Zzzzap! Pow! Boom!
It’s always been risky business to live in “the lightning capital.” But one business is looking at FSU to improve its odds against losses from Florida’s famous thunderstorms.
By Frank Stephenson
O, Mr. Franklin! Two hundred years dead, but one wonders: What would you have made of Florida? Had fate put you and your famous kite on the shores of Tampa Bay, you might have gotten more than you bargained for! Your clever experiment got you a nice spark on that stormy June day in 1752; in Florida it could have gotten you killed.
In his day, Ben Franklin never could have guessed that the best laboratory for studying lightning in the entire North American continent lay 1,200 miles south of his Boston, Mass. hometown, in a little-known spur of swamps and bugs called Florida.
For decades, along with its famous claim to sunshine Florida has borne the dubious distinction of being the “lightning capital” of the nation. The state’s remarkable geography, flanked by seas on three sides, conspires with its semi-tropical climate to create a natural cauldron for boiling up thunderstorms like no other state.
With more fire from the sky than any place in the country, combined with a swelling population, it’s little wonder that Florida holds the state record for deaths and injuries attributed to lightning. The most recent tally shows that between 1959 and 2004, lightning killed 428 people in Florida and injured another 1,689. The deaths more than double those killed during the same period in Texas, the No. 2 state for deadly strikes.
Florida even has its own “Lightning Alley,” a band of counties stretching across the state’s midsection from the east coast to the west. Along U.S. Interstate 4 from St. Petersburg to Orlando, it’s typical for residents to experience lightning at least 100 days out of every year (by comparison, Californians see five at best).
Just as they have with hurricanes, suffocating heat and humidity, and flying roaches the size of model aircraft, Floridians have learned to take nature’s annual electrical pounding in stride. But that’s not to say they don’t blow a fuse when their power goes outas it does every year for more than 450,000 customers of Florida’s biggest power providerFlorida Power and Lightthanks to lightning hits.
Next to interference from trees and animals, lightning typically accounts for power disruptions to the company’s 4.3 million customers more often than anything else. While it’s widespread outages caused by hurricanes that draw the big headlines, it’s the day-in, day-out toll that lightning takes on power service to homes and businessesnot to mention the company’s line crews already squeezed by Florida’s explosive growththat hurts the most.
And unlike hurricanes, which never come without days, even weeks, of notice, lightning storms in Florida can show up almost anywhere at any time, and with little or no warning. Residents have long since learned to plan their days around certain seasonal patterns to Florida’s annual barrage of thunderstorms (got an after-work tee-off time in June? Count on a rain-delay). But accurate, reliable forecasts of when and where lightning will strikeand how hardremain one of the sturdiest goals in meteorology.
Henry Fuelberg is an FSU meteorologist who truly believes that science is going to one day crack the whimsical code of lightning’s behavior, at least to the point of being able to predict with remarkable precision when and where the phenomenon is going to show up and with what kind of punch. If what he and his small team of grad students have managed to accomplish in the past five years suggests anything at all, it’s that this happy day may not be too far off.
In 2001, Florida Power and Light officials contacted Fuelberg and told him the company was interested in underwriting research into developing a better, more reliable method for forecasting lightning strikes in 11 of their most lightning-prone districts. The company’s network, which covers three-quarters of the state, extends from the Georgia line all the way to Miami and includes four districts on the west coast from Tampa Bay to Sarasota County. If Florida is indeed the nation’s number-one lightning target, the power grids run by FP&L are the bull’s-eye. Every year, up to 600,000 lightning strikes occur within the company’s territories.
Fuelberg said that the company’s chief concern is how to more effectively manage their line crews at crucial times during thunderstorm seasonwhich in Central and South Florida is typically from May through September, with peak times from July through August.
“The big issue with them is what to do with after-hour crews,” said Fuelberg. “Typically, lightning storms like to show up late in the afternoon when crews are ending their shifts. Once they go home, it’s harder to get them back for an emergency. So, having stand-by crews just sitting around on the chance they’ll be needed gets expensive.”
John Haupt, the company’s dispatch manager for the west coast, describes his job as something of a chess game. From roughly May through September, every afternoon around one-o’clock he has to gamble on where, or if, severe storms are going to show up in his 150-mile-long district and deploy his resources accordingly.
“We’re the quarterbacks, making the calls, telling our service centers whether to keep their people or let them go home,” Haupt said. “This can get to be a very risky, very expensive situation. We stand to spend a lot of money on nothing more than a guess.”
Typical service centers in his district carry 20 two-man crews, he said, men who get paid around $35 an hour for overtime paymuch higher over holidays. If Haupt makes the wrong call and sends crews home and a crisis happens, he risks making some very angry customers who can’t get their power restored as quickly as they think they should after a storm.
But Haupt’s neck is a lot safer these days, thanks to genuinely groundbreaking research by Fuelberg’s team.
“Before we had this tool, we were taking information from this source and that source and trying to come up with the best guess on what to do,” Haupt said. “The beauty of (FSU’s technique) is that it pulls all of this together for us, and now we can pretty much say whether (a storm) is going to be a non-event for us or not. The accuracy has been damned good.”
Sea Breeze Science
Now into the fifth phase of their research, Fuelberg’s team continues to focus on the most vexing thunderstorm delivery system in nature’s arsenalcoastal sea breezes.
Florida thunderstorms typically arrive in one of three formsas the vanguards of weather fronts tracked from hundreds of miles away; as noisy theme songs of tropical systems that signal the arrival of hurricane season (June–November); and most notoriously on the coattails of Florida’s storied sea breezes.
