisconsin? So here's a guy named Jonathan Carver who wrote a book called "Travels Through the Interior Parts of North America, in the Year 1766, 1767, and 1768." That's a long title. Look at this passage that he has in his book, and I'm going to read it for you because they have that funny "S" from the 18th century in there. "Here a most remarkable and astonishing sight presented itself to my view. In a wood, on the east of the river, which was about three-quarters of a mile in length, and in depth farther than my eye could reach," these guys were paid by the word, "I observed that every tree, many of which were more than six feet in circumference," so a foot in diameter, or a foot in radius, two feet in diameter, "was laying flat on the ground torn up by the roots. This appeared to have been done by some extraordinary hurricane that came from the west some years ago but how many I could not learn, as found no inhabitants near it." A hurricane. Here's a use in the 18th century of the word hurricane willy-nilly. No hurricanes in Chippewa Falls. So this was either a derecho or a tornado.

LAUGHTER

dinosaur

And he saw only the aftermath of it. And I think it could have actually been a tornado or a derecho if he had told us how the trees were laid down. It doesn't exactly say they're laid down all parallel. If that were the case, I'd say this was a straight-line wind event, a derecho. And this is often referred to as the first account of the aftermath of a tornadic storm in the state of Wisconsin. It may not be accurate because I think it might be a derecho. But anyway, it was near present day Chippewa Falls, Wisconsin. The most famous Wisconsin tornado, I think, of all time was probably the New Richmond, Wisconsin, tornado. June 12, 1899. Here's a picture of a bunch of guys after the storm went by, and the word that they use in the caption is this tree was disembarked. That means it had the bark ripped right off of it. Very interesting. And you can even see some pieces of it. And they're looking at it like wow, that's pretty amazing. And look at this house behind them. Totally destroyed. Here's part of the town. Of course the building structures were not, of course, as well fortified as they are today, but nonetheless, this is absolute devastation. And here's a poor lady and her daughter maybe looking at this house saying what do we do next? There's no FEMA.

LAUGHTER

dinosaur

Really. What did you do in 1899? We've come a long way in this country. We don't want to take any steps backwards. That's a disaster. Deadliest Wisconsin tornado of all time. 117 fatalities. 125 people injured. The next on the list is the Colfax tornado, June 4th, 1958. And here's the map for where it was. This was an E5 tornado, enhanced Fujita scale category 5 tornado. The enhanced is just to reflect the fact that now since building codes have changed since the Fujita scale was introduced, if it gets to F5 now it's really powerful because it's going against even better constructed buildings. That's one way to think of it. So this EF5 tornado was on the ground for about 30 miles near Menomonie and Chippewa Falls. 21 people died this day. 78 were injured. And here are other EF5 tornadoes in Wisconsin. There was one in Darlington the 22nd of May, 1893. Marathon County had one on the 18th of May, 1898. Waupaca had one, 26th of September, 1951. So look at the variation in time of year where these massive events can occur. Barneveld, of course, probably the most infamous tornado for those of us that live around here. It's just up the road. It happened in the middle of the night. That's somewhat unusual. Must have been horrifying for people who lived there. June 7th into the 8th, 1984. And then the Oakfield tornado, July 18th, 1996. And I can tell you I found a really great video on YouTube of the Oakfield tornado, but from 1996, we get so easily used to the fact that current day videos are outstanding no matter what they're of. If it's your kid's graduation or first birthday party, whatever it is, they're all better than they were in 1996. So this is kind of a grainy video by today's standards. You should look it up. Just put Oakfield, Wisconsin tornado video in Google. It will come up top of the list. I didn't bring it because I couldn't embed it in my presentation. But you can actually see the rotating super cell before it drops the funnel. So it's one of the most dramatic, scientifically useful videos I've ever seen of a tornado because it shows the origin of the tornado funnel as well as the aftermath. And since about 1950 there have been about 1300 tornadoes in Wisconsin. Most of them of the very weak, F0 to F1 to F2 variety. So only a very small fraction have been in the F3 or greater category. And these are some of the most famous F5 tornadoes. >> Do you know what the Stoughton tornado was? >> F4 I believe. So the Stoughton tornado, was that 19, what was that? No, it wasn't 19. It was 2001 or 2000. Right? Something like that. Well maybe 2004. That was an F4 tornado. And here is a compilation of Wisconsin tornadoes from 1844 to 2001. So the further back you go there may be a little bit of ambiguity on whether it was an actual tornado or not. This was put together by the Weather Service at Milwaukee some years ago. Counties with 30 or more are gray. So there's maybe a mini Tornado Alley from Polk County to Marathon. And then another one maybe from Dane County toward Fond du Lac. But we're splitting hairs. You can see, I think, something interesting is that right along the border with Michigan and Lake Superior there are fewer tornadoes in those bigger counties. These are not weighted by area of the county. So there are a couple of relative bald spots in the state, but, by and large, the state, you put a target on any one of those counties and you've got a pretty good chance that you're going to get as many there as in the neighboring counties. So Wisconsin gets a fair share of tornadoes. There's no doubt about that. And the average annual number of tornadoes per state, this is not a fair competition. Texas is huge so they've got 150. Florida is much smaller and gets 62. When you divide it by the area, I think Florida comes out on top, but then Oklahoma and Kansas are next in line. But this is average for the last 30 years through 2010. Wisconsin gets about 23 or so per year. And we do as well as any of the states neighboring us. Even better than those that are across the lake to the east. So, one could say we're probably a far northeastern extent of Tornado Alley, but we're far from the epicenter. But tornadoes are not the only kind of severe weather that these severe thunderstorms can bring. Here's two people who had a fight in the 1870s. One of them, Dr. Gustavus Hinrichs who was at the University of Iowa. The other, Lieutenant John Finley who was with the Army Signal Corps. And that's what the picture on top is. That's the insignia of the Army Signal Corps which President Grant mandated would become weather observing every single day, four times a day, on November 1st, 1870. So, that's when the National Weather Service was actually born. Finley was a really, he was in Iowa, he was very aggressive in trying to understand severe convection, summertime convection. And Hinrichs was actually a chemistry genius, had one of the most famous chemistry labs in the whole world at the University of Iowa in the late 1860s. But he was, by all accounts, a really kind of a pain in the neck type of guy to deal with. Nobody liked him. He didn't like anybody else.