These Elysian wafts of cool air, which typically are highly variable and localized, are created when air over land gets heated more quickly than air over waterwhich it always does. This land-heated air rises, creating a pressure shift that pulls in cooler sea air. Collisions between these air masses, with their contrasting temperatures and loads of moisture, can create spectacular cumulonimbus clouds where basic laws of physics breed all the necessary ingredients for lightning.
Nearly surrounded by water, Florida is subject to the whims of sea breezes more than any other state. These winds virtually dictate daily weather throughout the central and southern regions of the state from May through September.
What makes forecasting daily weathermuch less the probability of dangerous lightningso dicey for much of Florida during this period is the state’s peculiar geography, Fuelberg said. Sea breezes are highly influenced by the shapes of coastlines and by inland bodies of water, he said, which accounts for the Tampa area’s reign as Florida’s unrivaled lightning king.
“You don’t have to have a lake the size of Okeechobee, for example, to have a significant effect on sea breezes. And Florida, of course, is full of lakes.”
Still, Fuelberg’s team reasoned that given sufficient information about the weather history of a given area, one could get a reasonable idea of localized land-and-water effects. It follows that if you combine this historic data with daily information about wind speed and direction, humidity and temperature, you might reasonably approach a useful forecasting model for a specific locale, just as forecasts are routinely done every day for large regions across the country.
Phillip Shafer, one of Fuelberg’s four doctoral students, spent over two years teasing out the historic weather data for each of FP&L’s 11 coastal districts and collecting new data gathered from standard weather balloons sent aloft daily by the National Weather Service from stations in Jacksonville, Cape Canaveral, Miami and Tampa. Shafer eventually was able to coalesce all this information into sets of mathematical equations, or algorithms, tailor-made for each district.
“Once you get information from weather balloons such as wind speed and direction, water vapor and temperature at various heights, you can just plug this into one of these algorithms and get a forecast,” Shafer said. “Each of these 11 districts has its own set of equations.”
One of the key elements that goes into each equation is the number of lightning flashes that have been recorded for a specific area during a specific time of year, he said. That data comes from real-time ground observations that are automatically gathered and analyzed by the National Lightning Detection Network, a commercial firm operated by The Vaisala Group, an environmental technology firm based in Finland. The company maintains more than 100 highly sensitive lightning detectors around the country that constantly measure the frequency, intensity, precise location (reportedly accurate within 1,000 feet) and type (there are several) of lightning strikes and the time they occurred. The data stream is then sold to the National Weather Service, power companies, municipalities, researchers and a host of other customers around the country.
Sharpening the Tool
For the first time last year, Florida Power and Light had a polished product from FSU that it could distribute to its dispatch centers around the state. The process works like this: The company uses its own computers to create a lightning forecast based on daily weather balloon data supplied by the Weather Service and run through algorithms supplied by FSU. Forecasts are tailored specifically for each of the company’s 11 coastal districts, and e-mailed usually by mid-morning to dispatchers in each district.
The forecasts, says FP&L’s Raimundo Rey, are highly reliable estimates of the total number of strikes each district is likely to receive, and where most of the activity is likely to be concentrated. Rey serves as principal engineer in the company’s Reliability Engineering Group.
“Last year when we started, this information was very well received by the guys in our dispatch centers, especially those on the west coast,” Rey said. “It was used extensively, particularly to prepare for restoration in the summer.”
Rey is buoyed by what he hears from Haupt and others in the field, and is optimistic that the best is yet to come.
“This project is still in the experimental stage; it’s not complete. But we’re moving forward with it and are going to concentrate on improving both the timing and location of the strikes.”
For all its improvement over conventional methods of forecasting lightning, FSU’s method isn’t nearly as precise as the company’s dispatchers would like, Rey said. Right now, the algorithms are, as Haupt said, “damned accurate” for predicting the intensity of lightning in the various districts, but what operators in the field really want are accurate predictions on what hours of the day will the heaviest lightning occur.
For example, the company doesn’t want to be in the dark about when major activity is most likely to happen, nor do they want to maintain emergency crews in southern Dade County, for example, if they know that most of the lightning is going to strike in the county’s northern half, said Rey.
“The new model that Henry is working on now will take real-time data and allow us to use it by mid-day, just like a normal weather forecast, and we’ll be able to predict when and where and how much lightning will occur,” he said. “That will be a quantum leap forward, and will definitely be one of the main tools for Florida Power and Light this century.”
Indeed, Shafer’s quest to sharpen the resolution of his lightning algorithmsand complete his dissertationis now underway, Fuelberg said. The new model will be designed to produce a forecast every three hours, and instead of having a separate forecast for each of the company’s 11 districts, the forecast will be for the whole state, which may prove helpful to all Floridians, not just to FP&L customers. All of Fuelberg’s lightning forecasts already go not only to the company but to the National Weather Service offices in Miami, Tampa, Melbourne and Jacksonville as well. The agency incorporates FSU’s data into its own forecasting methods.
But Fuelberg cautioned against anyone jumping to irrational conclusions about what’s ultimately possible in the lightning forecasting game. It is, after all, weather.
“We’re going to be able to improve (the model) a great deal, but we’ll never reach the point of being able to predict the exact time and location of a lightning strike.”
That’s probably one thing that wouldn’t have come as a shock to Franklin, playing in the rain with a key on a kite string 244 Junes ago this year.
For more on Henry Fuelberg’s research, visit his Web site.