LAUGHTER

dinosaur

He couldn't cooperate with anyone and he went around picking fights and he picked a fight with Finley about tornadoes. Finley had the advantage of having President Grant on his side at the time. Hinrichs had the dean of the College of Letters and Science at Iowa. I think Grant was a little bit more substantial personality. But Hinrichs' main contention was, and Finley disagreed with him, was not all severe wind storms in the plains are tornadoes, and we have to stop talking about every single destructive event as if it were from a tornado. In fact, Hinrichs became aware that there were, and he calls them eastern interests, selling insurance to people on the plains, tornado insurance. And they were quite restricted in what they would call a tornado. And he kind of bought into that. He said, I think they're right. They were restricted because of their commercial interest. He was restricted because of scientific interest. And he said there are sometimes storms that are characterized by exceptionally strong, non-rotating winds. And who hasn't seen, in this part of the country, one of these massive thunderstorm lines that might, from a distance, have the appearance of this picture? Hinrichs writes a paper about one of these things in the Iowa Meteorological Journal or something, that he started actually. Oh, Iowa Weather Report in 1877. And these red lines across Iowa are supposed to isochrones, lines at which this line of storms was, at a certain point in time, going all the way across the state. 1877 and he proposes in this paper to use the word derecho, which is the Spanish word for straight, to describe these widespread, long-lived, straight-line wind storms and to differentiate them from tornadoes. It doesn't matter much maybe from a commercial point of view is your farm is destroyed by straight-line winds or by a tornado, but it matters a lot scientifically because they're different beasts. And so Hinrichs was right on about that, and he puts this word out there and it's in competition with the word tornado which was just beginning, of course, in about 1840, and this is now late 1870s, getting some substantial usage. And he found out that derechos are warm weather phenomenon. They occur mostly in the summer during June and July. Here's a picture of one from Hampshire, Illinois, July 2008. These are awesome. You see that and I think this is almost more awesome than a tornado in some ways because you can see so much of the geometry of the whole thing. It's just awesome to see one of these things roll by. I have seen one of these come by. They are just fantastic. Somebody named John Steuart Curry paints, in 1934, a painting called The Line Storm. Guess where Curry grew up? Kansas. So this is not a tornadic storm. And yet, it inspires this beautiful work of art because it carries a lot of power. Here's these derecho climatology. So, Hinrichs uses the word derecho, and for about 15 years or so there's a battle that goes on between how much damage is done by tornadoes and how much is derechos. Hinrichs, to his credit, did not think every single event that was spawned by severe convection in the plains was a derecho either. He gave plenty of room for tornadoes. But he said there is a different species of severe weather event. At around 1890, the Weather Service, which was at the time called the Weather Bureau, had succumbed to some pressure from people in Washington and in the insurance industry to stop using the word tornado. They were afraid it was going to inspire fear and calamity in the populous. So from 1890 until about 1944, the word tornado was not allowed to be used in any forecast issued by the weather service. You can't find it anywhere. That's a really bad mistake.

LAUGHTER

dinosaur

This is an honest to goodness weather phenomena that has a word that everyone knows what it means, and for reasons that were completely nonscientific, it was banished for almost 50 years. And so with it, the dispute about derecho versus tornado also disappeared. Tornado came back in the '40s. Derecho didn't come back until 1987. A guy named Robert Johns at the Storm Prediction Center in Oklahoma did a fantastic study of these straight-line wind storms, and he did laborious research to dig up the prior references, and he gave Hinrichs a lot of credit for having figured this out. And it's easy after a guy's been dead for 70 or 80 years and his work was forgotten in a dispute that's over a hundred years old to just forget about it entirely and pretend you made it up, but he didn't do that. That's real integrity, and that's what characterizes good science. So here's what Johns climatology has revealed since 1987 when he first reintroduced the term, so it's now about 25 years downstream or so. Tornado Alley, just a little to the east of the Tornado Alley is Derecho Alley. They get a few more derechos sort of near Branson, Missouri, in the Ozarks of Arkansas and Missouri. But notice there's a couple of really interesting features. If you look carefully, you might agree that there's an axis from Minneapolis all the way down to maybe far eastern Kentucky along which we have a pretty obvious maxim of derecho. It goes right through southern Wisconsin. So we are on one of the axes of maximum frequency for these features. Quite interesting and you'll see why in just a minute. They are a late spring/early summer phenomena. Hinrichs was right. Look at the spike in the frequency of events in May, June, and July. Almost nothing up through April. Nothing from August on. They really happen in May, June, July. Warmest time of the year almost. And here's one of the large scale, let's say three miles above the ground flow, picture across the continental US that is characteristic of these derechos. They form on the western and poleward edges of summertime anticyclones. And anticyclones are regions of broad scale clear skies. It tends to be quite warm at the center of them, but right along their periphery, and I think especially on the northern periphery of these anticyclones, for a variety of reasons it's really easy for thunderstorm complexes to grow and to grow into very deep cloud structures. Sometimes as deep as 60,000 feet. And you've got vigorous updrafts and downdrafts going on in those storms. You may not have substantial directional shear of the winds with height, but you've got a lot of speed shear. The speed of the winds is changing a lot with height. And air that's mixing up and down through 60,000 feet is going to take, sometimes the downdraft will take air that's starts out at 55,000 feet and drag it rapidly towards the surface. It doesn't have time for its momentum to slow down. It hits the ground with the same horizontal speed it had at 55,000 feet, and that's usually 70, 80, 90 miles per hours. And so this can happen in these complexes of thunderstorms that develop along the poleward edge of the big, hot anticyclones in the middle of the summer. The so-called ring of fire it's called in the central part of North America. And we saw it coming a week ago. I have to tell you that. In the office we were saying, a couple of my graduate students and I were saying it looks like that ring of fire is going to really get itself established early last weekend and into this week, and sure enough, that happened. Because it's a large scale feature, it's not hard to predict that. Hard to predict the thunderstorm. We'll get to that near the end. So, you get this vigorous mixing in these deep thunderstorms that simply drags high momentum air from high elevation right down to the ground. And it goes roaring across, horizontally across Michigan, Wisconsin, Ohio and creates these derechos. Those are so-called progressive derechos, and there are serial derechos where the line of thunderstorms keeps on regenerating itself in southerly flow. Progressive ones are ones that just maintain a certain structure and progress across a given area. I think this distinction is one of the brand new things that has come online about these derechos in the last 10 years. Here are some famous ones. July 4th, Independence Day of derecho of '77. Forest blow downs are a really big deal here in the summertime, and they are almost always associated with these derechos. If you go to the big Boundary Waters blow down from 1999, all the trees are pointing in the same direction. It's like Mount St. Helens went off or something. And that's because these winds are just coming straight out of the bottom of these thunderstorms, dragged from high elevation, hitting the ground, and racing forward, and they knock everything down in front of them. So, look at how long these things can last. This is over a period of how many? 12, 13, 14 hours. It goes all the way across three states. Unbelievable. 115 mile per hour gusts just north of Wausau that day. Unreal. July 4th again, 1980, the More Trees Down derecho. This was in Johns' paper. Yeah, here it is. They had a whole bunch of some fatalities unfortunately. Lots of injuries. Wind speed well in excess of 80 miles per hour in many locations. All the way from western Iowa to the mouth of the Chesapeake Bay. Unbelievable. July 19th derecho right down I-94. Didn't pay any tolls.

LAUGHTER

dinosaur

But, boy, did it go. All the way from northwest North Dakota down to Chicago, Gary, South Bend. 7th and 8th, 1991. They don't always come from the northwest to the southeast, although that's the preferred direction. But sometimes you'll go straight to the east. It must be, in this case the ridge was probably centered a little bit further south, and this was the northern periphery of that flow. And it goes all the way from Aberdeen, South Dakota, down to Buffalo, training camp for the Bills. 12th and 13th of July 1995. Anybody remember anything about July 13th, 1995? Heat wave. >> Wasn't it a Mississippi...? >> That was in..., no. It wasn't '95, it was '93. That was a bad summer of '93. '95 was 101 degrees here in Madison that day. The dew point was 81. Put them together and you get a variable called the equivalent potential temperature which is kind of a combination of temperature and humidity. It was 383 degrees theta E. All you geeks out there know what I'm talking about.

LAUGHTER

dinosaur

383 theta E was the theta E I've ever experienced. It's higher than you get in the middle of a tropical cyclone at 10 degrees north latitude in the west Pacific. Unbelievable. 830 people died in Chicago from respiratory failure during that heat wave. This was a massive social catastrophe. And look at just along the northern periphery of that heat. It was hot as Hades in Madison, the same in Chicago, just to the north of that is one of these derechos. And it was a devastating event on top of the heat wave to the south. Unbelievable. And then another one May 30th, 1998. One of my graduate students who had family in Minneapolis at the time, now he's a professor at University of Michigan. Great guy. Almost hired him here, it didn't work out. Would have if I had more pull. He's driving back from Minneapolis. This was late Memorial Day weekend. He's coming back and he saw several houses with the vinyl siding and regular siding stripped off the houses after this thing went through. It's unbelievable. Here's what maybe one of these derechos might look like in the radar reflectivity. It's becoming more common. Everybody sees this now days on their phones or however else they get it. This one stretches all the way from sort of Toledo, Ohio, down back towards the south of Indianapolis. It's kind of curved and the red and orange is where it's really high reflectivity, probably water coated hailstones. Intense thunderstorms going on there. This event occurred nearly a year ago to the day. The 29th of June 2012. This is a composite of all of the various times, and you can see the times listed there, of the line of thunderstorms that was expanding north and south as it headed eastward on the 29th of June 2012. Five million people lost power from Chicago to the mid-Atlantic coast as this thing struck nearly every metropolitan area from Chicago to the Tidewater in Virginia. >> What are those numbers? Wind speed? >> Yes. Wind speeds. Peak wind speeds. Those are peak wind speeds. Exactly right. So it gained strength even as it moves to the east. An incredible event. But we saw some other ones like it in the prior history. So, in the last few minutes, I'll just say a couple of words about where we stand with respect to our ability to tell you that these are coming. That event was well forecasted. That was part of the reason why, I think, this whole notion of derechos as increased its profile in the last couple of years. In fact, my next door neighbor asked me just the other day when one of these things went through a couple weeks ago, said what is this derecho? Did you guys just make that up?

LAUGHTER

dinosaur

And I was preparing this presentation but I hadn't gotten to the Hinrichs history, and I said I'm not really sure how far back that goes. I know of a paper from 1987 that seems to be when this word came into the literature, but I don't know yet if that's really the answer. And now I told him the other day, I said, no, it's totally different. It goes all the way back more than 120 years. So we didn't just make it up, but it has become more high profile. I think it's a combination of two things. First, how many people can't pull up a radar display within five minutes if they've got one of these phones or if they've got a computer at home? Just about everybody can do that now days. Let's not underestimate how revolutionary that is because that tells you what kind of weather is bearing down on you in the next couple of hours in the summertime. 20 years ago, 30 years ago certainly, 20 years ago this was impossible. I coached Little League Baseball for the last 10 years, this is the first summer I haven't in many years, and at every concession stand, every clubhouse, every sort of tool shed, on all the fields in the southern part of the state and the northern part of Illinois, and I've been to almost all of them, every one of them has a radar display, and everybody knows how to read it. That's a revolution. We've taken full advantage of that revolution. We can say something about this. So anybody can see this. Not everybody sees this. This is from May 20th this year in southern Oklahoma, central Oklahoma. It's a radar display, but if you saw this and you saw this extraordinary reflectivity, you'd say, wow, it's going to be raining like crazy or hailing like crazy there. What would you say about this hand of God? What would you say about that swirling around? That's the Moore tornado. And that's what they look like. That's what they look like on radar. And it's not a mystery that that's what they look like on radar. So here you go, with this revolution we know what they look like. The combination of forecast ability that I'll mention in a minute and remote surveillance of these storms has increased the warning time for a tornado from just under five minutes on June 7th, 1984, when people in Mount Horeb and Barneveld were under the gun, five minutes to over 15. And if you stop and you count to 600 at a reasonable pace, I think you will agree that's along time. Athletes who've played a game against the clock, you never get nervous when there's 10 minutes left in the third period or 10 minutes left in the quarter. 10 minutes? Come on. It's an eternity. 10 minutes. And think about all the lives that have been saved because of that increase in 10 minutes. That's an investment that our government made in these Doppler radars. That's an investment and that's paid off. So here's Moore May 20th, 2013. Can we forecast these things? It depends on what you mean by forecast. If you mean site-specific accurate depiction of something coming three days in advance, no. And the answer will probably be no for the rest of our lives. But if instead you acquiesce and you say I'll be satisfied if forecast means put me on alert seven days in advance, refine that alert three days in advance, and then remotely observe the thing as it's happening and give me 15 minutes warning, I'll say yeah, you bet we can forecast them. Here's why I say that.

This is 11

00 AM that day from the Norman, Oklahoma, Weather Service office. They're giving this weather briefing. It goes out to media outlets; it goes out to anybody who's got the web,

11

00 AM. Remember, this storm occurred at

about 3

30 in Moore. So this is four and a half hours

earlier Headline

Significant Severe Weather Expected Today! They're not trying to sell you a new and improved product.

LAUGHTER

earlier Headline

There's no profit motive here. They are interested in getting your attention and telling you that something serious is coming. Highest impacts expected along and south of I-44. Perfect. Tornadoes and giant hail likely. Giant? Again, this isn't just a casually thrown in adjective. This is to get your attention. We got something we've got to tell you. Concerns for schools and afternoon rush hour prescient. Here's the depiction. In orange, it does look like orange, most likely location for tornadoes in the afternoon. Three and a half hours early. They're telling people if you're south of Oklahoma City, be on alert. In fact, they can divide it even more than that. If you're south of Oklahoma City

00 and 5

00 PM, be on alert then. If you're further to the east, you can wait a little longer before you're nervous. This is unbelievable. Here's an experimental forecast run with a very expensive high resolution numerical forecast model that incorporates lots of data that we can't actually do every day in our routine forecasts. This is a three-hour forecast,

valid 3

00 PM, initialized at noon time that day. Now, in the next picture I'm not going to have the same scale, but look where this line of severe predicted forecast, predicted convection is. Right across central Oklahoma. Here's what actually was observed at the same time. Okay? Unbelievable. And there's the Moore tornado signature right in the middle. And that's what we saw in a close up. Three hours. That's a lot of time. You've got to be really distracted to not take a three-hour warning. How do we make such forecasts? And I'm just going to quickly wrap up here. I know I'm getting close to the end of the time, but I'm excited. I won't be upset if anybody leaves if they've got something to do. This is a map at three miles above the ground of a depiction of the atmosphere 7 o'clock in the morning on that same day. The blue lines are streamlines of the flow of air at three miles above the grounds. The red lines are lines of temperature. So you can see a couple of things that we've already talked about. Here's an upper trough from Nebraska down to the border of New Mexico and Arizona. Oklahoma City is downstream to the east of that. That puts it in a hotspot. There are local spots along this flow where the temperature contrast is very large like right here over Arizona, New Mexico, into the panhandle of Texas. Much weaker over Baja California. Weaker still over Idaho and western Montana. The temperature contrast is. In fact, I can highlight that region of high temperature contrast in yellow. That's just to the east, just to the west of Oklahoma City. It's not the only one. There are several other ones on the map that you can pick out by eye. And then if you got to a level that's six miles above the ground and you look at wind speeds colored in, you find out that Oklahoma City is just on the south side of a really strong jet streak. There's another strong jet over Quebec. There's another strong one going up along the coast of Greenland. And there's kind of a weak one spinning its way down from southern Alaska all the way down to Arizona. So there are some large scale features that point towards, watch out Oklahoma. There's something to worry about.

This is 7

00 AM. Any experienced meteorologist is going to see this map and say, okay, there's something that we've got to worry about this day. The day before, on Sunday, two-day forecast for moderate risk of severe weather was issued. Look where it was, in the pink. Eastern part of Oklahoma into the Ozarks and Missouri. That's just exactly where that line of thunderstorms developed more than 36 hours later. Can we forecast these large scale features that I'm pointing out long in advance? Well, let me just show you. I had another talk I had the opportunity to give, so that's why it's June 7th. I could have made it June 26th. This is a forecast 168 hours before the event,

valid at 7

00 AM on the 7th of June 2013. It was effectively picked at random. Of the streamlines of the flow at 500 millibars. So you can see a couple of different things three miles above the ground. There's a region of low pressure off the Canadian/British Columbia coast, there's a region of low pressure in the southern part of the Labrador Sea, and there's a broad trough in the central United States. Here is the actual observations from the same time 168 hours after the forecast was made. Now, let's be critical. The first reaction is wow, pretty good. But let's toggle back and forth. Forecast. Observations. Forecast. Observations. So the forecast of two lows, one in the Labrador Sea and one off the British Columbia coast, is right on. The troughiness in the central United States, that's maybe underdone in the forecast. It's a little bit flatter It's a little stronger in the actual data. And then there's this funny little thing down over South Carolina that's completely absent in the forecast. I'll show you again. Right? Not even there. Oh, but it's there in the forecast. That's Tropical Storm Andrea, the first tropical storm of the year, and it happened on that day. But by and large, over the hemisphere sort of view, that's a pretty darn good forecast. Wouldn't you say? Seven days in advance, 168 hours. Let's go to another level. Maybe just one mile above the ground. Same thing. Seven-day forecast, valid 7:00 AM, 7th of June. The red lines are streamlines of the flow in the lowest levels of the atmosphere. And this is the actual observations. It's kind of hard to draw an impression until you get critical. Forecast. Reality. Forecast. Reality. Boy, we nailed it in Canada in adjacent waters, didn't we? Both at 500 and 850. The low in the Labrador Sea is forecasted well. The low off the British Columbia coast is forecasted well. Both of these are pretty good because here's the actual observations. There's our low in the Labrador Sea and the low off the British Columbia coast, but what's that? Let's go back here. Nothing in the forecast for Tropical Storm Andrea. So there are features that are going to escape a seven-day detection, but does that really matter? Do you need to know seven days in advance that you're going to get a tornado or you're going to have a threat for severe weather? I think if you're brought to the alert stage at seven days and then at five days the alert is reinvigorated by new forecasts that seem to be telling the same story, then they're reinvigorated again by a three-day forecast, then a two-day and so on, you know that you're building momentum for a message that you're going to gradually put in louder voice as you get closer to that day zero. And that's exactly the success that we're beginning to have now in making what I call a forecast of this kind of weather. And I think we can continue to improve on that. We're really in quite an interesting position. We're living through a revolution in our science. And you all who consume weather information, forecast information especially, are the beneficiaries of this very revolution. And we're still trying to figure out how to convey that information so that there isn't a gap between what you recall the accuracy level being for numerical prediction and what it actually is today. I think most of us are walking around, maybe I'm not because I'm in the field, but if you're not, you might be walking around with a sense that our forecast ability is perceived to be about what it was in 1985. That's wrong. That is just flat out wrong. You are 30 years behind the times if you bring that attitude towards whether or not you ought to believe a forecast that you get today. I challenge all of you, go home, start doing this. Try to keep score every day in the sense that, use this as your standard, for five days out and longer, is the forecast painting the correct complexion of the day at five or more days out in the forecast? Is it giving you specific information at three days and shorter that's right on the money? And I'll bet you're going to find more than 90% of the time the answer is yes in both categories. And if you start to see that and sense that that's the new reality, then the information that's given out to you, like it was in Oklahoma City that day, is probably going to deliver you from some real harm, both in the wintertime as well as in the summertime. I'll just mention the wintertime. We had a snow day here in Madison December 20th, 2012. Schools were called off all across southern Wisconsin before a single flake of snow had fallen. Let me tell you something, when I was a kid I went to bed at 10 o'clock in New England, and I prayed to God that when I woke up the next day to deliver my newspapers that we actually would get the snowstorm that had

been forecasted at 6

00 PM. We never forecast, you never called school of without seeing the snowfall. Never. Now it's routine. It's happened three times in the last two winters. Unbelievable. Here's keeping track of our tornadoes for this year just as the last slide. We've had about 408 across the country, some of them deadly unfortunately. But they seem to conform this year much like they do in every other year, concentrated in the south central plains, maybe a little bit of a secondary spike somewhere in the southern states. In this case it looks like central Tennessee so far this year. So I want to thank you all for your attention, and I'm more than happy to answer questions. I especially thank you for your patience. I went a little bit over, but I'm happy to answer any more questions you might have.

APPLAUSE